slug: mlflow
provider: MLflow
generated_by: planning/capability-mapping/scripts/classify_capabilities.py
model: claude-opus-5
frame:
- Software & Technology
min_confidence: 0.7
capability_model:
  source: https://github.com/vincentmakes/turbo-ea-capabilities
  license: CC-BY-4.0
  attribution: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities,
    CC BY 4.0
  notice: NOTICE
edge_count: 6
edges:
- tag: Model Versions
  spec_file: mlflow-model-versions-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.82
  evidence: POST /api/2.0/mlflow/model-versions/create  createModelVersion  Create model version
  reason: Model registry versioning of ML models is core AI/ML model lifecycle management.
- tag: Registered Models
  spec_file: mlflow-registered-models-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.82
  evidence: createRegisteredModel  Create registered model ... "the model registry"
  reason: Model registry operations govern ML model artefacts through their lifecycle — AI/ML management.
- tag: Artifacts
  spec_file: mlflow-artifacts-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.75
  evidence: 'GET /api/2.0/mlflow/artifacts/list ''List artifacts for a run''; vendor: ''open-source platform
    for the end-to-end ML and GenAI lifecycle: tracking, model registry'''
  reason: Artifacts here are the outputs of an MLflow experiment run (models, plots, files) within ML
    experiment tracking, placing it in AI/ML model lifecycle and MLOps. Slight discount because the two
    operations are generic storage listing and presigned-upload plumbing rather than model management
    proper.
- tag: Experiments
  spec_file: mlflow-experiments-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.75
  evidence: POST /api/2.0/mlflow/experiments/create  createExperiment  Create experiment — "MLflow is
    an open-source platform for the end-to-end ML and GenAI lifecycle"
  reason: Experiments here are ML tracking experiments grouping training runs, metrics and params — squarely
    AI/ML model lifecycle (MLOps) tooling, not product A/B experimentation.
- tag: Runs
  spec_file: mlflow-runs-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.75
  evidence: POST /api/2.0/mlflow/runs/create  createRun; POST /api/2.0/mlflow/runs/search  searchRuns
  reason: Runs are ML training/evaluation executions tracked by MLflow; lifecycle operations on them realise
    AI/ML model lifecycle management.
- tag: Metrics
  spec_file: mlflow-metrics-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.72
  evidence: POST /api/2.0/mlflow/runs/log-metric  logMetric  Log a metric for a run; logBatch  Log a batch
    of metrics, params, and tags
  reason: Logging metrics and parameters for ML training runs is part of ML model lifecycle / MLOps tracking,
    not business KPI reporting.