{
  "id": "skill-area:embedding-optimization",
  "_kind": "SkillArea",
  "_file": "domain/skill-areas/skill-areas-knowledge-fabric.yaml",
  "_cluster": "domain",
  "attributes": {
    "displayName": "Embedding Optimization",
    "description": "Choosing, tuning, and evaluating embedding models for knowledge\nretrieval — model selection (OpenAI, Cohere, open-source E5/BGE/GTE),\ndimensionality trade-offs (higher dimensions = better quality but\nmore storage/compute), fine-tuning embeddings on domain-specific\ndata for improved retrieval relevance, quantization for memory\nefficiency, and benchmarking with MTEB (Massive Text Embedding\nBenchmark) and domain-specific evaluation sets.\n",
    "domains": [
      "specialization:ai-agents-conversational",
      "domain:data-science"
    ],
    "expertiseLevels": [
      "intermediate",
      "expert"
    ]
  },
  "outgoingEdges": [
    {
      "from": "skill-area:embedding-optimization",
      "to": "specialization:ai-agents-conversational",
      "kind": "applies_to",
      "attributes": {
        "confidence": "primary"
      }
    },
    {
      "from": "skill-area:embedding-optimization",
      "to": "domain:data-science",
      "kind": "applies_to",
      "attributes": {
        "confidence": "secondary"
      }
    },
    {
      "from": "skill-area:embedding-optimization",
      "to": "tool:pinecone",
      "kind": "uses_tool"
    },
    {
      "from": "skill-area:embedding-optimization",
      "to": "tool:weaviate",
      "kind": "uses_tool"
    },
    {
      "from": "skill-area:embedding-optimization",
      "to": "tool:chromadb",
      "kind": "uses_tool"
    },
    {
      "from": "skill-area:embedding-optimization",
      "to": "skill-area:retrieval-evaluation",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "required"
      }
    }
  ],
  "incomingEdges": [
    {
      "from": "skill-area:document-indexing",
      "to": "skill-area:embedding-optimization",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "required"
      }
    },
    {
      "from": "stack-profile:agentic-rag",
      "to": "skill-area:embedding-optimization",
      "kind": "requires_skill_area"
    },
    {
      "from": "stack-profile:rag-stack",
      "to": "skill-area:embedding-optimization",
      "kind": "requires_skill_area"
    },
    {
      "from": "tool:token-optimizer",
      "to": "skill-area:embedding-optimization",
      "kind": "tool_used_by",
      "attributes": {}
    },
    {
      "from": "tool:imptokens",
      "to": "skill-area:embedding-optimization",
      "kind": "tool_used_by",
      "attributes": {}
    }
  ]
}