{
  "id": "workflow:llm-eval-pipeline",
  "_kind": "Workflow",
  "_file": "workflows/workflows/workflows-ml-data.yaml",
  "_cluster": "workflows",
  "attributes": {
    "displayName": "LLM Evaluation Pipeline",
    "workflowKind": "data",
    "triggerType": "scheduled",
    "typicalCadence": "weekly",
    "complexity": "cross-team",
    "description": "Operates the continuous evaluation pipeline for LLM-powered features —\nmaintaining eval datasets, running benchmark suites across model versions,\ntracking quality trends, and producing comparative reports. Excludes model fine-tuning.\n"
  },
  "outgoingEdges": [
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "role:ml-engineer",
      "kind": "involves_role",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "role:ai-champion",
      "kind": "involves_role",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "role:data-scientist",
      "kind": "involves_role",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "skill-area:eval-driven-development",
      "kind": "requires_skill_area",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "skill-area:prompt-engineering",
      "kind": "requires_skill_area",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "skill-area:python-data-pipelines",
      "kind": "requires_skill_area",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "domain:ml-ops",
      "kind": "applies_to_domain",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "domain:data-science",
      "kind": "applies_to_domain",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "responsibility:ai-tooling-evaluation",
      "kind": "triggers_responsibility",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "responsibility:data-quality-monitoring",
      "kind": "triggers_responsibility",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "org-unit:ml-team",
      "kind": "performed_by_org_unit",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "org-unit:ai-enablement",
      "kind": "performed_by_org_unit",
      "attributes": {}
    },
    {
      "from": "workflow:llm-eval-pipeline",
      "to": "org-unit:research-engineering",
      "kind": "performed_by_org_unit",
      "attributes": {}
    }
  ],
  "incomingEdges": []
}