Pipeline run

{
  "final_skills": [
    {
      "is_primary": true,
      "skill_name": "Python"
    },
    {
      "is_primary": true,
      "skill_name": "SQL"
    },
    {
      "is_primary": true,
      "skill_name": "Snowflake"
    },
    {
      "is_primary": true,
      "skill_name": "Pinecone"
    },
    {
      "is_primary": true,
      "skill_name": "FAISS"
    },
    {
      "is_primary": true,
      "skill_name": "ChromaDB"
    },
    {
      "is_primary": true,
      "skill_name": "LangChain"
    },
    {
      "is_primary": true,
      "skill_name": "LlamaIndex"
    },
    {
      "is_primary": true,
      "skill_name": "OpenAI"
    },
    {
      "is_primary": true,
      "skill_name": "Azure OpenAI"
    },
    {
      "is_primary": true,
      "skill_name": "Anthropic"
    },
    {
      "is_primary": true,
      "skill_name": "Hugging Face"
    },
    {
      "is_primary": true,
      "skill_name": "PostgreSQL"
    },
    {
      "is_primary": true,
      "skill_name": "MySQL"
    },
    {
      "is_primary": true,
      "skill_name": "Git"
    },
    {
      "is_primary": true,
      "skill_name": "CI/CD"
    },
    {
      "is_primary": true,
      "skill_name": "Docker"
    },
    {
      "is_primary": true,
      "skill_name": "MLflow"
    },
    {
      "is_primary": true,
      "skill_name": "AWS"
    },
    {
      "is_primary": true,
      "skill_name": "Azure"
    },
    {
      "is_primary": true,
      "skill_name": "GCP"
    },
    {
      "is_primary": false,
      "skill_name": "Power BI"
    },
    {
      "is_primary": false,
      "skill_name": "Tableau"
    },
    {
      "is_primary": false,
      "skill_name": "CrewAI"
    },
    {
      "is_primary": false,
      "skill_name": "APIs"
    },
    {
      "is_primary": false,
      "skill_name": "LLMs"
    },
    {
      "is_primary": false,
      "skill_name": "RAG"
    },
    {
      "is_primary": false,
      "skill_name": "Embeddings"
    },
    {
      "is_primary": false,
      "skill_name": "MLOps"
    },
    {
      "is_primary": false,
      "skill_name": "Model Versioning"
    },
    {
      "is_primary": false,
      "skill_name": "Chatbots"
    },
    {
      "is_primary": false,
      "skill_name": "Virtual Assistants"
    },
    {
      "is_primary": false,
      "skill_name": "Intelligent Automation"
    }
  ],
  "jd_role": {
    "display_name": "AI Engineer",
    "rationale": null,
    "role_archetype": "Engineering",
    "slug": ""
  },
  "nano_parsed": {
    "JD_type": "pass",
    "about_company": null,
    "certifications": [],
    "company_name": null,
    "ctc": null,
    "domain": {
      "primary": {
        "aliases": [
          "SaaS",
          "Product Companies"
        ],
        "domain": "Software \u0026 SaaS Products"
      },
      "secondary": null
    },
    "education": [],
    "experience": {
      "max": 5,
      "min": 3,
      "raw": "3\u20135 years of hands-on experience in AI/ML, software engineering, or applied GenAI development."
    },
    "job_locations": [],
    "role": "AI Engineer",
    "role_archetype": "Engineering",
    "roles_and_responsibilities": [
      {
        "bullet_count": 0,
        "heading": "Responsibilities",
        "heading_was_present": true,
        "source_marker": {
          "first_5_words": "Design and build AI-powered",
          "last_5_words": "to real-world use cases."
        },
        "text": "Design and build AI-powered applications and conversational agents using LLMs to interact with structured data sources (SQL databases, Snowflake, etc.). \nDevelop and integrate RAG pipelines using vector databases such as Pinecone, FAISS, ChromaDB, or similar tools. \nBuild conversational workflows using AI orchestration frameworks like LangChain, LlamaIndex, or equivalent libraries. \nImplement natural language\u2013to\u2013SQL or data-querying capabilities under guidance from senior engineers. \nIntegrate AI applications with external tools, APIs, and data sources; exposure to agent frameworks or tool-based AI patterns is a plus. \nCollaborate with data engineers, product managers, and business stakeholders to translate requirements into working AI prototypes and features. \nSupport deployment, monitoring, and optimization of AI applications with guidance, following established engineering and MLOps practices. \nAssist with testing, evaluation, and iteration of AI models and workflows to improve accuracy, relevance, and user experience. \nStay current with emerging GenAI tools, models, and best practices, and apply learnings to real-world use cases.",
        "word_count": 202
      },
      {
        "bullet_count": 0,
        "heading": "Required Qualifications",
        "heading_was_present": true,
        "source_marker": {
          "first_5_words": "3\u20135 years of hands-on experience",
          "last_5_words": "in a fast-paced environment."
        },
        "text": "3\u20135 years of hands-on experience in AI/ML, software engineering, or applied GenAI development. \nStrong Python programming skills, including experience building APIs or backend services. \nHands-on experience working with LLMs and GenAI APIs (OpenAI, Azure OpenAI, Anthropic, Hugging Face, etc.). \nPractical experience or exposure to RAG concepts, embeddings, and vector databases. \nExperience working with structured databases (Postgres, MySQL, Snowflake, or similar) and writing SQL queries. \nFamiliarity with modern software development practices (Git, code reviews, basic CI/CD concepts). \nStrong problem-solving skills and an eagerness to learn in a fast-paced environment.",
        "word_count": 104
      },
      {
        "bullet_count": 0,
        "heading": "Nice To Have",
        "heading_was_present": true,
        "source_marker": {
          "first_5_words": "Exposure to AI orchestration or",
          "last_5_words": "downstream consumption of AI outputs."
        },
        "text": "Exposure to AI orchestration or agent frameworks (LangChain, LlamaIndex, CrewAI, etc.). \nBasic understanding of MLOps concepts (Docker, CI/CD, MLflow, model versioning). \nExperience deploying applications on AWS, Azure, or GCP. \nPrior work on chatbots, virtual assistants, or intelligent automation solutions. \nFamiliarity with BI or analytics tools (Power BI, Tableau) for downstream consumption of AI outputs.",
        "word_count": 66
      },
      {
        "bullet_count": 0,
        "heading": "Additional Responsibilities",
        "heading_was_present": true,
        "source_marker": {
          "first_5_words": "Participate in OP monthly team",
          "last_5_words": "to OP Account Management as requested."
        },
        "text": "Participate in OP monthly team meetings, and participate in team-building efforts. \nContribute to OP technical discussions, peer reviews, etc. \nContribute content and collaborate via the OP-Wiki/Knowledge Base. \nProvide status reports to OP Account Management as requested.",
        "word_count": 45
      }
    ],
    "urls": []
  },
  "run_id": "22f3e3d9-31b8-42af-98a2-39fb214dd4a2",
  "stage3_signals": {
    "alias_match_roles": [
      {
        "display_name": "AR/VR Engineer",
        "matched_count": null,
        "role_id": 8,
        "score": 0.7143,
        "slug": "ar-vr-engineer",
        "total_count": null
      },
      {
        "display_name": "Frontend Engineer",
        "matched_count": null,
        "role_id": 7,
        "score": 0.6,
        "slug": "frontend-engineer",
        "total_count": null
      },
      {
        "display_name": "ML Engineer",
        "matched_count": null,
        "role_id": 3,
        "score": 0.6,
        "slug": "ml-engineer",
        "total_count": null
      },
      {
        "display_name": "Ios engineer",
        "matched_count": null,
        "role_id": 6,
        "score": 0.5625,
        "slug": "ios-engineer",
        "total_count": null
      },
      {
        "display_name": "Data Engineer",
        "matched_count": null,
        "role_id": 2,
        "score": 0.5294,
        "slug": "data-engineer",
        "total_count": null
      }
    ],
    "kra_match_roles": [
      {
        "display_name": "AI Compliance Officer",
        "matched_count": null,
        "role_id": 12,
        "score": 0.492,
        "slug": "ai-compliance-officer",
        "total_count": null
      },
      {
        "display_name": "Android Engineer",
        "matched_count": null,
        "role_id": 4,
        "score": 0.4295,
        "slug": "android-engineer",
        "total_count": null
      },
      {
        "display_name": "DevOps Engineer",
        "matched_count": null,
        "role_id": 10,
        "score": 0.4269,
        "slug": "devops-engineer",
        "total_count": null
      },
      {
        "display_name": "AR/VR Engineer",
        "matched_count": null,
        "role_id": 8,
        "score": 0.4188,
        "slug": "ar-vr-engineer",
        "total_count": null
      },
      {
        "display_name": "ML Engineer",
        "matched_count": null,
        "role_id": 3,
        "score": 0.4121,
        "slug": "ml-engineer",
        "total_count": null
      }
    ],
    "skill_match_roles": [
      {
        "display_name": "Data Engineer",
        "matched_count": 8,
        "role_id": 2,
        "score": 0.2424,
        "slug": "data-engineer",
        "total_count": 33
      },
      {
        "display_name": "ML Engineer",
        "matched_count": 8,
        "role_id": 3,
        "score": 0.2424,
        "slug": "ml-engineer",
        "total_count": 33
      },
      {
        "display_name": "Backend Engineer",
        "matched_count": 7,
        "role_id": 1,
        "score": 0.2121,
        "slug": "backend-engineer",
        "total_count": 33
      },
      {
        "display_name": "Cybersecurity Engineer",
        "matched_count": 4,
        "role_id": 5,
        "score": 0.1212,
        "slug": "cybersecurity-engineer",
        "total_count": 33
      },
      {
        "display_name": "DevOps Engineer",
        "matched_count": 4,
        "role_id": 10,
        "score": 0.1212,
        "slug": "devops-engineer",
        "total_count": 33
      }
    ],
    "stage35_ran": false
  },
  "stage4_decision": {
    "alias_collision_detected": false,
    "case": "E",
    "chosen_role": null,
    "confidence": 0.0,
    "llm2_fired": false,
    "llm2_reasoning": null,
    "queued": true,
    "reasoning": "small_margin: KRA margin 0.06 \u003c 0.08"
  },
  "stage5_updates": {
    "centroid_n_after": null,
    "centroid_updated": false,
    "collision_log_id": null,
    "new_kra_attached": null,
    "new_skills_attached": [],
    "queue_entry_id": 27,
    "v3_pipeline_triggered": false,
    "v3_role_slug": null,
    "v3_run_id": null
  }
}

{
  "alias_matches": [
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 67,
      "existing_alias_text": "Python",
      "input_term": "Python",
      "matched_canonical": {
        "category_id": 6,
        "display_name": "Python",
        "id": 5,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "LANGUAGE",
        "slug": "python",
        "sub_category_id": 416,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 271,
      "existing_alias_text": "SQL",
      "input_term": "SQL",
      "matched_canonical": {
        "category_id": 6,
        "display_name": "SQL",
        "id": 101,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "LANGUAGE",
        "slug": "sql",
        "sub_category_id": 97,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 299,
      "existing_alias_text": "Snowflake",
      "input_term": "Snowflake",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "Snowflake",
        "id": 105,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "snowflake",
        "sub_category_id": 113,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 503,
      "existing_alias_text": "Pinecone",
      "input_term": "Pinecone",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "Pinecone",
        "id": 242,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "pinecone",
        "sub_category_id": 177,
        "typical_lifespan": "EVERGREEN",
        "volatility": "EMERGING"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 501,
      "existing_alias_text": "LangChain",
      "input_term": "LangChain",
      "matched_canonical": {
        "category_id": 5,
        "display_name": "LangChain",
        "id": 240,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "FRAMEWORK",
        "slug": "langchain",
        "sub_category_id": 146,
        "typical_lifespan": "EVERGREEN",
        "volatility": "EMERGING"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 505,
      "existing_alias_text": "LlamaIndex",
      "input_term": "LlamaIndex",
      "matched_canonical": {
        "category_id": 5,
        "display_name": "LlamaIndex",
        "id": 244,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "FRAMEWORK",
        "slug": "llamaindex",
        "sub_category_id": 146,
        "typical_lifespan": "EVERGREEN",
        "volatility": "EMERGING"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 121,
      "existing_alias_text": "PostgreSQL",
      "input_term": "PostgreSQL",
      "matched_canonical": {
        "category_id": 3,
        "display_name": "PostgreSQL",
        "id": 16,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "postgresql",
        "sub_category_id": 29,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 134,
      "existing_alias_text": "MySQL",
      "input_term": "MySQL",
      "matched_canonical": {
        "category_id": 3,
        "display_name": "MySQL",
        "id": 17,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "mysql",
        "sub_category_id": 29,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 1613,
      "existing_alias_text": "Git",
      "input_term": "Git",
      "matched_canonical": {
        "category_id": 13,
        "display_name": "Git",
        "id": 1002,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "git",
        "sub_category_id": 730,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 198,
      "existing_alias_text": "Docker",
      "input_term": "Docker",
      "matched_canonical": {
        "category_id": 13,
        "display_name": "Docker",
        "id": 61,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "docker",
        "sub_category_id": 654,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 470,
      "existing_alias_text": "MLflow",
      "input_term": "MLflow",
      "matched_canonical": {
        "category_id": 13,
        "display_name": "MLflow",
        "id": 214,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "mlflow",
        "sub_category_id": 193,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 406,
      "existing_alias_text": "AWS",
      "input_term": "AWS",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "AWS",
        "id": 187,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "aws",
        "sub_category_id": 46,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 407,
      "existing_alias_text": "Azure",
      "input_term": "Azure",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "Azure",
        "id": 188,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "azure",
        "sub_category_id": 46,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 405,
      "existing_alias_text": "GCP",
      "input_term": "GCP",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "GCP",
        "id": 186,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "gcp",
        "sub_category_id": 46,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 360,
      "existing_alias_text": "Power BI",
      "input_term": "Power BI",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "Power BI",
        "id": 151,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "power-bi",
        "sub_category_id": 111,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    },
    {
      "alias_persist_skipped_reason": "alias_text already exists for this canonical skill",
      "alias_persisted": false,
      "existing_alias_id": 359,
      "existing_alias_text": "Tableau",
      "input_term": "Tableau",
      "matched_canonical": {
        "category_id": 9,
        "display_name": "Tableau",
        "id": 150,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "tableau",
        "sub_category_id": 111,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "matched_via": "alias"
    }
  ],
  "candidate_roles": [
    {
      "display_name": "Backend Engineer",
      "id": 1,
      "rationale": null,
      "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
      "slug": "backend-engineer",
      "source": "db"
    },
    {
      "display_name": "Cybersecurity Engineer",
      "id": 5,
      "rationale": null,
      "role_archetype": null,
      "slug": "cybersecurity-engineer",
      "source": "db"
    },
    {
      "display_name": "Data Engineer",
      "id": 2,
      "rationale": null,
      "role_archetype": null,
      "slug": "data-engineer",
      "source": "db"
    },
    {
      "display_name": "ML Engineer",
      "id": 3,
      "rationale": null,
      "role_archetype": null,
      "slug": "ml-engineer",
      "source": "db"
    },
    {
      "display_name": "AR/VR Engineer",
      "id": 8,
      "rationale": null,
      "role_archetype": null,
      "slug": "ar-vr-engineer",
      "source": "db"
    },
    {
      "display_name": "DevOps Engineer",
      "id": 10,
      "rationale": null,
      "role_archetype": null,
      "slug": "devops-engineer",
      "source": "db"
    },
    {
      "display_name": "Cloud Architect",
      "id": 9,
      "rationale": null,
      "role_archetype": null,
      "slug": "cloud-architect",
      "source": "db"
    }
  ],
  "chosen_role": {
    "display_name": "Data Engineer",
    "id": 2,
    "rationale": "The primary skills involve data manipulation and cloud data warehousing, which are essential for a Data Engineer.",
    "role_archetype": null,
    "slug": "data-engineer",
    "source": "db"
  },
  "dimensions": [
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Programming Languages",
        "id": 1,
        "rationale": "Core server-side languages used to implement backend business logic, integrations, and service internals. This is the primary coding surface for the role across application layers.",
        "slug": "programming-languages",
        "source": "db"
      },
      "input_skill": "Python",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Programming Languages and Scripting",
        "id": 59,
        "rationale": "Languages used to write security automation, analysis scripts, detection logic, and remediation helpers. This is the primary implementation surface for a cybersecurity engineer across tooling and response workflows.",
        "slug": "programming-languages-and-scripting",
        "source": "db"
      },
      "input_skill": "Python",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Programming Languages for Data Work",
        "id": 21,
        "rationale": "Languages used to implement data pipelines, transformations, and operational glue. This is the primary coding surface for building ingestion, enrichment, and automation logic in data engineering.",
        "slug": "programming-languages-for-data-work",
        "source": "db"
      },
      "input_skill": "Python",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Programming Languages for ML Systems",
        "id": 39,
        "rationale": "Languages used to build training code, inference services, evaluation jobs, and ML glue code. This is the primary implementation surface for ML engineers across experimentation and productionization.",
        "slug": "programming-languages-for-ml-systems",
        "source": "db"
      },
      "input_skill": "Python",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Programming Languages for XR",
        "id": 97,
        "rationale": "Primary implementation languages used to build immersive client features, interaction logic, and device-specific runtime behavior. This is the core coding surface for AR/VR experiences.",
        "slug": "programming-languages-for-xr",
        "source": "db"
      },
      "input_skill": "Python",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "AR/VR Engineer",
          "id": 8,
          "rationale": null,
          "role_archetype": null,
          "slug": "ar-vr-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Programming Languages for Data Work",
        "id": 21,
        "rationale": "Languages used to implement data pipelines, transformations, and operational glue. This is the primary coding surface for building ingestion, enrichment, and automation logic in data engineering.",
        "slug": "programming-languages-for-data-work",
        "source": "db"
      },
      "input_skill": "SQL",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Data Warehouses",
        "id": 22,
        "rationale": "Managed analytical storage and compute platforms used for curated datasets, reporting, and downstream analytics. These systems are central to data modeling, performance tuning, and cost-aware query design.",
        "slug": "cloud-data-warehouses",
        "source": "db"
      },
      "input_skill": "Snowflake",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "LLM Operations and Orchestration",
        "id": 49,
        "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
        "slug": "llm-operations-and-orchestration",
        "source": "db"
      },
      "input_skill": "Pinecone",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "LLM Operations and Orchestration",
        "id": 49,
        "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
        "slug": "llm-operations-and-orchestration",
        "source": "db"
      },
      "input_skill": "LangChain",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "LLM Operations and Orchestration",
        "id": 49,
        "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
        "slug": "llm-operations-and-orchestration",
        "source": "db"
      },
      "input_skill": "LlamaIndex",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Relational Database Design",
        "id": 4,
        "rationale": "Modeling and operating relational persistence for backend services. Includes schema design, normalization, indexing, transactions, and query tuning for operational data stores.",
        "slug": "relational-database-design",
        "source": "db"
      },
      "input_skill": "PostgreSQL",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Relational Database Design",
        "id": 4,
        "rationale": "Modeling and operating relational persistence for backend services. Includes schema design, normalization, indexing, transactions, and query tuning for operational data stores.",
        "slug": "relational-database-design",
        "source": "db"
      },
      "input_skill": "MySQL",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Git",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Containerization and Image Builds",
        "id": 152,
        "rationale": "Container image creation, tagging, hardening, and registry workflows used to package services for deployment. This is coherent because DevOps often owns the build-to-image path that feeds runtime environments.",
        "slug": "containerization-and-image-builds",
        "source": "db"
      },
      "input_skill": "Docker",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Deployment and Runtime Configuration",
        "id": 13,
        "rationale": "Configuration and release artifacts that control how backend services run in environments. Includes environment variables, manifests, feature flags, and release-safe configuration management.",
        "slug": "deployment-and-runtime-configuration",
        "source": "db"
      },
      "input_skill": "Docker",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "MLOps Platforms and Lifecycle",
        "id": 43,
        "rationale": "End-to-end managed platforms used to train, deploy, register, and govern models across their lifecycle. This is the operational control plane for production ML workflows.",
        "slug": "mlops-platforms-and-lifecycle",
        "source": "db"
      },
      "input_skill": "MLflow",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Platforms",
        "id": 20,
        "rationale": "Proficiency in major cloud service provider platforms and their core services.",
        "slug": "cloud-platforms",
        "source": "db"
      },
      "input_skill": "AWS",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        },
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        },
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        },
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        },
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Provider Platforms",
        "id": 131,
        "rationale": "Major cloud platforms and their core service ecosystems used to design target-state architectures, choose deployment boundaries, and evaluate managed capabilities. This is the primary substrate for cloud architecture decisions.",
        "slug": "cloud-provider-platforms",
        "source": "db"
      },
      "input_skill": "AWS",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Cloud Architect",
          "id": 9,
          "rationale": null,
          "role_archetype": null,
          "slug": "cloud-architect",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Security Posture Tools",
        "id": 64,
        "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
        "slug": "cloud-security-posture-tools",
        "source": "db"
      },
      "input_skill": "AWS",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Platforms",
        "id": 20,
        "rationale": "Proficiency in major cloud service provider platforms and their core services.",
        "slug": "cloud-platforms",
        "source": "db"
      },
      "input_skill": "Azure",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        },
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        },
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        },
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        },
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Provider Platforms",
        "id": 131,
        "rationale": "Major cloud platforms and their core service ecosystems used to design target-state architectures, choose deployment boundaries, and evaluate managed capabilities. This is the primary substrate for cloud architecture decisions.",
        "slug": "cloud-provider-platforms",
        "source": "db"
      },
      "input_skill": "Azure",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Cloud Architect",
          "id": 9,
          "rationale": null,
          "role_archetype": null,
          "slug": "cloud-architect",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Security Posture Tools",
        "id": 64,
        "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
        "slug": "cloud-security-posture-tools",
        "source": "db"
      },
      "input_skill": "Azure",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Platforms",
        "id": 20,
        "rationale": "Proficiency in major cloud service provider platforms and their core services.",
        "slug": "cloud-platforms",
        "source": "db"
      },
      "input_skill": "GCP",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        },
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        },
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        },
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        },
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Security Posture Tools",
        "id": 64,
        "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
        "slug": "cloud-security-posture-tools",
        "source": "db"
      },
      "input_skill": "GCP",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "BI and Visualization Tools",
        "id": 31,
        "rationale": "Tools used to expose curated data to analysts and business users through dashboards, reports, and semantic exploration. Data engineers support these tools by shaping reliable datasets and performant models.",
        "slug": "bi-and-visualization-tools",
        "source": "db"
      },
      "input_skill": "Power BI",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "BI and Visualization Tools",
        "id": 31,
        "rationale": "Tools used to expose curated data to analysts and business users through dashboards, reports, and semantic exploration. Data engineers support these tools by shaping reliable datasets and performant models.",
        "slug": "bi-and-visualization-tools",
        "source": "db"
      },
      "input_skill": "Tableau",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "FAISS",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "ML Frameworks and Libraries",
        "id": 40,
        "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
        "slug": "ml-frameworks-and-libraries",
        "source": "db"
      },
      "input_skill": "FAISS",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "ML Frameworks and Libraries",
        "id": 40,
        "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
        "slug": "ml-frameworks-and-libraries",
        "source": "db"
      },
      "input_skill": "FAISS",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "ChromaDB",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "OpenAI",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Cloud Platforms",
        "id": 20,
        "rationale": "Proficiency in major cloud service provider platforms and their core services.",
        "slug": "cloud-platforms",
        "source": "db"
      },
      "input_skill": "Azure OpenAI",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Backend Engineer",
          "id": 1,
          "rationale": null,
          "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
          "slug": "backend-engineer",
          "source": "db"
        },
        {
          "display_name": "Cybersecurity Engineer",
          "id": 5,
          "rationale": null,
          "role_archetype": null,
          "slug": "cybersecurity-engineer",
          "source": "db"
        },
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        },
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        },
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Azure OpenAI",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Anthropic",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "ML Frameworks and Libraries",
        "id": 40,
        "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
        "slug": "ml-frameworks-and-libraries",
        "source": "db"
      },
      "input_skill": "Hugging Face",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Hugging Face",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "ML Frameworks and Libraries",
        "id": 40,
        "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
        "slug": "ml-frameworks-and-libraries",
        "source": "db"
      },
      "input_skill": "Hugging Face",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD Pipeline Platforms",
        "id": 150,
        "rationale": "Systems used to define, run, and maintain automated build and deployment workflows. This cluster is coherent because the role owns delivery automation end to end, including pipeline reliability and promotion logic.",
        "slug": "ci-cd-pipeline-platforms",
        "source": "db"
      },
      "input_skill": "CI/CD",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD for Machine Learning",
        "id": 56,
        "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
        "slug": "ci-cd-for-machine-learning",
        "source": "db"
      },
      "input_skill": "CI/CD",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD Pipeline Platforms",
        "id": 150,
        "rationale": "Systems used to define, run, and maintain automated build and deployment workflows. This cluster is coherent because the role owns delivery automation end to end, including pipeline reliability and promotion logic.",
        "slug": "ci-cd-pipeline-platforms",
        "source": "db"
      },
      "input_skill": "CI/CD",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "DevOps Engineer",
          "id": 10,
          "rationale": null,
          "role_archetype": null,
          "slug": "devops-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD for Machine Learning",
        "id": 56,
        "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
        "slug": "ci-cd-for-machine-learning",
        "source": "db"
      },
      "input_skill": "CI/CD",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "CrewAI",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "APIs",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "LLMs",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "RAG",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "ML Frameworks and Libraries",
        "id": 40,
        "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
        "slug": "ml-frameworks-and-libraries",
        "source": "db"
      },
      "input_skill": "Embeddings",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Embeddings",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Systems Programming",
        "id": 166,
        "rationale": "Systems programming covers low-level software development where performance, memory safety, and direct control over resources matter. Rust fits here because it is commonly used for OS-adjacent services, infrastructure components, and other performance-sensitive systems code.",
        "slug": "d_init_02",
        "source": "db"
      },
      "input_skill": "Embeddings",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "ML Frameworks and Libraries",
        "id": 40,
        "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
        "slug": "ml-frameworks-and-libraries",
        "source": "db"
      },
      "input_skill": "Embeddings",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD for Machine Learning",
        "id": 56,
        "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
        "slug": "ci-cd-for-machine-learning",
        "source": "db"
      },
      "input_skill": "MLOps",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Deployment Rollouts and Release Control",
        "id": 51,
        "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
        "slug": "deployment-rollouts-and-release-control",
        "source": "db"
      },
      "input_skill": "MLOps",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Data Lineage and Metadata",
        "id": 28,
        "rationale": "Cataloging, documenting, and tracing how data moves and changes across systems. This dimension supports impact analysis, governance, discoverability, and operational understanding of datasets.",
        "slug": "data-lineage-and-metadata",
        "source": "db"
      },
      "input_skill": "MLOps",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD for Machine Learning",
        "id": 56,
        "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
        "slug": "ci-cd-for-machine-learning",
        "source": "db"
      },
      "input_skill": "MLOps",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Deployment Rollouts and Release Control",
        "id": 51,
        "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
        "slug": "deployment-rollouts-and-release-control",
        "source": "db"
      },
      "input_skill": "MLOps",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Data Lineage and Metadata",
        "id": 28,
        "rationale": "Cataloging, documenting, and tracing how data moves and changes across systems. This dimension supports impact analysis, governance, discoverability, and operational understanding of datasets.",
        "slug": "data-lineage-and-metadata",
        "source": "db"
      },
      "input_skill": "MLOps",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "Data Engineer",
          "id": 2,
          "rationale": null,
          "role_archetype": null,
          "slug": "data-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Deployment Rollouts and Release Control",
        "id": 51,
        "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
        "slug": "deployment-rollouts-and-release-control",
        "source": "db"
      },
      "input_skill": "Model Versioning",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD for Machine Learning",
        "id": 56,
        "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
        "slug": "ci-cd-for-machine-learning",
        "source": "db"
      },
      "input_skill": "Model Versioning",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "CI/CD for Machine Learning",
        "id": 56,
        "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
        "slug": "ci-cd-for-machine-learning",
        "source": "db"
      },
      "input_skill": "Model Versioning",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "Deployment Rollouts and Release Control",
        "id": 51,
        "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
        "slug": "deployment-rollouts-and-release-control",
        "source": "db"
      },
      "input_skill": "Model Versioning",
      "llm_role": null,
      "roles_from_db": [
        {
          "display_name": "ML Engineer",
          "id": 3,
          "rationale": null,
          "role_archetype": null,
          "slug": "ml-engineer",
          "source": "db"
        }
      ]
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Chatbots",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Virtual Assistants",
      "llm_role": null,
      "roles_from_db": []
    },
    {
      "dimension": {
        "difficulty_hint": "well_known",
        "display_name": "React Frontend Development",
        "id": 96,
        "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
        "slug": "d_init_01",
        "source": "db"
      },
      "input_skill": "Intelligent Automation",
      "llm_role": null,
      "roles_from_db": []
    }
  ],
  "input_final_skills": [
    "Python",
    "SQL",
    "Snowflake",
    "Pinecone",
    "FAISS",
    "ChromaDB",
    "LangChain",
    "LlamaIndex",
    "OpenAI",
    "Azure OpenAI",
    "Anthropic",
    "Hugging Face",
    "PostgreSQL",
    "MySQL",
    "Git",
    "CI/CD",
    "Docker",
    "MLflow",
    "AWS",
    "Azure",
    "GCP",
    "Power BI",
    "Tableau",
    "CrewAI",
    "APIs",
    "LLMs",
    "RAG",
    "Embeddings",
    "MLOps",
    "Model Versioning",
    "Chatbots",
    "Virtual Assistants",
    "Intelligent Automation"
  ],
  "input_llm_skills": [
    "Python",
    "SQL",
    "Snowflake",
    "Pinecone",
    "FAISS",
    "ChromaDB",
    "LangChain",
    "LlamaIndex",
    "OpenAI",
    "Azure OpenAI",
    "Anthropic",
    "Hugging Face",
    "PostgreSQL",
    "MySQL",
    "Git",
    "CI/CD",
    "Docker",
    "MLflow",
    "AWS",
    "Azure",
    "GCP",
    "Power BI",
    "Tableau",
    "CrewAI",
    "APIs",
    "LLMs",
    "RAG",
    "Embeddings",
    "MLOps",
    "Model Versioning",
    "Chatbots",
    "Virtual Assistants",
    "Intelligent Automation"
  ],
  "new_aliases_persisted": 0,
  "run_id": "22f3e3d9-31b8-42af-98a2-39fb214dd4a2",
  "skills_detail": [
    {
      "aliases_in_db": [
        {
          "alias_text": "Python",
          "alias_type": "CANONICAL",
          "id": 67,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 2",
          "alias_type": "VERSION",
          "id": 72,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 2.x",
          "alias_type": "VERSION",
          "id": 74,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 3",
          "alias_type": "VERSION",
          "id": 73,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 3.10",
          "alias_type": "VERSION",
          "id": 76,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 3.11",
          "alias_type": "VERSION",
          "id": 77,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 3.12",
          "alias_type": "VERSION",
          "id": 78,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Python 3.x",
          "alias_type": "VERSION",
          "id": 75,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "py2",
          "alias_type": "VERSION",
          "id": 68,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "py3",
          "alias_type": "VERSION",
          "id": 69,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "python2",
          "alias_type": "VERSION",
          "id": 70,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "python3",
          "alias_type": "VERSION",
          "id": 71,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 6,
        "display_name": "Python",
        "id": 5,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "LANGUAGE",
        "slug": "python",
        "sub_category_id": 416,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Programming Languages",
            "id": 1,
            "rationale": "Core server-side languages used to implement backend business logic, integrations, and service internals. This is the primary coding surface for the role across application layers.",
            "slug": "programming-languages",
            "source": "db"
          },
          "input_skill": "Python",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Programming Languages and Scripting",
            "id": 59,
            "rationale": "Languages used to write security automation, analysis scripts, detection logic, and remediation helpers. This is the primary implementation surface for a cybersecurity engineer across tooling and response workflows.",
            "slug": "programming-languages-and-scripting",
            "source": "db"
          },
          "input_skill": "Python",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Programming Languages for Data Work",
            "id": 21,
            "rationale": "Languages used to implement data pipelines, transformations, and operational glue. This is the primary coding surface for building ingestion, enrichment, and automation logic in data engineering.",
            "slug": "programming-languages-for-data-work",
            "source": "db"
          },
          "input_skill": "Python",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Programming Languages for ML Systems",
            "id": 39,
            "rationale": "Languages used to build training code, inference services, evaluation jobs, and ML glue code. This is the primary implementation surface for ML engineers across experimentation and productionization.",
            "slug": "programming-languages-for-ml-systems",
            "source": "db"
          },
          "input_skill": "Python",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Programming Languages for XR",
            "id": 97,
            "rationale": "Primary implementation languages used to build immersive client features, interaction logic, and device-specific runtime behavior. This is the core coding surface for AR/VR experiences.",
            "slug": "programming-languages-for-xr",
            "source": "db"
          },
          "input_skill": "Python",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "AR/VR Engineer",
              "id": 8,
              "rationale": null,
              "role_archetype": null,
              "slug": "ar-vr-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Python",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "SQL",
          "alias_type": "CANONICAL",
          "id": 271,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 6,
        "display_name": "SQL",
        "id": 101,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "LANGUAGE",
        "slug": "sql",
        "sub_category_id": 97,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Programming Languages for Data Work",
            "id": 21,
            "rationale": "Languages used to implement data pipelines, transformations, and operational glue. This is the primary coding surface for building ingestion, enrichment, and automation logic in data engineering.",
            "slug": "programming-languages-for-data-work",
            "source": "db"
          },
          "input_skill": "SQL",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "SQL",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Snowflake",
          "alias_type": "CANONICAL",
          "id": 299,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "Snowflake",
        "id": 105,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "snowflake",
        "sub_category_id": 113,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Data Warehouses",
            "id": 22,
            "rationale": "Managed analytical storage and compute platforms used for curated datasets, reporting, and downstream analytics. These systems are central to data modeling, performance tuning, and cost-aware query design.",
            "slug": "cloud-data-warehouses",
            "source": "db"
          },
          "input_skill": "Snowflake",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Snowflake",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Pinecone",
          "alias_type": "CANONICAL",
          "id": 503,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "Pinecone",
        "id": 242,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "pinecone",
        "sub_category_id": 177,
        "typical_lifespan": "EVERGREEN",
        "volatility": "EMERGING"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "LLM Operations and Orchestration",
            "id": 49,
            "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
            "slug": "llm-operations-and-orchestration",
            "source": "db"
          },
          "input_skill": "Pinecone",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Pinecone",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "FAISS",
          "llm_role": null,
          "roles_from_db": []
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "ML Frameworks and Libraries",
            "id": 40,
            "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "slug": "ml-frameworks-and-libraries",
            "source": "db"
          },
          "input_skill": "FAISS",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "ML Frameworks and Libraries",
            "id": 40,
            "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "slug": "ml-frameworks-and-libraries",
            "source": "db"
          },
          "input_skill": "FAISS",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "FAISS",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Library",
          "skill_nature": "LIBRARY",
          "sub_category": "vector_search_library",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "FAISS is a specific, well-known vector search library; typical JDs won\u2019t confuse it with other unrelated skills."
          },
          "context_keywords": {
            "context_keywords": [
              "vector search",
              "nearest neighbor",
              "indexing",
              "clustering",
              "similarity search",
              "high-dimensional",
              "approximate nearest neighbors",
              "GPU acceleration",
              "quantization",
              "search algorithms",
              "embedding",
              "scalability",
              "performance tuning",
              "data retrieval",
              "machine learning"
            ]
          },
          "maturity": {
            "confidence": 0.84,
            "maturity": "emerging",
            "reasoning": "FAISS appears in growing number of AI/vector-search job descriptions and is widely used in RAG stacks, but it is still less universal than PostgreSQL or Kubernetes and often paired with managed vector DBs."
          },
          "skill_id": "faiss",
          "vendor_license": {
            "confidence": 0.95,
            "license": "apache_2",
            "vendor": "Facebook",
            "year_introduced": 2017
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [
          {
            "a_dim_id": "d_init_01",
            "a_name": "Vector Search Indexing",
            "a_role": "__skill_focal__",
            "b_dim_id": "ml-frameworks-and-libraries",
            "b_name": "ML Frameworks and Libraries",
            "b_role": "__skill_focal__",
            "pair_kind": "intra_role",
            "reasoning": "Dim A is about vector search indexing/ANN structures (FAISS, HNSW, IVF, PQ) for similarity retrieval over embeddings. Dim B is about general ML frameworks/libraries for defining models, training loops, inference, and evaluation. A senior FAISS/ANN engineer is not naturally a senior PyTorch/TensorFlow-style framework engineer; different daily work and ecosystem. The similarity is a false positive from broad ML-library wording.",
            "similarity": 0.621129319975389
          }
        ],
        "locked_dimensions": [
          {
            "description": "Indexes and retrieval structures for fast nearest-neighbor search over embeddings and other high-dimensional vectors. FAISS belongs here because it is a core library for building and querying similarity search indexes at scale.",
            "exemplar_skills": [
              "FAISS",
              "approximate nearest neighbor search",
              "vector indexing",
              "similarity search",
              "HNSW",
              "IVF",
              "product quantization"
            ],
            "in_scope": "FAISS, approximate nearest neighbor search, vector indexing, similarity search, embedding retrieval, IVF indexes, HNSW indexes, PQ compression, index training, index tuning",
            "name": "Vector Search Indexing",
            "out_of_scope": "Dense model training, embedding generation, reranking models, database schema design, full-text search engines, general SQL query optimization",
            "overlap_flags": [
              {
                "reason": "FAISS is often used alongside ML frameworks to serve embeddings, but it is specifically a retrieval/indexing library rather than a model framework.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          },
          {
            "description": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. FAISS can fit here when the emphasis is on ML tooling used in embedding-based systems, though its main identity is retrieval indexing.",
            "exemplar_skills": [
              "FAISS",
              "PyTorch",
              "TensorFlow",
              "scikit-learn",
              "JAX",
              "XGBoost"
            ],
            "in_scope": "FAISS, PyTorch, TensorFlow, scikit-learn, JAX, XGBoost, model inference, embedding pipelines, similarity search tooling",
            "name": "ML Frameworks and Libraries",
            "out_of_scope": "Database engines, application UI frameworks, cloud infrastructure, orchestration platforms, general software testing tools",
            "overlap_flags": [
              {
                "reason": "FAISS is more specifically a vector search index library, so retrieval-focused work should be classified there when possible.",
                "with_dim_id": "d_init_01",
                "with_dim_name": null,
                "with_role": null
              }
            ],
            "tentative_id": "ml-frameworks-and-libraries"
          },
          {
            "description": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "exemplar_skills": [
              "ML Frameworks and Libraries"
            ],
            "in_scope": "Skills, tools, and practices that belong under ML Frameworks and Libraries for the target role, including items implied by the dimension rationale.",
            "name": "ML Frameworks and Libraries",
            "out_of_scope": "Adjacent clusters explicitly not owned by ML Frameworks and Libraries, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ml-frameworks-and-libraries"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "FAISS",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 3 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "ml-frameworks-and-libraries"
          ],
          "skill_id": "faiss"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [],
          "requires": [],
          "skill_id": "faiss",
          "suppress_on_match": []
        },
        "skill_id": "faiss",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "FAISS",
          "reasoning": "FAISS is fundamentally a code package imported by applications for similarity search, so by the Tool vs Framework rule it is a Library rather than a user-operated tool or hosted platform.",
          "skill_id": "faiss",
          "subtype": "vector_search_library",
          "type": "Library"
        },
        "warnings": [
          "stage3_post_filter_dropped_catalog_only_locked_dims:42-\u003e3"
        ]
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "ChromaDB",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "ChromaDB",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Datastore",
          "skill_nature": "TOOL",
          "sub_category": "vector_database",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "ChromaDB is a specific vector database product name; typical JDs won\u2019t confuse it with other distinct vector DBs."
          },
          "context_keywords": {
            "context_keywords": [
              "vector search",
              "embedding",
              "similarity search",
              "data retrieval",
              "scalability",
              "real-time analytics",
              "machine learning",
              "data indexing",
              "query optimization",
              "metadata management",
              "high-dimensional data",
              "distributed architecture",
              "data pipelines",
              "performance tuning",
              "API integration"
            ]
          },
          "maturity": {
            "confidence": 0.84,
            "maturity": "emerging",
            "reasoning": "ChromaDB appears in a growing number of AI/vector-search job postings and GitHub usage, but it is still far less common than Pinecone, Weaviate, or pgvector in hiring pipelines."
          },
          "skill_id": "chromadb",
          "vendor_license": {
            "confidence": 0.9,
            "license": "apache_2",
            "vendor": "Chroma",
            "year_introduced": 2021
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Systems for storing embeddings and performing similarity search over unstructured data. ChromaDB belongs here because it is a vector database used to index embeddings and retrieve nearest neighbors for RAG and semantic search workflows.",
            "exemplar_skills": [
              "ChromaDB",
              "vector databases",
              "embedding search",
              "similarity search",
              "retrieval-augmented generation",
              "semantic search",
              "ANN indexing"
            ],
            "in_scope": "ChromaDB, vector storage, embedding indexes, similarity search, nearest-neighbor retrieval, metadata filtering, semantic search, retrieval-augmented generation, ANN search, document chunk indexing",
            "name": "Vector Databases and Retrieval",
            "out_of_scope": "Traditional relational databases, key-value caches, full-text search engines without embeddings, model training frameworks, orchestration of LLM prompts and agents",
            "overlap_flags": [
              {
                "reason": "Vector databases are often used alongside ML libraries to generate and consume embeddings, but the storage/retrieval layer is distinct.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "ChromaDB",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "chromadb"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "sqlite",
            "watermelondb",
            "managed-databases",
            "grafana",
            "datadog",
            "kubernetes",
            "docker",
            "google-cloud-platform"
          ],
          "requires": [],
          "skill_id": "chromadb",
          "suppress_on_match": []
        },
        "skill_id": "chromadb",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.94,
          "name": "ChromaDB",
          "reasoning": "By the Datastore vs Format rule, ChromaDB is a system that persists and serves embeddings/data, so it is fundamentally a datastore rather than a tool or library.",
          "skill_id": "chromadb",
          "subtype": "vector_database",
          "type": "Datastore"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "LangChain",
          "alias_type": "CANONICAL",
          "id": 501,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 5,
        "display_name": "LangChain",
        "id": 240,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "FRAMEWORK",
        "slug": "langchain",
        "sub_category_id": 146,
        "typical_lifespan": "EVERGREEN",
        "volatility": "EMERGING"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "LLM Operations and Orchestration",
            "id": 49,
            "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
            "slug": "llm-operations-and-orchestration",
            "source": "db"
          },
          "input_skill": "LangChain",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "LangChain",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "LlamaIndex",
          "alias_type": "CANONICAL",
          "id": 505,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 5,
        "display_name": "LlamaIndex",
        "id": 244,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "FRAMEWORK",
        "slug": "llamaindex",
        "sub_category_id": 146,
        "typical_lifespan": "EVERGREEN",
        "volatility": "EMERGING"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "LLM Operations and Orchestration",
            "id": 49,
            "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
            "slug": "llm-operations-and-orchestration",
            "source": "db"
          },
          "input_skill": "LlamaIndex",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "LlamaIndex",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "OpenAI",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "OpenAI",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Platform",
          "skill_nature": "PLATFORM",
          "sub_category": "ai_platform",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cOpenAI\u201d in JDs typically refers specifically to the OpenAI company/models, not another distinct catalog skill."
          },
          "context_keywords": {
            "context_keywords": [
              "GPT-3",
              "ChatGPT",
              "DALL-E",
              "Codex",
              "API",
              "fine-tuning",
              "transformer",
              "machine learning",
              "natural language processing",
              "AI ethics",
              "prompt engineering",
              "model training",
              "reinforcement learning",
              "data augmentation",
              "neural networks"
            ]
          },
          "maturity": {
            "confidence": 0.84,
            "maturity": "emerging",
            "reasoning": "OpenAI appears in a growing number of job postings for LLM/app integration, but it is not yet a universal baseline skill like AWS or Python; market demand is rising alongside API adoption."
          },
          "skill_id": "openai",
          "vendor_license": {
            "confidence": 0.95,
            "license": "other_open",
            "vendor": "OpenAI",
            "year_introduced": 2015
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Working with OpenAI as a platform for building LLM-powered features, including API usage, prompt design, tool calling, and response handling. This fits AI Engineer work because OpenAI is commonly used to integrate generative models into products and workflows.",
            "exemplar_skills": [
              "OpenAI",
              "OpenAI API",
              "prompt engineering",
              "function calling",
              "structured outputs",
              "embeddings",
              "streaming responses"
            ],
            "in_scope": "OpenAI API, Chat Completions, Responses API, prompt engineering, system and user messages, function calling, tool use, embeddings, structured outputs, rate limits, model selection, streaming responses",
            "name": "LLM APIs and Prompting",
            "out_of_scope": "Training foundation models from scratch, general MLOps pipelines, non-OpenAI model hosting, browser UI integration, mobile SDK UI concerns",
            "overlap_flags": [
              {
                "reason": "OpenAI is often used alongside ML libraries, but this dimension is specifically about using the OpenAI platform rather than model implementation.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "OpenAI-powered features may be deployed through ML pipelines, but the core skill here is API integration and prompting, not pipeline automation.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "OpenAI",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "openai"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "aws-iam",
            "aws-cdk",
            "azure-virtual-network",
            "kubernetes",
            "google-cloud-platform"
          ],
          "requires": [],
          "skill_id": "openai",
          "suppress_on_match": []
        },
        "skill_id": "openai",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "OpenAI",
          "reasoning": "By the Vendor SaaS = Platform rule, OpenAI is a hosted multi-tenant environment with APIs and managed capabilities rather than software you run yourself.",
          "skill_id": "openai",
          "subtype": "ai_platform",
          "type": "Platform"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Platforms",
            "id": 20,
            "rationale": "Proficiency in major cloud service provider platforms and their core services.",
            "slug": "cloud-platforms",
            "source": "db"
          },
          "input_skill": "Azure OpenAI",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            },
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            },
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            },
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            },
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Azure OpenAI",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "Azure OpenAI",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Service",
          "skill_nature": "CLOUD_SERVICE",
          "sub_category": "ai_model_service",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cAzure OpenAI\u201d is a specific Azure service name; unlikely to be confused with other catalog skills."
          },
          "context_keywords": {
            "context_keywords": [
              "GPT-3",
              "ChatGPT",
              "Azure Cognitive Services",
              "machine learning",
              "natural language processing",
              "AI integration",
              "Azure Functions",
              "data preprocessing",
              "model deployment",
              "API management",
              "Azure Machine Learning",
              "Azure DevOps",
              "scalability",
              "security compliance",
              "cost optimization"
            ]
          },
          "maturity": {
            "confidence": 0.86,
            "maturity": "emerging",
            "reasoning": "Appears increasingly in job postings for GenAI/Azure roles, but is still far less universal than core cloud services like AWS or Azure itself; market demand is growing rather than ubiquitous."
          },
          "skill_id": "azure-openai",
          "vendor_license": {
            "confidence": 0.95,
            "license": "proprietary",
            "vendor": "Microsoft",
            "year_introduced": 2021
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Managed cloud services used to deploy and consume AI capabilities on a provider platform. Azure OpenAI fits here because it is an Azure-hosted service for accessing foundation models, embeddings, and related AI APIs.",
            "exemplar_skills": [
              "Azure OpenAI",
              "Azure AI Services",
              "Azure Cognitive Services",
              "Azure AI Studio",
              "OpenAI API on Azure"
            ],
            "in_scope": "Azure OpenAI, Azure AI services, managed model endpoints, embeddings APIs, chat/completions APIs, Azure resource setup for AI services, API keys and managed identity access",
            "name": "Cloud AI Platform Services",
            "out_of_scope": "Custom model training frameworks, on-prem inference servers, generic cloud networking, non-AI Azure services such as storage or compute provisioning",
            "overlap_flags": [
              {
                "reason": "Azure OpenAI is also a third-party AI platform that may require vendor review before adoption.",
                "with_dim_id": "ai-vendor-and-third-party-due-diligence",
                "with_dim_name": null,
                "with_role": "AI Compliance Officer"
              },
              {
                "reason": "Using Azure OpenAI often triggers policy review for approved AI use cases and data handling.",
                "with_dim_id": "ai-use-case-compliance-review",
                "with_dim_name": null,
                "with_role": "AI Compliance Officer"
              }
            ],
            "tentative_id": "cloud-platforms"
          },
          {
            "description": "Application-side integration of hosted large language model APIs for chat, embeddings, and generation workflows. Azure OpenAI belongs here because it is commonly used as the API layer for building AI-enabled products.",
            "exemplar_skills": [
              "Azure OpenAI",
              "OpenAI API integration",
              "LLM API integration",
              "prompt engineering",
              "embeddings APIs"
            ],
            "in_scope": "Azure OpenAI, prompt construction, chat completions, embeddings, function calling, streaming responses, rate limits, retries, SDK integration, API authentication",
            "name": "Foundation Model API Integration",
            "out_of_scope": "Training transformer models from scratch, cloud account governance, model deployment infrastructure, prompt policy review and legal approval workflows",
            "overlap_flags": [
              {
                "reason": "This overlaps with ML application tooling, but the focus here is hosted model API usage rather than model development.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "Azure OpenAI is delivered as an Azure service, so platform knowledge may also be relevant.",
                "with_dim_id": "cloud-platforms",
                "with_dim_name": null,
                "with_role": "Backend Engineer, Cybersecurity Engineer, Data Engineer, DevOps Engineer, ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Azure OpenAI",
          "placement_confidence": 0.92,
          "primary_dimension": "cloud-platforms",
          "reasoning": "Deterministic JD placement: locked_dimensions has 2 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "d_init_01"
          ],
          "skill_id": "azure-openai"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [
            "azure",
            "microsoft-azure"
          ],
          "related_to": [
            "google-cloud-platform",
            "aws",
            "azure-blob-storage",
            "azure-virtual-machines",
            "azure-virtual-network",
            "azure-devops-pipelines"
          ],
          "requires": [
            "azure-entra-id",
            "azure-key-vault",
            "azure-policy"
          ],
          "skill_id": "azure-openai",
          "suppress_on_match": []
        },
        "skill_id": "azure-openai",
        "split_log": [],
        "typed": {
          "alternatives_considered": [
            "Platform: ruled out \u2014 Azure is the platform, while Azure OpenAI is one hosted service within it.",
            "Tool: ruled out \u2014 it is consumed as a managed hosted offering, not software you run yourself."
          ],
          "confidence": 0.93,
          "name": "Azure OpenAI",
          "reasoning": "By the Service vs Platform rule, Azure OpenAI is a specific managed capability offered inside the Azure platform rather than the platform itself.",
          "skill_id": "azure-openai",
          "subtype": "ai_model_service",
          "type": "Service"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Anthropic",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "Anthropic",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Platform",
          "skill_nature": "PLATFORM",
          "sub_category": "ai_model_platform",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cAnthropic\u201d in JDs typically refers specifically to the Anthropic AI model platform/vendor, not another similarly named skill in the catalog."
          },
          "context_keywords": {
            "context_keywords": [
              "Claude",
              "AI safety",
              "alignment",
              "transformers",
              "machine learning",
              "natural language processing",
              "reinforcement learning",
              "ethical AI",
              "model interpretability",
              "scalability",
              "prompt engineering",
              "fine-tuning",
              "multi-modal",
              "data privacy",
              "human-in-the-loop"
            ]
          },
          "maturity": {
            "confidence": 0.84,
            "maturity": "emerging",
            "reasoning": "Anthropic/Claude is increasingly listed in AI engineer JDs and vendor docs, but it is not yet as universal as AWS/OpenAI; GitHub and job-market signals show rapid growth rather than saturation."
          },
          "skill_id": "anthropic",
          "vendor_license": {
            "confidence": 0.8,
            "license": "unknown",
            "vendor": "Anthropic",
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Vendor-specific model APIs and hosted AI services used to build applications on top of third-party foundation models. Anthropic belongs here because it refers to the Anthropic provider and its Claude model family, which AI engineers integrate through hosted inference and tooling APIs.",
            "exemplar_skills": [
              "Anthropic",
              "Claude API",
              "Claude models",
              "prompt engineering for Claude",
              "Anthropic SDK",
              "tool use with Claude"
            ],
            "in_scope": "Anthropic, Claude API, Claude models, hosted LLM inference, prompt/response APIs, tool use, model parameters, safety settings, rate limits, SDK integration",
            "name": "AI Model Provider Platforms",
            "out_of_scope": "Self-hosted model training, generic ML frameworks, cloud infrastructure, and vendor risk review processes owned by other dimensions",
            "overlap_flags": [
              {
                "reason": "Anthropic can also be discussed as a third-party AI vendor during procurement and compliance review, but this skill most directly refers to using the provider\u0027s model platform.",
                "with_dim_id": "ai-vendor-and-third-party-due-diligence",
                "with_dim_name": null,
                "with_role": "AI Compliance Officer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Anthropic",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "anthropic"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "aws-organizations-scps",
            "terraform",
            "pulumi",
            "github",
            "github-actions",
            "datadog",
            "sentry",
            "prometheus"
          ],
          "requires": [],
          "skill_id": "anthropic",
          "suppress_on_match": []
        },
        "skill_id": "anthropic",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.9,
          "name": "Anthropic",
          "reasoning": "By the Vendor SaaS = Platform rule, Anthropic is a hosted multi-tenant AI provider consumed via APIs rather than software you run yourself.",
          "skill_id": "anthropic",
          "subtype": "ai_model_platform",
          "type": "Platform"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "ML Frameworks and Libraries",
            "id": 40,
            "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "slug": "ml-frameworks-and-libraries",
            "source": "db"
          },
          "input_skill": "Hugging Face",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Hugging Face",
          "llm_role": null,
          "roles_from_db": []
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "ML Frameworks and Libraries",
            "id": 40,
            "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "slug": "ml-frameworks-and-libraries",
            "source": "db"
          },
          "input_skill": "Hugging Face",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Hugging Face",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Platform",
          "skill_nature": "PLATFORM",
          "sub_category": "ai_model_hub_platform",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cHugging Face\u201d in JDs typically refers to the specific AI model hub/platform (Transformers ecosystem), not another similarly named catalog skill."
          },
          "context_keywords": {
            "context_keywords": [
              "Transformers",
              "Tokenizers",
              "Datasets",
              "Model Hub",
              "Fine-tuning",
              "Pre-trained models",
              "Pipelines",
              "AutoModel",
              "Trainer",
              "Hugging Face Spaces",
              "Inference API",
              "BERT",
              "GPT-2",
              "T5",
              "NLU"
            ]
          },
          "maturity": {
            "confidence": 0.93,
            "maturity": "well_known",
            "reasoning": "Hugging Face appears in many ML/LLM job descriptions and has strong GitHub adoption; its Transformers ecosystem is a common default for model sharing and fine-tuning."
          },
          "skill_id": "hugging-face",
          "vendor_license": {
            "confidence": 0.95,
            "license": "apache_2",
            "vendor": "Hugging Face",
            "year_introduced": 2016
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Core libraries and SDKs used to build, train, fine-tune, evaluate, and run machine learning models. Hugging Face belongs here because it is a primary ecosystem for transformer models, model hubs, tokenizers, and inference workflows.",
            "exemplar_skills": [
              "Hugging Face",
              "Hugging Face Transformers",
              "Hugging Face Datasets",
              "Hugging Face Tokenizers",
              "model hub",
              "transformer fine-tuning",
              "text generation pipelines"
            ],
            "in_scope": "Hugging Face, Transformers, tokenizers, datasets, model loading, fine-tuning, inference pipelines, text generation, embeddings, model hub usage",
            "name": "ML Frameworks and Libraries",
            "out_of_scope": "Cloud deployment platforms, container orchestration, CI/CD automation, and production rollout controls, which belong to infrastructure and delivery dimensions",
            "overlap_flags": [
              {
                "reason": "Hugging Face models are often packaged and deployed through ML delivery pipelines, but the library itself is a framework skill.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "ml-frameworks-and-libraries"
          },
          {
            "description": "Libraries, hubs, and tooling centered on transformer-based NLP and multimodal model workflows. This fits Hugging Face when the emphasis is on model selection, fine-tuning, tokenization, and pretrained model reuse rather than general ML frameworks.",
            "exemplar_skills": [
              "Hugging Face",
              "Transformers",
              "tokenizers",
              "model cards",
              "pretrained checkpoints",
              "PEFT",
              "inference endpoints"
            ],
            "in_scope": "Hugging Face, Transformers library, pretrained checkpoints, tokenizers, model cards, pipelines, PEFT adapters, inference endpoints, embeddings models",
            "name": "Transformer Model Ecosystem",
            "out_of_scope": "General-purpose deep learning frameworks like PyTorch or TensorFlow, which own low-level tensor programming and training loops",
            "overlap_flags": [
              {
                "reason": "This is a narrower specialization of ML libraries focused on transformer-centric workflows.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          },
          {
            "description": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "exemplar_skills": [
              "ML Frameworks and Libraries"
            ],
            "in_scope": "Skills, tools, and practices that belong under ML Frameworks and Libraries for the target role, including items implied by the dimension rationale.",
            "name": "ML Frameworks and Libraries",
            "out_of_scope": "Adjacent clusters explicitly not owned by ML Frameworks and Libraries, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ml-frameworks-and-libraries"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Hugging Face",
          "placement_confidence": 0.92,
          "primary_dimension": "ml-frameworks-and-libraries",
          "reasoning": "Deterministic JD placement: locked_dimensions has 3 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "d_init_01"
          ],
          "skill_id": "hugging-face"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [],
          "requires": [],
          "skill_id": "hugging-face",
          "suppress_on_match": []
        },
        "skill_id": "hugging-face",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "Hugging Face",
          "reasoning": "By the Platform vs Tool rule, Hugging Face is primarily a hosted multi-tenant environment with APIs and managed services (e.g., model hub/inference), so it fits Platform rather than a user-run tool.",
          "skill_id": "hugging-face",
          "subtype": "ai_model_hub_platform",
          "type": "Platform"
        },
        "warnings": [
          "stage3_post_filter_dropped_catalog_only_locked_dims:42-\u003e3"
        ]
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "PostgreSQL",
          "alias_type": "CANONICAL",
          "id": 121,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PG 13",
          "alias_type": "VERSION",
          "id": 122,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PG 14",
          "alias_type": "VERSION",
          "id": 123,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PG 15",
          "alias_type": "VERSION",
          "id": 124,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PG 16",
          "alias_type": "VERSION",
          "id": 125,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PostgreSQL 13",
          "alias_type": "VERSION",
          "id": 130,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PostgreSQL 14",
          "alias_type": "VERSION",
          "id": 131,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PostgreSQL 15",
          "alias_type": "VERSION",
          "id": 132,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "PostgreSQL 16",
          "alias_type": "VERSION",
          "id": 133,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Postgres 13",
          "alias_type": "VERSION",
          "id": 126,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Postgres 14",
          "alias_type": "VERSION",
          "id": 127,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Postgres 15",
          "alias_type": "VERSION",
          "id": 128,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        },
        {
          "alias_text": "Postgres 16",
          "alias_type": "VERSION",
          "id": 129,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 3,
        "display_name": "PostgreSQL",
        "id": 16,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "postgresql",
        "sub_category_id": 29,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Relational Database Design",
            "id": 4,
            "rationale": "Modeling and operating relational persistence for backend services. Includes schema design, normalization, indexing, transactions, and query tuning for operational data stores.",
            "slug": "relational-database-design",
            "source": "db"
          },
          "input_skill": "PostgreSQL",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "PostgreSQL",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "MySQL",
          "alias_type": "CANONICAL",
          "id": 134,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 3,
        "display_name": "MySQL",
        "id": 17,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "mysql",
        "sub_category_id": 29,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Relational Database Design",
            "id": 4,
            "rationale": "Modeling and operating relational persistence for backend services. Includes schema design, normalization, indexing, transactions, and query tuning for operational data stores.",
            "slug": "relational-database-design",
            "source": "db"
          },
          "input_skill": "MySQL",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "MySQL",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Git",
          "alias_type": "CANONICAL",
          "id": 1613,
          "is_primary": false,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 13,
        "display_name": "Git",
        "id": 1002,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "git",
        "sub_category_id": 730,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Git",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "Git",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD Pipeline Platforms",
            "id": 150,
            "rationale": "Systems used to define, run, and maintain automated build and deployment workflows. This cluster is coherent because the role owns delivery automation end to end, including pipeline reliability and promotion logic.",
            "slug": "ci-cd-pipeline-platforms",
            "source": "db"
          },
          "input_skill": "CI/CD",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD for Machine Learning",
            "id": 56,
            "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "slug": "ci-cd-for-machine-learning",
            "source": "db"
          },
          "input_skill": "CI/CD",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD Pipeline Platforms",
            "id": 150,
            "rationale": "Systems used to define, run, and maintain automated build and deployment workflows. This cluster is coherent because the role owns delivery automation end to end, including pipeline reliability and promotion logic.",
            "slug": "ci-cd-pipeline-platforms",
            "source": "db"
          },
          "input_skill": "CI/CD",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD for Machine Learning",
            "id": 56,
            "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "slug": "ci-cd-for-machine-learning",
            "source": "db"
          },
          "input_skill": "CI/CD",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "CI/CD",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Methodology",
          "skill_nature": "METHODOLOGY",
          "sub_category": "ci_cd_process",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "CI/CD is a standard, unambiguous term for continuous integration and delivery/deployment processes."
          },
          "context_keywords": {
            "context_keywords": [
              "Jenkins",
              "GitLab CI",
              "CircleCI",
              "Travis CI",
              "Docker",
              "Kubernetes",
              "Terraform",
              "Ansible",
              "automated testing",
              "deployment pipelines",
              "continuous integration",
              "continuous deployment",
              "version control",
              "build automation",
              "monitoring"
            ]
          },
          "maturity": {
            "confidence": 0.97,
            "maturity": "well_known",
            "reasoning": "CI/CD appears in a large share of software engineering JDs and is a standard requirement across DevOps, platform, and backend roles; major vendors like GitHub, GitLab, and AWS all center product roadmaps on CI/CD pipelines."
          },
          "skill_id": "ci-cd",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [
          {
            "a_dim_id": "ci-cd-pipeline-platforms",
            "a_name": "CI/CD Pipeline Platforms",
            "a_role": "__skill_focal__",
            "b_dim_id": "ci-cd-for-machine-learning",
            "b_name": "CI/CD for Machine Learning",
            "b_role": "__skill_focal__",
            "pair_kind": "intra_role",
            "reasoning": "Dim A is generic delivery automation: build/test/deploy workflows with GitHub Actions, Jenkins, CircleCI, and pipeline-as-code. Dim B is ML-specific automation for model integration, testing, and deployment. A senior CI/CD platform engineer is not automatically a senior ML pipeline engineer because ML pipelines involve model lifecycle and ML tooling beyond standard app delivery. career-track: no, because generic CI/CD expertise does not naturally transfer to ML pipeline specialization.",
            "similarity": 0.6895055623802784
          },
          {
            "a_dim_id": "ci-cd-pipeline-platforms",
            "a_name": "CI/CD Pipeline Platforms",
            "a_role": "__skill_focal__",
            "b_dim_id": "ci-cd-pipeline-platforms",
            "b_name": "CI/CD Pipeline Platforms",
            "b_role": "DevOps Engineer",
            "pair_kind": "cross_role",
            "reasoning": "A is the platform/tooling cluster for CI/CD systems like GitHub Actions, GitLab CI, Jenkins, CircleCI, and pipeline-as-code, with release gating/canary/rollback explicitly out of scope. B extends to DevOps ownership of delivery automation end to end, including pipeline reliability and promotion logic. Those extra concerns are not in A. career-track: no, because a senior CI/CD platform engineer is not automatically a senior DevOps engineer handling release/promotion policy.",
            "similarity": 0.8623776544904218
          },
          {
            "a_dim_id": "ci-cd-pipeline-platforms",
            "a_name": "CI/CD Pipeline Platforms",
            "a_role": "__skill_focal__",
            "b_dim_id": "ci-cd-for-machine-learning",
            "b_name": "CI/CD for Machine Learning",
            "b_role": "ML Engineer",
            "pair_kind": "cross_role",
            "reasoning": "Dim A is general CI/CD platform engineering: automated build, test, and deployment workflows with exemplars like GitHub Actions, GitLab CI, Jenkins, Azure DevOps Pipelines, and CircleCI. Dim B is ML-specific CI/CD for automating model integration, testing, and deployment pipelines. The overlap is only at the delivery-automation layer; ML CI/CD adds model/data validation and ML release workflows. career-track: no, because a senior general CI/CD engineer is not automatically a senior ML pipeline engineer.",
            "similarity": 0.6635829321971303
          }
        ],
        "locked_dimensions": [
          {
            "description": "Systems used to define, run, and maintain automated build, test, and deployment workflows. CI/CD belongs here because it refers to the delivery automation layer that orchestrates code integration and release execution.",
            "exemplar_skills": [
              "CI/CD",
              "GitHub Actions",
              "GitLab CI",
              "Jenkins",
              "Azure DevOps Pipelines",
              "CircleCI",
              "pipeline-as-code",
              "build automation",
              "deployment automation"
            ],
            "in_scope": "CI/CD, GitHub Actions, GitLab CI, Jenkins, Azure DevOps Pipelines, CircleCI, build automation, test automation, deployment automation, pipeline-as-code",
            "name": "CI/CD Pipeline Platforms",
            "out_of_scope": "Release gating, canary rollout strategy, rollback policy, and environment promotion rules, which belong to deployment and release patterns; cloud infrastructure provisioning, which belongs to infrastructure as code",
            "overlap_flags": [
              {
                "reason": "CI/CD pipelines often trigger releases, but rollout strategy and promotion controls are a separate concern.",
                "with_dim_id": "deployment-and-release-patterns",
                "with_dim_name": null,
                "with_role": "Cloud Architect"
              },
              {
                "reason": "Pipelines may provision environments, but declarative infrastructure definition is owned by IaC.",
                "with_dim_id": "infrastructure-as-code",
                "with_dim_name": null,
                "with_role": "Cloud Architect, DevOps Engineer"
              }
            ],
            "tentative_id": "ci-cd-pipeline-platforms"
          },
          {
            "description": "Automation for validating, packaging, and deploying ML code, models, and related artifacts. CI/CD can fit here in an AI Engineer context when the focus is on model training, evaluation, and model release workflows rather than generic software delivery.",
            "exemplar_skills": [
              "CI/CD",
              "ML pipeline automation",
              "model validation",
              "model packaging",
              "model deployment",
              "model registry integration",
              "experiment-to-production workflows"
            ],
            "in_scope": "CI/CD, ML pipeline automation, model validation, model packaging, model deployment, experiment-to-production workflows, training pipeline checks, model registry integration",
            "name": "CI/CD for Machine Learning",
            "out_of_scope": "General application build and deploy pipelines, which belong to CI/CD pipeline platforms; model rollout governance and rollback policy, which belong to deployment rollouts and release control",
            "overlap_flags": [
              {
                "reason": "ML delivery pipelines use the same automation concepts as general CI/CD, but are specialized around models and training artifacts.",
                "with_dim_id": "ci-cd-pipeline-platforms",
                "with_dim_name": null,
                "with_role": "DevOps Engineer"
              },
              {
                "reason": "Model promotion and rollback are related, but this dimension focuses on the automation pipeline rather than release governance.",
                "with_dim_id": "deployment-rollouts-and-release-control",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "ci-cd-for-machine-learning"
          },
          {
            "description": "Systems used to define, run, and maintain automated build and deployment workflows. This cluster is coherent because the role owns delivery automation end to end, including pipeline reliability and promotion logic.",
            "exemplar_skills": [
              "CI/CD Pipeline Platforms"
            ],
            "in_scope": "Skills, tools, and practices that belong under CI/CD Pipeline Platforms for the target role, including items implied by the dimension rationale.",
            "name": "CI/CD Pipeline Platforms",
            "out_of_scope": "Adjacent clusters explicitly not owned by CI/CD Pipeline Platforms, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ci-cd-pipeline-platforms"
          },
          {
            "description": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "exemplar_skills": [
              "CI/CD for Machine Learning"
            ],
            "in_scope": "Skills, tools, and practices that belong under CI/CD for Machine Learning for the target role, including items implied by the dimension rationale.",
            "name": "CI/CD for Machine Learning",
            "out_of_scope": "Adjacent clusters explicitly not owned by CI/CD for Machine Learning, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ci-cd-for-machine-learning"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "CI/CD",
          "placement_confidence": 0.92,
          "primary_dimension": "ci-cd-pipeline-platforms",
          "reasoning": "Deterministic JD placement: locked_dimensions has 4 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "ci-cd-for-machine-learning"
          ],
          "skill_id": "ci-cd"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "jenkins",
            "gitlab-ci",
            "azure-devops-pipelines",
            "github-actions",
            "blue-green-deployment",
            "docker",
            "kubernetes",
            "terraform-cdk"
          ],
          "requires": [],
          "skill_id": "ci-cd",
          "suppress_on_match": []
        },
        "skill_id": "ci-cd",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "CI/CD",
          "reasoning": "CI/CD is fundamentally a way of working for automating build, test, and deployment pipelines, so by the Concept vs Methodology rule it is a Methodology.",
          "skill_id": "ci-cd",
          "subtype": "ci_cd_process",
          "type": "Methodology"
        },
        "warnings": [
          "stage3_post_filter_dropped_catalog_only_locked_dims:42-\u003e4"
        ]
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Docker",
          "alias_type": "CANONICAL",
          "id": 198,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 13,
        "display_name": "Docker",
        "id": 61,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "docker",
        "sub_category_id": 654,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Containerization and Image Builds",
            "id": 152,
            "rationale": "Container image creation, tagging, hardening, and registry workflows used to package services for deployment. This is coherent because DevOps often owns the build-to-image path that feeds runtime environments.",
            "slug": "containerization-and-image-builds",
            "source": "db"
          },
          "input_skill": "Docker",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Deployment and Runtime Configuration",
            "id": 13,
            "rationale": "Configuration and release artifacts that control how backend services run in environments. Includes environment variables, manifests, feature flags, and release-safe configuration management.",
            "slug": "deployment-and-runtime-configuration",
            "source": "db"
          },
          "input_skill": "Docker",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Docker",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "MLflow",
          "alias_type": "CANONICAL",
          "id": 470,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 13,
        "display_name": "MLflow",
        "id": 214,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "TOOL",
        "slug": "mlflow",
        "sub_category_id": 193,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "MLOps Platforms and Lifecycle",
            "id": 43,
            "rationale": "End-to-end managed platforms used to train, deploy, register, and govern models across their lifecycle. This is the operational control plane for production ML workflows.",
            "slug": "mlops-platforms-and-lifecycle",
            "source": "db"
          },
          "input_skill": "MLflow",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "MLflow",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "AWS",
          "alias_type": "CANONICAL",
          "id": 406,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "AWS",
        "id": 187,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "aws",
        "sub_category_id": 46,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Platforms",
            "id": 20,
            "rationale": "Proficiency in major cloud service provider platforms and their core services.",
            "slug": "cloud-platforms",
            "source": "db"
          },
          "input_skill": "AWS",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            },
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            },
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            },
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            },
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Provider Platforms",
            "id": 131,
            "rationale": "Major cloud platforms and their core service ecosystems used to design target-state architectures, choose deployment boundaries, and evaluate managed capabilities. This is the primary substrate for cloud architecture decisions.",
            "slug": "cloud-provider-platforms",
            "source": "db"
          },
          "input_skill": "AWS",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Cloud Architect",
              "id": 9,
              "rationale": null,
              "role_archetype": null,
              "slug": "cloud-architect",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Security Posture Tools",
            "id": 64,
            "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
            "slug": "cloud-security-posture-tools",
            "source": "db"
          },
          "input_skill": "AWS",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "AWS",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Azure",
          "alias_type": "CANONICAL",
          "id": 407,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "Azure",
        "id": 188,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "azure",
        "sub_category_id": 46,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Platforms",
            "id": 20,
            "rationale": "Proficiency in major cloud service provider platforms and their core services.",
            "slug": "cloud-platforms",
            "source": "db"
          },
          "input_skill": "Azure",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            },
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            },
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            },
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            },
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Provider Platforms",
            "id": 131,
            "rationale": "Major cloud platforms and their core service ecosystems used to design target-state architectures, choose deployment boundaries, and evaluate managed capabilities. This is the primary substrate for cloud architecture decisions.",
            "slug": "cloud-provider-platforms",
            "source": "db"
          },
          "input_skill": "Azure",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Cloud Architect",
              "id": 9,
              "rationale": null,
              "role_archetype": null,
              "slug": "cloud-architect",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Security Posture Tools",
            "id": 64,
            "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
            "slug": "cloud-security-posture-tools",
            "source": "db"
          },
          "input_skill": "Azure",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Azure",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "GCP",
          "alias_type": "CANONICAL",
          "id": 405,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "GCP",
        "id": 186,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "gcp",
        "sub_category_id": 46,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Platforms",
            "id": 20,
            "rationale": "Proficiency in major cloud service provider platforms and their core services.",
            "slug": "cloud-platforms",
            "source": "db"
          },
          "input_skill": "GCP",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Backend Engineer",
              "id": 1,
              "rationale": null,
              "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
              "slug": "backend-engineer",
              "source": "db"
            },
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            },
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            },
            {
              "display_name": "DevOps Engineer",
              "id": 10,
              "rationale": null,
              "role_archetype": null,
              "slug": "devops-engineer",
              "source": "db"
            },
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Cloud Security Posture Tools",
            "id": 64,
            "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
            "slug": "cloud-security-posture-tools",
            "source": "db"
          },
          "input_skill": "GCP",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Cybersecurity Engineer",
              "id": 5,
              "rationale": null,
              "role_archetype": null,
              "slug": "cybersecurity-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "GCP",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Power BI",
          "alias_type": "CANONICAL",
          "id": 360,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "Power BI",
        "id": 151,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "power-bi",
        "sub_category_id": 111,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "BI and Visualization Tools",
            "id": 31,
            "rationale": "Tools used to expose curated data to analysts and business users through dashboards, reports, and semantic exploration. Data engineers support these tools by shaping reliable datasets and performant models.",
            "slug": "bi-and-visualization-tools",
            "source": "db"
          },
          "input_skill": "Power BI",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Power BI",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [
        {
          "alias_text": "Tableau",
          "alias_type": "CANONICAL",
          "id": 359,
          "is_primary": true,
          "match_strategy": "CASE_INSENSITIVE"
        }
      ],
      "canonical": {
        "category_id": 9,
        "display_name": "Tableau",
        "id": 150,
        "is_also_category": false,
        "is_extractable": true,
        "skill_nature": "PLATFORM",
        "slug": "tableau",
        "sub_category_id": 111,
        "typical_lifespan": "EVERGREEN",
        "volatility": "STABLE"
      },
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "BI and Visualization Tools",
            "id": 31,
            "rationale": "Tools used to expose curated data to analysts and business users through dashboards, reports, and semantic exploration. Data engineers support these tools by shaping reliable datasets and performant models.",
            "slug": "bi-and-visualization-tools",
            "source": "db"
          },
          "input_skill": "Tableau",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Tableau",
      "matched_via": "alias",
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": null,
      "source_tag": "db",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "CrewAI",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "CrewAI",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Framework",
          "skill_nature": "FRAMEWORK",
          "sub_category": "agent_orchestration_framework",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "CrewAI is a specific agent orchestration framework name; unlikely to be confused with other catalog skills."
          },
          "context_keywords": {
            "context_keywords": [
              "agent orchestration",
              "multi-agent systems",
              "AI coordination",
              "task management",
              "workflow automation",
              "decision-making",
              "real-time collaboration",
              "scalability",
              "intelligent agents",
              "API integration",
              "event-driven architecture",
              "data flow",
              "system interoperability",
              "performance optimization",
              "cloud deployment"
            ]
          },
          "maturity": {
            "confidence": 0.84,
            "maturity": "emerging",
            "reasoning": "CrewAI is appearing in a growing number of AI-agent job postings and GitHub repos, but it is far from a universal hiring staple compared with established frameworks."
          },
          "skill_id": "crewai",
          "vendor_license": {
            "confidence": 0.9,
            "license": "apache_2",
            "vendor": "CrewAI, Inc.",
            "year_introduced": 2021
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Frameworks for building and coordinating LLM-powered agents that can plan, delegate tasks, call tools, and collaborate in workflows. CrewAI belongs here because it is specifically used to define multi-agent systems and orchestrate agent interactions.",
            "exemplar_skills": [
              "CrewAI",
              "LangChain agents",
              "AutoGen",
              "agent orchestration",
              "multi-agent systems",
              "tool use for LLM agents"
            ],
            "in_scope": "CrewAI, agent roles and delegation, multi-agent workflows, tool calling, task planning, agent collaboration, autonomous agent loops, LLM agent orchestration, agent memory patterns",
            "name": "Agent Orchestration Frameworks",
            "out_of_scope": "Core model training and fine-tuning, prompt engineering for single-turn chat, vector database operations, UI chat components, cloud deployment plumbing",
            "overlap_flags": [
              {
                "reason": "Agent frameworks often sit on top of ML/LLM libraries, but this dimension is about orchestration rather than model implementation.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "Agent applications may be deployed through ML pipelines, but deployment automation is not the core skill here.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "CrewAI",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "crewai"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "apollo-client",
            "istio",
            "jenkins",
            "gitlab",
            "docker",
            "ansible",
            "axios",
            "datadog"
          ],
          "requires": [],
          "skill_id": "crewai",
          "suppress_on_match": []
        },
        "skill_id": "crewai",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.88,
          "name": "CrewAI",
          "reasoning": "CrewAI is best classified as a Framework because users build applications and agent workflows inside it rather than merely operating it as standalone software.",
          "skill_id": "crewai",
          "subtype": "agent_orchestration_framework",
          "type": "Framework"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "APIs",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "APIs",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Protocol",
          "skill_nature": "PROTOCOL",
          "sub_category": "application_programming_interfaces",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cAPIs\u201d is a broad, standard term for application programming interfaces and is unlikely to be confused with another distinct catalog skill."
          },
          "context_keywords": {
            "context_keywords": [
              "REST",
              "SOAP",
              "GraphQL",
              "JSON",
              "XML",
              "OAuth",
              "JWT",
              "API Gateway",
              "Webhooks",
              "Rate Limiting",
              "Microservices",
              "Endpoint",
              "Swagger",
              "Postman",
              "Throttling"
            ]
          },
          "maturity": {
            "confidence": 0.98,
            "maturity": "well_known",
            "reasoning": "APIs are a hiring-pipeline staple across backend, mobile, and platform JDs; REST/GraphQL/API design appears in large volumes of job postings and vendor docs, indicating broad adoption."
          },
          "skill_id": "apis",
          "vendor_license": {
            "confidence": 0.9,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Designing, consuming, and integrating application programming interfaces across services and clients. This fits the target skill because APIs are the primary contract for exchanging data and invoking capabilities between systems.",
            "exemplar_skills": [
              "APIs",
              "REST APIs",
              "GraphQL",
              "gRPC",
              "Webhook design",
              "API integration"
            ],
            "in_scope": "APIs, REST APIs, GraphQL APIs, gRPC endpoints, webhook contracts, request and response schemas, versioning, pagination, authentication for API access, API client integration",
            "name": "API Design and Integration",
            "out_of_scope": "UI component frameworks, database schema design, container orchestration, model training pipelines, observability tooling, which belong to other dimensions",
            "overlap_flags": [
              {
                "reason": "API work often includes auth headers, tokens, and session-bound access control.",
                "with_dim_id": "authentication-and-session-handling",
                "with_dim_name": null,
                "with_role": "Android Engineer, Frontend Engineer, Hybrid Mobile Developer, Ios engineer"
              },
              {
                "reason": "Platform services frequently expose and consume APIs, but the core skill here is interface design and integration.",
                "with_dim_id": "cloud-platforms",
                "with_dim_name": null,
                "with_role": "Backend Engineer, Cybersecurity Engineer, Data Engineer, DevOps Engineer, ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "APIs",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "apis"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "fetch-api",
            "axios",
            "apollo-client",
            "oauth-2-0",
            "refresh-tokens",
            "apns",
            "history-api",
            "repository-pattern"
          ],
          "requires": [],
          "skill_id": "apis",
          "suppress_on_match": []
        },
        "skill_id": "apis",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "APIs",
          "reasoning": "APIs are a communication interface between systems, so by the Protocol rule they fit best as a protocol-like standard for interaction rather than a tool or platform.",
          "skill_id": "apis",
          "subtype": "application_programming_interfaces",
          "type": "Protocol"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "LLMs",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "LLMs",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Concept",
          "skill_nature": "CONCEPT",
          "sub_category": "large_language_models",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cLLMs\u201d is a specific, widely used abbreviation for Large Language Models and is unlikely to be confused with other catalog skills."
          },
          "context_keywords": {
            "context_keywords": [
              "transformers",
              "GPT",
              "BERT",
              "fine-tuning",
              "tokenization",
              "NLP",
              "prompt engineering",
              "zero-shot learning",
              "transfer learning",
              "model training",
              "language generation",
              "contextual embeddings",
              "attention mechanism",
              "pre-trained models",
              "text classification"
            ]
          },
          "maturity": {
            "confidence": 0.91,
            "maturity": "emerging",
            "reasoning": "LLMs are increasingly listed in job descriptions for AI/ML and product roles, and major vendors (OpenAI, Anthropic, Google) are shipping APIs and platforms, but they are not yet universal across engineering hiring."
          },
          "skill_id": "llms",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Models and techniques for building, adapting, and using transformer-based language models. This fits the target skill because LLMs are the core model family behind modern generative AI systems, prompting, fine-tuning, and inference workflows.",
            "exemplar_skills": [
              "LLMs",
              "prompt engineering",
              "instruction tuning",
              "fine-tuning LLMs",
              "transformer models",
              "text generation",
              "embeddings"
            ],
            "in_scope": "LLMs, transformer language models, foundation models, prompt engineering, fine-tuning, instruction tuning, embeddings, tokenization, context windows, text generation, chat completion APIs",
            "name": "Large Language Models",
            "out_of_scope": "Traditional machine learning models such as linear regression or random forests, distributed training infrastructure for scaling across GPUs, deployment pipelines for ML systems",
            "overlap_flags": [
              {
                "reason": "LLMs are implemented and trained with ML libraries, but this dimension is about the model family rather than the general framework stack.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "LLM workflows may be deployed through ML pipelines, but release automation is separate from the model concept itself.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "LLMs",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "llms"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "managed-databases",
            "service-workers",
            "testing-library",
            "javascript",
            "jvm",
            "gradle",
            "jenkins",
            "mtls"
          ],
          "requires": [],
          "skill_id": "llms",
          "suppress_on_match": []
        },
        "skill_id": "llms",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.96,
          "name": "LLMs",
          "reasoning": "LLMs are a named knowledge unit about a class of models, so by the Concept vs Methodology rule they are a Concept rather than a tool, framework, or platform.",
          "skill_id": "llms",
          "subtype": "large_language_models",
          "type": "Concept"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "RAG",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "RAG",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Concept",
          "skill_nature": "CONCEPT",
          "sub_category": "retrieval_augmented_generation",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cRAG\u201d in JDs typically and specifically refers to Retrieval-Augmented Generation; unlikely to be confused with other catalog skills."
          },
          "context_keywords": {
            "context_keywords": [
              "retrieval",
              "generation",
              "contextualization",
              "fine-tuning",
              "prompt engineering",
              "knowledge integration",
              "data augmentation",
              "model training",
              "information retrieval",
              "transformer models",
              "semantic search",
              "natural language processing",
              "machine learning",
              "AI applications",
              "user intent"
            ]
          },
          "maturity": {
            "confidence": 0.89,
            "maturity": "emerging",
            "reasoning": "RAG appears in many recent AI/ML job descriptions and vendor docs, but it is still not a universal baseline skill like Python or SQL; market demand is rising fast rather than fully standardized."
          },
          "skill_id": "rag",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Techniques for grounding LLM outputs in retrieved external context. This includes building retrieval pipelines, chunking and indexing content, and combining search results with generation, which is exactly what RAG refers to.",
            "exemplar_skills": [
              "RAG",
              "retrieval augmented generation",
              "vector search",
              "embedding retrieval",
              "reranking",
              "chunking strategies",
              "hybrid search",
              "context grounding"
            ],
            "in_scope": "RAG, retrieval augmented generation, document chunking, embedding-based retrieval, vector search, reranking, context assembly, prompt grounding, hybrid search, retrieval pipelines, knowledge-grounded generation",
            "name": "Retrieval Augmented Generation",
            "out_of_scope": "Core LLM model training, distributed training systems, generic NLP preprocessing, UI integration for chat apps, data governance and lineage",
            "overlap_flags": [
              {
                "reason": "RAG implementations often use ML libraries and model APIs, but the dimension is about retrieval-grounded generation rather than model definition or training.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "RAG systems may index governed content, but lineage and metadata management are separate from the retrieval-generation pattern itself.",
                "with_dim_id": "data-lineage-and-metadata",
                "with_dim_name": null,
                "with_role": "Data Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "RAG",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "rag"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "git",
            "github",
            "gitlab",
            "redux",
            "linkerd",
            "puppet",
            "saltstack",
            "ecs"
          ],
          "requires": [],
          "skill_id": "rag",
          "suppress_on_match": []
        },
        "skill_id": "rag",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "RAG",
          "reasoning": "RAG is fundamentally a named AI knowledge pattern for combining retrieval with generation, so it fits the Concept category rather than a tool, framework, or architecture.",
          "skill_id": "rag",
          "subtype": "retrieval_augmented_generation",
          "type": "Concept"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "ML Frameworks and Libraries",
            "id": 40,
            "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "slug": "ml-frameworks-and-libraries",
            "source": "db"
          },
          "input_skill": "Embeddings",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Embeddings",
          "llm_role": null,
          "roles_from_db": []
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Systems Programming",
            "id": 166,
            "rationale": "Systems programming covers low-level software development where performance, memory safety, and direct control over resources matter. Rust fits here because it is commonly used for OS-adjacent services, infrastructure components, and other performance-sensitive systems code.",
            "slug": "d_init_02",
            "source": "db"
          },
          "input_skill": "Embeddings",
          "llm_role": null,
          "roles_from_db": []
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "ML Frameworks and Libraries",
            "id": 40,
            "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "slug": "ml-frameworks-and-libraries",
            "source": "db"
          },
          "input_skill": "Embeddings",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Embeddings",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Concept",
          "skill_nature": "CONCEPT",
          "sub_category": "vector_representation",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cEmbeddings\u201d in JDs typically refers to vector representations for ML/NLP, not a distinct catalog skill with a similar name."
          },
          "context_keywords": {
            "context_keywords": [
              "vector space",
              "semantic similarity",
              "word embeddings",
              "sentence embeddings",
              "transformers",
              "BERT",
              "GloVe",
              "fastText",
              "dimensionality reduction",
              "nearest neighbors",
              "contextual embeddings",
              "feature extraction",
              "transfer learning",
              "natural language processing",
              "deep learning"
            ]
          },
          "maturity": {
            "confidence": 0.93,
            "maturity": "well_known",
            "reasoning": "Embeddings are a standard ML concept and appear widely in JDs for search, recommendation, and LLM/RAG roles; major vendors like OpenAI, Cohere, and AWS expose embedding APIs, signaling broad adoption."
          },
          "skill_id": "embeddings",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Core ML libraries and model representations used to build, train, and run predictive systems. Embeddings belong here because they are a fundamental learned representation used by ML models and downstream inference workflows.",
            "exemplar_skills": [
              "Embeddings",
              "PyTorch",
              "TensorFlow",
              "JAX",
              "scikit-learn",
              "representation learning"
            ],
            "in_scope": "Embeddings, vector representations, feature tensors, PyTorch, TensorFlow, JAX, scikit-learn model APIs, training and inference model code, representation learning",
            "name": "Machine Learning Model Representations",
            "out_of_scope": "Vector databases and ANN indexing, prompt engineering, deployment pipelines, cloud infrastructure, data governance, application UI frameworks",
            "overlap_flags": [
              {
                "reason": "Embedding datasets and feature provenance may be tracked for governance, but the core skill is model representation rather than metadata management.",
                "with_dim_id": "data-lineage-and-metadata",
                "with_dim_name": null,
                "with_role": "Data Engineer"
              }
            ],
            "tentative_id": "ml-frameworks-and-libraries"
          },
          {
            "description": "Techniques for storing, indexing, and querying dense vectors to support semantic search and retrieval-augmented systems. Embeddings belong here because they are the primary vector representation used in similarity search and retrieval pipelines.",
            "exemplar_skills": [
              "Embeddings",
              "semantic search",
              "vector similarity search",
              "nearest-neighbor retrieval",
              "ANN indexing",
              "vector databases"
            ],
            "in_scope": "Embeddings, vector similarity search, nearest-neighbor retrieval, cosine similarity, ANN indexes, semantic search, retrieval-augmented generation, vector databases, reranking inputs",
            "name": "Vector Search and Retrieval",
            "out_of_scope": "Model training frameworks, tokenization, prompt design, database schema design, general search engine ranking, cloud hosting and deployment",
            "overlap_flags": [
              {
                "reason": "Embeddings are learned in ML frameworks, but this dimension focuses on using them for retrieval and search.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          },
          {
            "description": "Methods for comparing meaning across text, images, or other modalities using dense representations. Embeddings fit here because they encode semantic relationships that power clustering, matching, classification, and recommendation.",
            "exemplar_skills": [
              "Embeddings",
              "sentence embeddings",
              "document embeddings",
              "cosine similarity",
              "contrastive learning",
              "semantic matching"
            ],
            "in_scope": "Embeddings, similarity scoring, cosine distance, contrastive learning, sentence embeddings, document embeddings, multimodal embeddings, clustering with vectors, semantic matching",
            "name": "Semantic Similarity Modeling",
            "out_of_scope": "Indexing infrastructure, database operations, model serving platforms, prompt templates, UI search components, compliance or governance workflows",
            "overlap_flags": [
              {
                "reason": "The modeling techniques are implemented with ML frameworks, but the dimension centers on semantic representation and comparison.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_02"
          },
          {
            "description": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
            "exemplar_skills": [
              "ML Frameworks and Libraries"
            ],
            "in_scope": "Skills, tools, and practices that belong under ML Frameworks and Libraries for the target role, including items implied by the dimension rationale.",
            "name": "ML Frameworks and Libraries",
            "out_of_scope": "Adjacent clusters explicitly not owned by ML Frameworks and Libraries, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ml-frameworks-and-libraries"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Embeddings",
          "placement_confidence": 0.92,
          "primary_dimension": "ml-frameworks-and-libraries",
          "reasoning": "Deterministic JD placement: locked_dimensions has 4 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "d_init_01",
            "d_init_02"
          ],
          "skill_id": "embeddings"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "data-structures",
            "algorithms",
            "repository-pattern",
            "deep-links",
            "feature-modules",
            "memory-leaks",
            "idempotent-configuration"
          ],
          "requires": [],
          "skill_id": "embeddings",
          "suppress_on_match": []
        },
        "skill_id": "embeddings",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.9,
          "name": "Embeddings",
          "reasoning": "Embeddings are a named knowledge unit in machine learning representing how items are mapped into vector space, so by the Concept vs Methodology rule they are a Concept rather than a tool or format.",
          "skill_id": "embeddings",
          "subtype": "vector_representation",
          "type": "Concept"
        },
        "warnings": [
          "stage3_post_filter_dropped_catalog_only_locked_dims:43-\u003e4"
        ]
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD for Machine Learning",
            "id": 56,
            "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "slug": "ci-cd-for-machine-learning",
            "source": "db"
          },
          "input_skill": "MLOps",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Deployment Rollouts and Release Control",
            "id": 51,
            "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
            "slug": "deployment-rollouts-and-release-control",
            "source": "db"
          },
          "input_skill": "MLOps",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Data Lineage and Metadata",
            "id": 28,
            "rationale": "Cataloging, documenting, and tracing how data moves and changes across systems. This dimension supports impact analysis, governance, discoverability, and operational understanding of datasets.",
            "slug": "data-lineage-and-metadata",
            "source": "db"
          },
          "input_skill": "MLOps",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD for Machine Learning",
            "id": 56,
            "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "slug": "ci-cd-for-machine-learning",
            "source": "db"
          },
          "input_skill": "MLOps",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Deployment Rollouts and Release Control",
            "id": 51,
            "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
            "slug": "deployment-rollouts-and-release-control",
            "source": "db"
          },
          "input_skill": "MLOps",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Data Lineage and Metadata",
            "id": 28,
            "rationale": "Cataloging, documenting, and tracing how data moves and changes across systems. This dimension supports impact analysis, governance, discoverability, and operational understanding of datasets.",
            "slug": "data-lineage-and-metadata",
            "source": "db"
          },
          "input_skill": "MLOps",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "Data Engineer",
              "id": 2,
              "rationale": null,
              "role_archetype": null,
              "slug": "data-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "MLOps",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Methodology",
          "skill_nature": "METHODOLOGY",
          "sub_category": "mlops",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "MLOps is a specific, commonly used term for ML deployment/operations; unlikely to be confused with other catalog skills."
          },
          "context_keywords": {
            "context_keywords": [
              "Kubeflow",
              "MLflow",
              "Docker",
              "Kubernetes",
              "CI/CD",
              "data pipeline",
              "model deployment",
              "versioning",
              "monitoring",
              "automation",
              "scalability",
              "reproducibility",
              "cloud-native",
              "A/B testing",
              "data governance"
            ]
          },
          "maturity": {
            "confidence": 0.92,
            "maturity": "well_known",
            "reasoning": "MLOps appears in many job descriptions for ML/platform roles and is a standard practice in major cloud vendor docs (AWS, GCP, Azure) for CI/CD, model monitoring, and deployment."
          },
          "skill_id": "mlops",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [
          {
            "a_dim_id": "ci-cd-for-machine-learning",
            "a_name": "CI/CD for Machine Learning",
            "a_role": "__skill_focal__",
            "b_dim_id": "deployment-rollouts-and-release-control",
            "b_name": "Deployment Rollouts and Release Control",
            "b_role": "__skill_focal__",
            "pair_kind": "intra_role",
            "reasoning": "Dim A covers ML delivery pipelines: MLOps, model build pipelines, training job automation, model packaging, validation gates, artifact versioning, and model registry integration. Dim B covers release governance after a model is ready: safe promotion through environments, release gating, version pinning, rollout strategies, and rollback. A senior MLOps/pipeline engineer is not automatically a senior rollout/rollback specialist. career-track: no, because pipeline automation and release-control are adjacent but distinct.",
            "similarity": 0.7275503211597232
          },
          {
            "a_dim_id": "ci-cd-for-machine-learning",
            "a_name": "CI/CD for Machine Learning",
            "a_role": "__skill_focal__",
            "b_dim_id": "data-lineage-and-metadata",
            "b_name": "Data Lineage and Metadata",
            "b_role": "__skill_focal__",
            "pair_kind": "intra_role",
            "reasoning": "Dim A is ML delivery automation: build/test/validate/deploy pipelines, model packaging, model registry integration, and training workflow orchestration (e.g. MLOps, ML pipeline automation, automated ML testing). Dim B is data lineage/metadata: cataloging and tracing how data moves and changes for governance and impact analysis. A senior ML CI/CD engineer would not naturally be a senior lineage/metadata specialist; these are different career tracks and daily work. The similarity is broad data/ML platform overlap, not the same skill cluster.",
            "similarity": 0.7256127599825558
          },
          {
            "a_dim_id": "deployment-rollouts-and-release-control",
            "a_name": "Deployment Rollouts and Release Control",
            "a_role": "__skill_focal__",
            "b_dim_id": "data-lineage-and-metadata",
            "b_name": "Data Lineage and Metadata",
            "b_role": "__skill_focal__",
            "pair_kind": "intra_role",
            "reasoning": "Dim A centers on safe ML model release operations: canary releases, shadow deployments, blue-green rollout, rollback, traffic splitting, and staged promotion. Dim B centers on data governance: cataloging, documenting, and tracing data movement/changes for impact analysis and discoverability. They share an MLOps umbrella but the core skills and artifacts differ. career-track: no, because model rollout/release control is not the same senior track as data lineage/metadata management.",
            "similarity": 0.645282029688236
          },
          {
            "a_dim_id": "data-lineage-and-metadata",
            "a_name": "Data Lineage and Metadata",
            "a_role": "__skill_focal__",
            "b_dim_id": "data-lineage-and-metadata",
            "b_name": "Data Lineage and Metadata",
            "b_role": "Data Engineer",
            "pair_kind": "cross_role",
            "reasoning": "Names overlap, but clusters differ. Dim A is ML/MLOps lineage: dataset versioning, feature lineage, model provenance, experiment tracking, artifact tracking, and training data audit trails for reproducibility/auditability of model inputs/outputs. Dim B is data-engineering lineage/metadata: cataloging, documenting, and tracing how data moves across systems for impact analysis, governance, and discoverability. career-track: no, because senior ML lineage/MLOps work is not the same as senior enterprise data catalog/governance work.",
            "similarity": 0.7021450806313893
          }
        ],
        "locked_dimensions": [
          {
            "description": "Automated build, test, validation, and deployment workflows for ML code, models, and related artifacts. MLOps belongs here because it operationalizes the ML lifecycle with repeatable pipelines and promotion steps.",
            "exemplar_skills": [
              "MLOps",
              "ML pipeline automation",
              "model packaging",
              "model validation",
              "model registry",
              "training workflow orchestration",
              "automated ML testing"
            ],
            "in_scope": "MLOps, model build pipelines, training job automation, model packaging, model validation gates, artifact versioning, reproducible ML workflows, automated testing for ML code, pipeline orchestration, model registry integration",
            "name": "CI/CD for Machine Learning",
            "out_of_scope": "Distributed training frameworks and multi-node optimization, which belong to distributed-training-systems; cloud infrastructure provisioning, which belongs to infrastructure-as-code; model rollout policy and rollback strategy, which belong to deployment-rollouts-and-release-control",
            "overlap_flags": [
              {
                "reason": "MLOps often includes promotion and rollback mechanics, but that dimension owns release gating and rollout strategy.",
                "with_dim_id": "deployment-rollouts-and-release-control",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "MLOps pipelines frequently track artifacts and provenance, which can overlap with lineage and metadata management.",
                "with_dim_id": "data-lineage-and-metadata",
                "with_dim_name": null,
                "with_role": "Data Engineer"
              }
            ],
            "tentative_id": "ci-cd-for-machine-learning"
          },
          {
            "description": "Practices for safely promoting ML models through environments and controlling production exposure. MLOps fits here when the focus is on canarying, rollback, approvals, and release safety for model deployments.",
            "exemplar_skills": [
              "MLOps",
              "model canary release",
              "shadow deployment",
              "blue-green deployment",
              "model rollback",
              "staged model promotion",
              "traffic splitting"
            ],
            "in_scope": "MLOps, model canary releases, shadow deployments, blue-green model rollout, rollback procedures, release approvals, traffic splitting, staged model promotion, production monitoring for model changes",
            "name": "Deployment Rollouts and Release Control",
            "out_of_scope": "Training pipeline automation and CI/CD mechanics, which belong to ci-cd-for-machine-learning; model development frameworks and training APIs, which belong to ml-frameworks-and-libraries; infrastructure provisioning, which belongs to infrastructure-as-code",
            "overlap_flags": [
              {
                "reason": "Release control is often the final stage of an ML delivery pipeline, so the two dimensions overlap on deployment automation.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "Safe model rollout depends on monitoring and incident response signals during production changes.",
                "with_dim_id": "observability-and-incident-triage",
                "with_dim_name": null,
                "with_role": "DevOps Engineer"
              }
            ],
            "tentative_id": "deployment-rollouts-and-release-control"
          },
          {
            "description": "Tracking datasets, features, artifacts, and provenance across the ML lifecycle. MLOps belongs here when it emphasizes traceability, reproducibility, and auditability of model inputs and outputs.",
            "exemplar_skills": [
              "MLOps",
              "dataset versioning",
              "feature lineage",
              "model provenance",
              "experiment tracking",
              "artifact tracking",
              "training data audit trails"
            ],
            "in_scope": "MLOps, dataset versioning, feature lineage, model provenance, experiment metadata, artifact tracking, reproducibility records, pipeline lineage, training data audit trails",
            "name": "Data Lineage and Metadata",
            "out_of_scope": "Model training frameworks and inference APIs, which belong to ml-frameworks-and-libraries; deployment orchestration and release gating, which belong to ci-cd-for-machine-learning or deployment-rollouts-and-release-control; cloud infrastructure setup, which belongs to infrastructure-as-code",
            "overlap_flags": [
              {
                "reason": "ML delivery pipelines often emit lineage and metadata, so operational automation and traceability can intersect.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "data-lineage-and-metadata"
          },
          {
            "description": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "exemplar_skills": [
              "CI/CD for Machine Learning"
            ],
            "in_scope": "Skills, tools, and practices that belong under CI/CD for Machine Learning for the target role, including items implied by the dimension rationale.",
            "name": "CI/CD for Machine Learning",
            "out_of_scope": "Adjacent clusters explicitly not owned by CI/CD for Machine Learning, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ci-cd-for-machine-learning"
          },
          {
            "description": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
            "exemplar_skills": [
              "Deployment Rollouts and Release Control"
            ],
            "in_scope": "Skills, tools, and practices that belong under Deployment Rollouts and Release Control for the target role, including items implied by the dimension rationale.",
            "name": "Deployment Rollouts and Release Control",
            "out_of_scope": "Adjacent clusters explicitly not owned by Deployment Rollouts and Release Control, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "deployment-rollouts-and-release-control"
          },
          {
            "description": "Cataloging, documenting, and tracing how data moves and changes across systems. This dimension supports impact analysis, governance, discoverability, and operational understanding of datasets.",
            "exemplar_skills": [
              "Data Lineage and Metadata"
            ],
            "in_scope": "Skills, tools, and practices that belong under Data Lineage and Metadata for the target role, including items implied by the dimension rationale.",
            "name": "Data Lineage and Metadata",
            "out_of_scope": "Adjacent clusters explicitly not owned by Data Lineage and Metadata, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "data-lineage-and-metadata"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "MLOps",
          "placement_confidence": 0.92,
          "primary_dimension": "ci-cd-for-machine-learning",
          "reasoning": "Deterministic JD placement: locked_dimensions has 6 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "deployment-rollouts-and-release-control",
            "data-lineage-and-metadata"
          ],
          "skill_id": "mlops"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "github-actions",
            "jenkins",
            "kubernetes",
            "opentelemetry",
            "datadog",
            "autoscaling",
            "git",
            "runbooks"
          ],
          "requires": [],
          "skill_id": "mlops",
          "suppress_on_match": []
        },
        "skill_id": "mlops",
        "split_log": [],
        "typed": {
          "alternatives_considered": [
            "Concept: ruled out \u2014 it is not just a knowledge unit but an operational practice/process.",
            "Architecture: ruled out \u2014 it does not describe a system shape or deployment topology."
          ],
          "confidence": 0.93,
          "name": "MLOps",
          "reasoning": "MLOps is fundamentally a way of working that combines machine learning development and operations practices, so by the Concept vs Methodology rule it fits Methodology.",
          "skill_id": "mlops",
          "subtype": "mlops",
          "type": "Methodology"
        },
        "warnings": [
          "stage3_post_filter_dropped_catalog_only_locked_dims:43-\u003e6"
        ]
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Deployment Rollouts and Release Control",
            "id": 51,
            "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
            "slug": "deployment-rollouts-and-release-control",
            "source": "db"
          },
          "input_skill": "Model Versioning",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD for Machine Learning",
            "id": 56,
            "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "slug": "ci-cd-for-machine-learning",
            "source": "db"
          },
          "input_skill": "Model Versioning",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "CI/CD for Machine Learning",
            "id": 56,
            "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "slug": "ci-cd-for-machine-learning",
            "source": "db"
          },
          "input_skill": "Model Versioning",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        },
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "Deployment Rollouts and Release Control",
            "id": 51,
            "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
            "slug": "deployment-rollouts-and-release-control",
            "source": "db"
          },
          "input_skill": "Model Versioning",
          "llm_role": null,
          "roles_from_db": [
            {
              "display_name": "ML Engineer",
              "id": 3,
              "rationale": null,
              "role_archetype": null,
              "slug": "ml-engineer",
              "source": "db"
            }
          ]
        }
      ],
      "input_skill": "Model Versioning",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Concept",
          "skill_nature": "CONCEPT",
          "sub_category": "model_versioning",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cModel Versioning\u201d in JDs typically refers to tracking and managing ML model releases; it\u2019s not commonly confused with other catalog skills."
          },
          "context_keywords": {
            "context_keywords": [
              "MLflow",
              "DVC",
              "Git",
              "model registry",
              "version control",
              "reproducibility",
              "artifact management",
              "experiment tracking",
              "data lineage",
              "rollback",
              "deployment pipeline",
              "model governance",
              "continuous integration",
              "A/B testing",
              "model drift",
              "hyperparameter tuning"
            ]
          },
          "maturity": {
            "confidence": 0.86,
            "maturity": "well_known",
            "reasoning": "Common in MLOps job descriptions and platform docs; tools like MLflow, DVC, and SageMaker Model Registry are widely used for tracking model artifacts and rollout control."
          },
          "skill_id": "model-versioning",
          "vendor_license": {
            "confidence": 1.0,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [
          {
            "a_dim_id": "deployment-rollouts-and-release-control",
            "a_name": "Model Release Rollouts and Control",
            "a_role": "__skill_focal__",
            "b_dim_id": "ci-cd-for-machine-learning",
            "b_name": "Machine Learning Delivery Pipelines",
            "b_role": "__skill_focal__",
            "pair_kind": "intra_role",
            "reasoning": "Dim A is specifically about release governance for already-built ML models: model versioning, promotion, canary/shadow/blue-green rollout, rollback, release gating, champion-challenger deployment, and model registry release states. Dim B is broader pipeline automation for ML: automating integration, testing, and deployment pipelines. A senior practitioner in model rollout control is not necessarily a senior practitioner in CI/CD platform automation, because the former centers on operational release-state management of models while the latter centers on building and operating end-to-end delivery pipelines and tooling. career-track: no, because model release governance and ML CI/CD platform automation are adjacent but distinct specialties with different day-to-day work and depth of tooling.",
            "similarity": 0.6568248069903599
          }
        ],
        "locked_dimensions": [
          {
            "description": "Practices for promoting ML models safely through environments and controlling when a new model becomes active. Model versioning belongs here because version identifiers, staged promotion, rollback, and release gating are the operational mechanisms used to manage model changes.",
            "exemplar_skills": [
              "Model Versioning",
              "model promotion",
              "canary releases for models",
              "shadow deployments",
              "blue-green model rollout",
              "rollback to prior model versions",
              "model registry release states"
            ],
            "in_scope": "model versioning, model promotion, canary releases for models, shadow deployments, blue-green model rollout, rollback to prior model versions, release gating, champion-challenger deployment, model registry release states",
            "name": "Model Release Rollouts and Control",
            "out_of_scope": "training algorithms and hyperparameter tuning, feature engineering, inference serving internals, experiment design and A/B analysis, data versioning and dataset lineage",
            "overlap_flags": [
              {
                "reason": "Model versioning is often implemented inside ML delivery pipelines, but this dimension is about the release-control semantics rather than pipeline automation itself.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "Version identifiers may be tracked alongside lineage metadata, but lineage focuses on tracing data and artifacts rather than release promotion.",
                "with_dim_id": "data-lineage-and-metadata",
                "with_dim_name": null,
                "with_role": "Data Engineer"
              }
            ],
            "tentative_id": "deployment-rollouts-and-release-control"
          },
          {
            "description": "Automation for building, testing, registering, and deploying ML models through repeatable delivery workflows. Model versioning fits here when it is used as part of CI/CD mechanics for packaging artifacts, tagging builds, and moving models across environments.",
            "exemplar_skills": [
              "Model Versioning",
              "model registry integration",
              "artifact tagging",
              "pipeline-driven model promotion",
              "automated validation before deployment",
              "CI/CD workflows for ML models",
              "environment-specific model releases"
            ],
            "in_scope": "Model Versioning, model registry integration, artifact tagging, pipeline-driven model promotion, automated validation before deployment, CI/CD workflows for ML models, build and deploy orchestration, environment-specific model releases",
            "name": "Machine Learning Delivery Pipelines",
            "out_of_scope": "manual release governance and rollout strategy, training-time optimization, distributed training infrastructure, feature store design, data quality monitoring",
            "overlap_flags": [
              {
                "reason": "Both dimensions cover moving models through environments; this one emphasizes automation pipelines while the other emphasizes release strategy and control.",
                "with_dim_id": "deployment-rollouts-and-release-control",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "Versioned model artifacts may be tracked in metadata systems, but this dimension is about delivery automation rather than cataloging provenance.",
                "with_dim_id": "data-lineage-and-metadata",
                "with_dim_name": null,
                "with_role": "Data Engineer"
              }
            ],
            "tentative_id": "ci-cd-for-machine-learning"
          },
          {
            "description": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
            "exemplar_skills": [
              "CI/CD for Machine Learning"
            ],
            "in_scope": "Skills, tools, and practices that belong under CI/CD for Machine Learning for the target role, including items implied by the dimension rationale.",
            "name": "CI/CD for Machine Learning",
            "out_of_scope": "Adjacent clusters explicitly not owned by CI/CD for Machine Learning, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "ci-cd-for-machine-learning"
          },
          {
            "description": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
            "exemplar_skills": [
              "Deployment Rollouts and Release Control"
            ],
            "in_scope": "Skills, tools, and practices that belong under Deployment Rollouts and Release Control for the target role, including items implied by the dimension rationale.",
            "name": "Deployment Rollouts and Release Control",
            "out_of_scope": "Adjacent clusters explicitly not owned by Deployment Rollouts and Release Control, including unrelated platforms, roles, and skill families per library policy.",
            "overlap_flags": [],
            "tentative_id": "deployment-rollouts-and-release-control"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Model Versioning",
          "placement_confidence": 0.92,
          "primary_dimension": "deployment-rollouts-and-release-control",
          "reasoning": "Deterministic JD placement: locked_dimensions has 4 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [
            "ci-cd-for-machine-learning"
          ],
          "skill_id": "model-versioning"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "rolling-update",
            "rollback-automation",
            "rollback-plan",
            "build-variants",
            "git",
            "github",
            "idempotent-configuration",
            "autoscaling"
          ],
          "requires": [],
          "skill_id": "model-versioning",
          "suppress_on_match": []
        },
        "skill_id": "model-versioning",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.93,
          "name": "Model Versioning",
          "reasoning": "Model Versioning is fundamentally a named knowledge unit about tracking and managing model revisions, so it fits the Concept category rather than a tool or methodology.",
          "skill_id": "model-versioning",
          "subtype": "model_versioning",
          "type": "Concept"
        },
        "warnings": [
          "stage3_post_filter_dropped_catalog_only_locked_dims:42-\u003e4"
        ]
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Chatbots",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "Chatbots",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Concept",
          "skill_nature": "CONCEPT",
          "sub_category": "conversational_ai_systems",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "STABLE"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cChatbots\u201d in JDs typically refers to conversational agents; it\u2019s distinct from other AI concepts in the catalog."
          },
          "context_keywords": {
            "context_keywords": [
              "NLP",
              "dialogflow",
              "Rasa",
              "intent recognition",
              "entity extraction",
              "user intent",
              "conversational flow",
              "machine learning",
              "chatbot frameworks",
              "natural language understanding",
              "voice assistants",
              "customer support automation",
              "sentiment analysis",
              "API integration",
              "user experience design"
            ]
          },
          "maturity": {
            "confidence": 0.93,
            "maturity": "well_known",
            "reasoning": "Chatbots are broadly adopted and commonly appear in job postings across customer support, sales, and AI product roles; major vendors like Intercom, Zendesk, and Microsoft ship chatbot tooling."
          },
          "skill_id": "chatbots",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Systems for building chat-based assistants that understand user intent, maintain dialogue context, and generate responses. Chatbots fit here because they are interactive conversational interfaces rather than generic UI or backend services.",
            "exemplar_skills": [
              "Chatbots",
              "Conversational AI",
              "Dialog Management",
              "Intent Recognition",
              "Slot Filling",
              "Retrieval-Augmented Generation",
              "Assistant Design"
            ],
            "in_scope": "Chatbots, conversational agents, dialog management, intent handling, slot filling, multi-turn conversation design, response generation, retrieval-augmented chatbots, chatbot orchestration, ChatGPT-style assistants",
            "name": "Conversational AI Systems",
            "out_of_scope": "Traditional web UI components, mobile screen navigation, voice-only IVR systems, generic NLP model training, and backend API design not focused on conversation flow",
            "overlap_flags": [
              {
                "reason": "Chatbots often use ML/NLP libraries for model inference and response generation, but the dimension here is the conversational product/system layer.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "Chatbots can be embedded in web or app interfaces, but UI rendering is not the core skill being assessed.",
                "with_dim_id": "ui-frameworks-and-rendering",
                "with_dim_name": null,
                "with_role": "Frontend Engineer, Hybrid Mobile Developer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Chatbots",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "chatbots"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [],
          "requires": [],
          "skill_id": "chatbots",
          "suppress_on_match": []
        },
        "skill_id": "chatbots",
        "split_log": [],
        "typed": {
          "alternatives_considered": [
            "Architecture: ruled out \u2014 chatbots are not primarily a system-shape pattern like microservices.",
            "Domain: ruled out \u2014 this is a capability/concept, not a vertical industry body of knowledge.",
            "Tool: ruled out \u2014 the term is generic and does not refer to a specific user-operated application."
          ],
          "confidence": 0.78,
          "name": "Chatbots",
          "reasoning": "Chatbots are best treated as a named knowledge unit describing conversational AI systems rather than a specific software product, so they fit the Concept type.",
          "skill_id": "chatbots",
          "subtype": "conversational_ai_systems",
          "type": "Concept"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Virtual Assistants",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "Virtual Assistants",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Domain",
          "skill_nature": "CONCEPT",
          "sub_category": "virtual_assistants",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cVirtual Assistants\u201d is a distinct domain term (AI assistants/chatbots) and isn\u2019t commonly confused with another specific catalog skill."
          },
          "context_keywords": {
            "context_keywords": [
              "AI",
              "chatbots",
              "NLP",
              "automation",
              "voice recognition",
              "task management",
              "dialog systems",
              "machine learning",
              "customer support",
              "personalization",
              "integration",
              "cloud services",
              "data analysis",
              "user experience",
              "workflow optimization"
            ]
          },
          "maturity": {
            "confidence": 0.84,
            "maturity": "emerging",
            "reasoning": "Job postings increasingly mention building AI virtual assistants/copilots, and vendor ecosystems like OpenAI Assistants API and Microsoft Copilot show strong adoption, but it is not yet a universal hiring staple."
          },
          "skill_id": "virtual-assistants",
          "vendor_license": {
            "confidence": 0.9,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Systems for building conversational assistants that understand user requests, manage dialog, and trigger actions across apps or services. This skill belongs here because virtual assistants are the core product surface for assistant-driven interaction, orchestration, and response generation.",
            "exemplar_skills": [
              "Virtual Assistants",
              "Conversational AI",
              "Dialog Management",
              "Intent Recognition",
              "Slot Filling",
              "Assistant Orchestration",
              "Tool Calling",
              "Task Automation"
            ],
            "in_scope": "Virtual Assistants, conversational assistants, assistant orchestration, dialog management, intent handling, slot filling, tool calling, task completion flows, voice or text assistant UX, assistant routing, multi-turn conversations",
            "name": "Virtual Assistant Systems",
            "out_of_scope": "Chatbot marketing sites and lead-gen bots, speech recognition internals, text-to-speech engine implementation, generic NLP model training, mobile UI frameworks, which belong to speech, ML, or frontend dimensions",
            "overlap_flags": [
              {
                "reason": "Assistant implementations often rely on ML models and LLM frameworks for intent, response generation, and tool selection.",
                "with_dim_id": "ml-frameworks-and-libraries",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              },
              {
                "reason": "If the assistant is embedded in a browser or app, the presentation layer may overlap with UI implementation.",
                "with_dim_id": "ui-frameworks-and-rendering",
                "with_dim_name": null,
                "with_role": "Frontend Engineer, Hybrid Mobile Developer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Virtual Assistants",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "virtual-assistants"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "voiceover",
            "location-services",
            "service-workers",
            "ui-automator",
            "talkback",
            "apollo-client",
            "workmanager",
            "rollback-automation"
          ],
          "requires": [],
          "skill_id": "virtual-assistants",
          "suppress_on_match": []
        },
        "skill_id": "virtual-assistants",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.9,
          "name": "Virtual Assistants",
          "reasoning": "Virtual Assistants is best treated as a domain/problem-space rather than a specific product, language, or methodology, so it fits the Domain type.",
          "skill_id": "virtual-assistants",
          "subtype": "virtual_assistants",
          "type": "Domain"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    },
    {
      "aliases_in_db": [],
      "canonical": null,
      "dimensions": [
        {
          "dimension": {
            "difficulty_hint": "well_known",
            "display_name": "React Frontend Development",
            "id": 96,
            "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
            "slug": "d_init_01",
            "source": "db"
          },
          "input_skill": "Intelligent Automation",
          "llm_role": null,
          "roles_from_db": []
        }
      ],
      "input_skill": "Intelligent Automation",
      "matched_via": null,
      "new_alias_persisted": false,
      "new_alias_text": null,
      "new_skill_meta": {
        "derived": {
          "category": "Methodology",
          "skill_nature": "METHODOLOGY",
          "sub_category": "automation_methodology",
          "typical_lifespan": "EVERGREEN",
          "version_strategy": "NOT_APPLICABLE",
          "volatility": "EMERGING"
        },
        "enrichment": {
          "ambiguity": {
            "ambiguity_flag": false,
            "confused_with": [],
            "reasoning": "\u201cIntelligent Automation\u201d is a specific methodology term; unlikely to be confused with other distinct catalog skills in typical JDs."
          },
          "context_keywords": {
            "context_keywords": [
              "RPA",
              "machine learning",
              "AI",
              "process mining",
              "chatbots",
              "workflow automation",
              "NLP",
              "predictive analytics",
              "digital workforce",
              "cognitive automation",
              "data integration",
              "self-healing systems",
              "hyperautomation",
              "intelligent agents",
              "business process management"
            ]
          },
          "maturity": {
            "confidence": 0.78,
            "maturity": "emerging",
            "reasoning": "Job postings increasingly mention intelligent automation alongside RPA and AI, but it is not yet a universal hiring staple; market demand is still concentrated in enterprise transformation roles."
          },
          "skill_id": "intelligent-automation",
          "vendor_license": {
            "confidence": 0.95,
            "license": null,
            "vendor": null,
            "year_introduced": null
          },
          "versioning": {
            "current_version": null,
            "version_aliases": {},
            "versioned": false
          }
        },
        "keep_log": [],
        "locked_dimensions": [
          {
            "description": "Automation systems that combine rules, workflows, and AI-driven decisioning to execute business or operational tasks with minimal human intervention. This fits the target skill because it refers to building and orchestrating automated processes that can reason, route, and act.",
            "exemplar_skills": [
              "Intelligent Automation",
              "workflow orchestration",
              "RPA integration",
              "decision automation",
              "human-in-the-loop approvals",
              "agentic task execution"
            ],
            "in_scope": "Intelligent Automation, workflow orchestration, decision automation, RPA integration, agentic task execution, human-in-the-loop approvals, rule-based triggers, AI-assisted process routing",
            "name": "Intelligent Automation",
            "out_of_scope": "Pure robotic process automation without AI decisioning, general machine learning model training, cloud infrastructure automation, CI/CD pipeline automation",
            "overlap_flags": [
              {
                "reason": "Both involve automation workflows, but CI/CD is specifically software delivery automation rather than business or AI-driven process automation.",
                "with_dim_id": "ci-cd-pipeline-platforms",
                "with_dim_name": null,
                "with_role": "DevOps Engineer"
              },
              {
                "reason": "ML pipeline automation can resemble intelligent automation, but this dimension is focused on model delivery workflows rather than broader task automation.",
                "with_dim_id": "ci-cd-for-machine-learning",
                "with_dim_name": null,
                "with_role": "ML Engineer"
              }
            ],
            "tentative_id": "d_init_01"
          }
        ],
        "merge_log": [],
        "placed": {
          "name": "Intelligent Automation",
          "placement_confidence": 0.92,
          "primary_dimension": "d_init_01",
          "reasoning": "Deterministic JD placement: locked_dimensions has 1 dimension(s) from skill-driven dimension generation after reconciliation; primary_dimension is the first locked dim.",
          "secondary_dimensions": [],
          "skill_id": "intelligent-automation"
        },
        "relationships": {
          "child_skills": [],
          "parent_skills": [],
          "related_to": [
            "ansible",
            "jenkins",
            "rollback-automation",
            "autoscaling",
            "ui-automator",
            "robolectric",
            "aws-iam",
            "dynamic-type"
          ],
          "requires": [],
          "skill_id": "intelligent-automation",
          "suppress_on_match": []
        },
        "skill_id": "intelligent-automation",
        "split_log": [],
        "typed": {
          "alternatives_considered": [],
          "confidence": 0.88,
          "name": "Intelligent Automation",
          "reasoning": "Intelligent Automation is best treated as a way of working that combines automation with AI-driven decisioning, so it fits the Methodology category rather than a tool or platform.",
          "skill_id": "intelligent-automation",
          "subtype": "automation_methodology",
          "type": "Methodology"
        },
        "warnings": []
      },
      "source_tag": "llm",
      "was_in_llm_skills": true
    }
  ],
  "unmatched_skills": [
    "FAISS",
    "ChromaDB",
    "OpenAI",
    "Azure OpenAI",
    "Anthropic",
    "Hugging Face",
    "CI/CD",
    "CrewAI",
    "APIs",
    "LLMs",
    "RAG",
    "Embeddings",
    "MLOps",
    "Model Versioning",
    "Chatbots",
    "Virtual Assistants",
    "Intelligent Automation"
  ]
}

{
  "chosen_role": {
    "display_name": "Data Engineer",
    "id": 2,
    "rationale": "The primary skills involve data manipulation and cloud data warehousing, which are essential for a Data Engineer.",
    "role_archetype": null,
    "slug": "data-engineer",
    "source": "db"
  },
  "chosen_role_resolution": "in_db",
  "final_input_skills": [
    {
      "skill": "Python",
      "tag": "in_db"
    },
    {
      "skill": "SQL",
      "tag": "in_db"
    },
    {
      "skill": "Snowflake",
      "tag": "in_db"
    },
    {
      "skill": "Pinecone",
      "tag": "in_db"
    },
    {
      "skill": "FAISS",
      "tag": "new"
    },
    {
      "skill": "ChromaDB",
      "tag": "new"
    },
    {
      "skill": "LangChain",
      "tag": "in_db"
    },
    {
      "skill": "LlamaIndex",
      "tag": "in_db"
    },
    {
      "skill": "OpenAI",
      "tag": "new"
    },
    {
      "skill": "Azure OpenAI",
      "tag": "new"
    },
    {
      "skill": "Anthropic",
      "tag": "new"
    },
    {
      "skill": "Hugging Face",
      "tag": "new"
    },
    {
      "skill": "PostgreSQL",
      "tag": "in_db"
    },
    {
      "skill": "MySQL",
      "tag": "in_db"
    },
    {
      "skill": "Git",
      "tag": "in_db"
    },
    {
      "skill": "CI/CD",
      "tag": "new"
    },
    {
      "skill": "Docker",
      "tag": "in_db"
    },
    {
      "skill": "MLflow",
      "tag": "in_db"
    },
    {
      "skill": "AWS",
      "tag": "in_db"
    },
    {
      "skill": "Azure",
      "tag": "in_db"
    },
    {
      "skill": "GCP",
      "tag": "in_db"
    },
    {
      "skill": "Power BI",
      "tag": "in_db"
    },
    {
      "skill": "Tableau",
      "tag": "in_db"
    },
    {
      "skill": "CrewAI",
      "tag": "new"
    },
    {
      "skill": "APIs",
      "tag": "new"
    },
    {
      "skill": "LLMs",
      "tag": "new"
    },
    {
      "skill": "RAG",
      "tag": "new"
    },
    {
      "skill": "Embeddings",
      "tag": "new"
    },
    {
      "skill": "MLOps",
      "tag": "new"
    },
    {
      "skill": "Model Versioning",
      "tag": "new"
    },
    {
      "skill": "Chatbots",
      "tag": "new"
    },
    {
      "skill": "Virtual Assistants",
      "tag": "new"
    },
    {
      "skill": "Intelligent Automation",
      "tag": "new"
    }
  ],
  "persistence": {
    "items": [
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Programming Languages",
          "id": 1,
          "rationale": "Core server-side languages used to implement backend business logic, integrations, and service internals. This is the primary coding surface for the role across application layers.",
          "slug": "programming-languages",
          "source": "db"
        },
        "dimension_id": 1,
        "input_skill": "Python",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 5,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Programming Languages and Scripting",
          "id": 59,
          "rationale": "Languages used to write security automation, analysis scripts, detection logic, and remediation helpers. This is the primary implementation surface for a cybersecurity engineer across tooling and response workflows.",
          "slug": "programming-languages-and-scripting",
          "source": "db"
        },
        "dimension_id": 59,
        "input_skill": "Python",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 5,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Programming Languages for Data Work",
          "id": 21,
          "rationale": "Languages used to implement data pipelines, transformations, and operational glue. This is the primary coding surface for building ingestion, enrichment, and automation logic in data engineering.",
          "slug": "programming-languages-for-data-work",
          "source": "db"
        },
        "dimension_id": 21,
        "input_skill": "Python",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 5,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Programming Languages for ML Systems",
          "id": 39,
          "rationale": "Languages used to build training code, inference services, evaluation jobs, and ML glue code. This is the primary implementation surface for ML engineers across experimentation and productionization.",
          "slug": "programming-languages-for-ml-systems",
          "source": "db"
        },
        "dimension_id": 39,
        "input_skill": "Python",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 5,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Programming Languages for XR",
          "id": 97,
          "rationale": "Primary implementation languages used to build immersive client features, interaction logic, and device-specific runtime behavior. This is the core coding surface for AR/VR experiences.",
          "slug": "programming-languages-for-xr",
          "source": "db"
        },
        "dimension_id": 97,
        "input_skill": "Python",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "AR/VR Engineer",
            "id": 8,
            "rationale": null,
            "role_archetype": null,
            "slug": "ar-vr-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 5,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Programming Languages for Data Work",
          "id": 21,
          "rationale": "Languages used to implement data pipelines, transformations, and operational glue. This is the primary coding surface for building ingestion, enrichment, and automation logic in data engineering.",
          "slug": "programming-languages-for-data-work",
          "source": "db"
        },
        "dimension_id": 21,
        "input_skill": "SQL",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 101,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Data Warehouses",
          "id": 22,
          "rationale": "Managed analytical storage and compute platforms used for curated datasets, reporting, and downstream analytics. These systems are central to data modeling, performance tuning, and cost-aware query design.",
          "slug": "cloud-data-warehouses",
          "source": "db"
        },
        "dimension_id": 22,
        "input_skill": "Snowflake",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 105,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "LLM Operations and Orchestration",
          "id": 49,
          "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
          "slug": "llm-operations-and-orchestration",
          "source": "db"
        },
        "dimension_id": 49,
        "input_skill": "Pinecone",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 242,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "LLM Operations and Orchestration",
          "id": 49,
          "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
          "slug": "llm-operations-and-orchestration",
          "source": "db"
        },
        "dimension_id": 49,
        "input_skill": "LangChain",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 240,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "LLM Operations and Orchestration",
          "id": 49,
          "rationale": "Operational stack for building, serving, evaluating, and orchestrating LLM-based systems. This includes vector retrieval, prompt workflows, LLM serving, and observability for generative applications.",
          "slug": "llm-operations-and-orchestration",
          "source": "db"
        },
        "dimension_id": 49,
        "input_skill": "LlamaIndex",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 244,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Relational Database Design",
          "id": 4,
          "rationale": "Modeling and operating relational persistence for backend services. Includes schema design, normalization, indexing, transactions, and query tuning for operational data stores.",
          "slug": "relational-database-design",
          "source": "db"
        },
        "dimension_id": 4,
        "input_skill": "PostgreSQL",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 16,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Relational Database Design",
          "id": 4,
          "rationale": "Modeling and operating relational persistence for backend services. Includes schema design, normalization, indexing, transactions, and query tuning for operational data stores.",
          "slug": "relational-database-design",
          "source": "db"
        },
        "dimension_id": 4,
        "input_skill": "MySQL",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 17,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Git",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1002,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Containerization and Image Builds",
          "id": 152,
          "rationale": "Container image creation, tagging, hardening, and registry workflows used to package services for deployment. This is coherent because DevOps often owns the build-to-image path that feeds runtime environments.",
          "slug": "containerization-and-image-builds",
          "source": "db"
        },
        "dimension_id": 152,
        "input_skill": "Docker",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "DevOps Engineer",
            "id": 10,
            "rationale": null,
            "role_archetype": null,
            "slug": "devops-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 61,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Deployment and Runtime Configuration",
          "id": 13,
          "rationale": "Configuration and release artifacts that control how backend services run in environments. Includes environment variables, manifests, feature flags, and release-safe configuration management.",
          "slug": "deployment-and-runtime-configuration",
          "source": "db"
        },
        "dimension_id": 13,
        "input_skill": "Docker",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 61,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "MLOps Platforms and Lifecycle",
          "id": 43,
          "rationale": "End-to-end managed platforms used to train, deploy, register, and govern models across their lifecycle. This is the operational control plane for production ML workflows.",
          "slug": "mlops-platforms-and-lifecycle",
          "source": "db"
        },
        "dimension_id": 43,
        "input_skill": "MLflow",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 214,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Platforms",
          "id": 20,
          "rationale": "Proficiency in major cloud service provider platforms and their core services.",
          "slug": "cloud-platforms",
          "source": "db"
        },
        "dimension_id": 20,
        "input_skill": "AWS",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          },
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          },
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          },
          {
            "display_name": "DevOps Engineer",
            "id": 10,
            "rationale": null,
            "role_archetype": null,
            "slug": "devops-engineer",
            "source": "db"
          },
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 187,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Provider Platforms",
          "id": 131,
          "rationale": "Major cloud platforms and their core service ecosystems used to design target-state architectures, choose deployment boundaries, and evaluate managed capabilities. This is the primary substrate for cloud architecture decisions.",
          "slug": "cloud-provider-platforms",
          "source": "db"
        },
        "dimension_id": 131,
        "input_skill": "AWS",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Cloud Architect",
            "id": 9,
            "rationale": null,
            "role_archetype": null,
            "slug": "cloud-architect",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 187,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Security Posture Tools",
          "id": 64,
          "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
          "slug": "cloud-security-posture-tools",
          "source": "db"
        },
        "dimension_id": 64,
        "input_skill": "AWS",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 187,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Platforms",
          "id": 20,
          "rationale": "Proficiency in major cloud service provider platforms and their core services.",
          "slug": "cloud-platforms",
          "source": "db"
        },
        "dimension_id": 20,
        "input_skill": "Azure",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          },
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          },
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          },
          {
            "display_name": "DevOps Engineer",
            "id": 10,
            "rationale": null,
            "role_archetype": null,
            "slug": "devops-engineer",
            "source": "db"
          },
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 188,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Provider Platforms",
          "id": 131,
          "rationale": "Major cloud platforms and their core service ecosystems used to design target-state architectures, choose deployment boundaries, and evaluate managed capabilities. This is the primary substrate for cloud architecture decisions.",
          "slug": "cloud-provider-platforms",
          "source": "db"
        },
        "dimension_id": 131,
        "input_skill": "Azure",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Cloud Architect",
            "id": 9,
            "rationale": null,
            "role_archetype": null,
            "slug": "cloud-architect",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 188,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Security Posture Tools",
          "id": 64,
          "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
          "slug": "cloud-security-posture-tools",
          "source": "db"
        },
        "dimension_id": 64,
        "input_skill": "Azure",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 188,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Platforms",
          "id": 20,
          "rationale": "Proficiency in major cloud service provider platforms and their core services.",
          "slug": "cloud-platforms",
          "source": "db"
        },
        "dimension_id": 20,
        "input_skill": "GCP",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          },
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          },
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          },
          {
            "display_name": "DevOps Engineer",
            "id": 10,
            "rationale": null,
            "role_archetype": null,
            "slug": "devops-engineer",
            "source": "db"
          },
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 186,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Security Posture Tools",
          "id": 64,
          "rationale": "Cloud-native security platforms used to assess misconfiguration, workload exposure, and cloud control coverage. This dimension includes the major CNAPP/CSPM/CWPP vendors and cloud security services the role reviews and tunes.",
          "slug": "cloud-security-posture-tools",
          "source": "db"
        },
        "dimension_id": 64,
        "input_skill": "GCP",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 186,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "BI and Visualization Tools",
          "id": 31,
          "rationale": "Tools used to expose curated data to analysts and business users through dashboards, reports, and semantic exploration. Data engineers support these tools by shaping reliable datasets and performant models.",
          "slug": "bi-and-visualization-tools",
          "source": "db"
        },
        "dimension_id": 31,
        "input_skill": "Power BI",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 151,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "BI and Visualization Tools",
          "id": 31,
          "rationale": "Tools used to expose curated data to analysts and business users through dashboards, reports, and semantic exploration. Data engineers support these tools by shaping reliable datasets and performant models.",
          "slug": "bi-and-visualization-tools",
          "source": "db"
        },
        "dimension_id": 31,
        "input_skill": "Tableau",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 150,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "FAISS",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1184,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "ML Frameworks and Libraries",
          "id": 40,
          "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
          "slug": "ml-frameworks-and-libraries",
          "source": "db"
        },
        "dimension_id": 40,
        "input_skill": "FAISS",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1184,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "ChromaDB",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1185,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "OpenAI",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1186,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Cloud Platforms",
          "id": 20,
          "rationale": "Proficiency in major cloud service provider platforms and their core services.",
          "slug": "cloud-platforms",
          "source": "db"
        },
        "dimension_id": 20,
        "input_skill": "Azure OpenAI",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Backend Engineer",
            "id": 1,
            "rationale": null,
            "role_archetype": "A Backend Engineer designs, builds, and maintains the server-side logic and data handling that power applications and services. They focus on implementing reliable business functionality, integrating with other systems, and ensuring the backend is scalable, maintainable, and observable.",
            "slug": "backend-engineer",
            "source": "db"
          },
          {
            "display_name": "Cybersecurity Engineer",
            "id": 5,
            "rationale": null,
            "role_archetype": null,
            "slug": "cybersecurity-engineer",
            "source": "db"
          },
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          },
          {
            "display_name": "DevOps Engineer",
            "id": 10,
            "rationale": null,
            "role_archetype": null,
            "slug": "devops-engineer",
            "source": "db"
          },
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1187,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Azure OpenAI",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1187,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Anthropic",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1188,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "ML Frameworks and Libraries",
          "id": 40,
          "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
          "slug": "ml-frameworks-and-libraries",
          "source": "db"
        },
        "dimension_id": 40,
        "input_skill": "Hugging Face",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1189,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Hugging Face",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1189,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "CI/CD Pipeline Platforms",
          "id": 150,
          "rationale": "Systems used to define, run, and maintain automated build and deployment workflows. This cluster is coherent because the role owns delivery automation end to end, including pipeline reliability and promotion logic.",
          "slug": "ci-cd-pipeline-platforms",
          "source": "db"
        },
        "dimension_id": 150,
        "input_skill": "CI/CD",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "DevOps Engineer",
            "id": 10,
            "rationale": null,
            "role_archetype": null,
            "slug": "devops-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1190,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "CI/CD for Machine Learning",
          "id": 56,
          "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
          "slug": "ci-cd-for-machine-learning",
          "source": "db"
        },
        "dimension_id": 56,
        "input_skill": "CI/CD",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1190,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "CrewAI",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1191,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "APIs",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1192,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "LLMs",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1193,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "RAG",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1194,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "ML Frameworks and Libraries",
          "id": 40,
          "rationale": "Core libraries used to define models, train them, run inference, and evaluate predictive performance. These frameworks shape how ML engineers express model architectures and training loops.",
          "slug": "ml-frameworks-and-libraries",
          "source": "db"
        },
        "dimension_id": 40,
        "input_skill": "Embeddings",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1195,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Embeddings",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1195,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Systems Programming",
          "id": 166,
          "rationale": "Systems programming covers low-level software development where performance, memory safety, and direct control over resources matter. Rust fits here because it is commonly used for OS-adjacent services, infrastructure components, and other performance-sensitive systems code.",
          "slug": "d_init_02",
          "source": "db"
        },
        "dimension_id": 166,
        "input_skill": "Embeddings",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1195,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "CI/CD for Machine Learning",
          "id": 56,
          "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
          "slug": "ci-cd-for-machine-learning",
          "source": "db"
        },
        "dimension_id": 56,
        "input_skill": "MLOps",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1196,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Deployment Rollouts and Release Control",
          "id": 51,
          "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
          "slug": "deployment-rollouts-and-release-control",
          "source": "db"
        },
        "dimension_id": 51,
        "input_skill": "MLOps",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1196,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Data Lineage and Metadata",
          "id": 28,
          "rationale": "Cataloging, documenting, and tracing how data moves and changes across systems. This dimension supports impact analysis, governance, discoverability, and operational understanding of datasets.",
          "slug": "data-lineage-and-metadata",
          "source": "db"
        },
        "dimension_id": 28,
        "input_skill": "MLOps",
        "llm_role": null,
        "matched_chosen_role": true,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension saved",
        "role_dimension_saved": true,
        "roles_from_db": [
          {
            "display_name": "Data Engineer",
            "id": 2,
            "rationale": null,
            "role_archetype": null,
            "slug": "data-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1196,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "Deployment Rollouts and Release Control",
          "id": 51,
          "rationale": "Practices for safely promoting models through environments and managing rollback when production behavior changes. This dimension covers release gating, version pinning, and rollout strategies specific to ML systems.",
          "slug": "deployment-rollouts-and-release-control",
          "source": "db"
        },
        "dimension_id": 51,
        "input_skill": "Model Versioning",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1197,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "CI/CD for Machine Learning",
          "id": 56,
          "rationale": "Tools and platforms for automating ML model integration, testing, and deployment pipelines.",
          "slug": "ci-cd-for-machine-learning",
          "source": "db"
        },
        "dimension_id": 56,
        "input_skill": "Model Versioning",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [
          {
            "display_name": "ML Engineer",
            "id": 3,
            "rationale": null,
            "role_archetype": null,
            "slug": "ml-engineer",
            "source": "db"
          }
        ],
        "skill_dimension_saved": true,
        "skill_id": 1197,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Chatbots",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1198,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Virtual Assistants",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1199,
        "skill_tag": "in_db",
        "skipped_reason": null
      },
      {
        "chosen_role_id": 2,
        "dimension": {
          "difficulty_hint": "well_known",
          "display_name": "React Frontend Development",
          "id": 96,
          "rationale": "Building interactive web user interfaces with React.js, including component composition, state management, hooks, and rendering patterns. React.js belongs here because it is a core library for client-side UI development in modern web applications.",
          "slug": "d_init_01",
          "source": "db"
        },
        "dimension_id": 96,
        "input_skill": "Intelligent Automation",
        "llm_role": null,
        "matched_chosen_role": false,
        "outcome_line": "New skill saved \u00b7 Existing dimension (library) \u00b7 Role\u2194dimension skipped (dimension not under chosen role)",
        "role_dimension_saved": false,
        "roles_from_db": [],
        "skill_dimension_saved": true,
        "skill_id": 1200,
        "skill_tag": "in_db",
        "skipped_reason": null
      }
    ],
    "new_skills_created": 17,
    "role_dimension_saved": 0,
    "skill_dimension_saved": 26,
    "skipped": 0
  },
  "planner_output": null,
  "run_id": "22f3e3d9-31b8-42af-98a2-39fb214dd4a2"
}