{
  "id": "skill-area:model-serving",
  "_kind": "SkillArea",
  "_file": "domain/skill-areas/skill-areas-ecosystem-expansion.yaml",
  "_cluster": "domain",
  "attributes": {
    "displayName": "Model Serving",
    "description": "Deploying and operating machine learning models in production — inference\nservers, batching strategies, hardware-aware optimization, autoscaling,\nand low-latency endpoint design for ML and LLM workloads.\n",
    "domains": [
      "domain:ml-ops"
    ],
    "expertiseLevels": [
      "intermediate",
      "expert"
    ]
  },
  "outgoingEdges": [
    {
      "from": "skill-area:model-serving",
      "to": "domain:ml-ops",
      "kind": "applies_to",
      "attributes": {
        "confidence": "primary"
      }
    },
    {
      "from": "skill-area:model-serving",
      "to": "skill-area:llm-infrastructure",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "recommended"
      }
    },
    {
      "from": "skill-area:model-serving",
      "to": "skill-area:model-serving-deployment",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "recommended"
      }
    },
    {
      "from": "skill-area:model-serving",
      "to": "skill-area:model-serving-operations",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "recommended"
      }
    },
    {
      "from": "skill-area:model-serving",
      "to": "skill-area:model-optimisation",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "recommended"
      }
    }
  ],
  "incomingEdges": [
    {
      "from": "skill-area:model-evaluation",
      "to": "skill-area:model-serving",
      "kind": "prerequisite_for_learning",
      "attributes": {
        "strength": "recommended"
      }
    },
    {
      "from": "specialization:recommendation-infrastructure",
      "to": "skill-area:model-serving",
      "kind": "contains"
    },
    {
      "from": "tool:mlflow",
      "to": "skill-area:model-serving",
      "kind": "used_for",
      "attributes": {}
    },
    {
      "from": "tool:bentoml",
      "to": "skill-area:model-serving",
      "kind": "used_for",
      "attributes": {}
    },
    {
      "from": "tool:vllm",
      "to": "skill-area:model-serving",
      "kind": "tool_used_by",
      "attributes": {}
    },
    {
      "from": "tool:vllm",
      "to": "skill-area:model-serving",
      "kind": "used_for"
    },
    {
      "from": "tool:tensorrt",
      "to": "skill-area:model-serving",
      "kind": "tool_used_by",
      "attributes": {}
    },
    {
      "from": "tool:tensorrt",
      "to": "skill-area:model-serving",
      "kind": "used_for"
    },
    {
      "from": "tool:triton-inference",
      "to": "skill-area:model-serving",
      "kind": "tool_used_by",
      "attributes": {}
    },
    {
      "from": "tool:triton-inference",
      "to": "skill-area:model-serving",
      "kind": "used_for"
    },
    {
      "from": "tool:onnx-runtime",
      "to": "skill-area:model-serving",
      "kind": "tool_used_by",
      "attributes": {}
    },
    {
      "from": "tool:onnx-runtime",
      "to": "skill-area:model-serving",
      "kind": "used_for"
    },
    {
      "from": "responsibility:inference-latency-sla",
      "to": "skill-area:model-serving",
      "kind": "requires_expertise",
      "attributes": {}
    },
    {
      "from": "role:machine-learning-ops-engineer",
      "to": "skill-area:model-serving",
      "kind": "requires_expertise",
      "attributes": {}
    }
  ]
}