What problem does it solve? Turning notebook experiments into production ML systems often fails due to data leakage, irreproducible training, missing promotion gates, and absent monitoring. This Skill provides a structured workflow that converts model work into reviewable, deployable, and operable systems. ## Core Features & Use Cases - Data and Prediction Contracts: Define entity grain, label timing, feature freshness, split policy, and leakage checks before writing model code. - Reproducible Training and Promotion Gates: Pin configs, seeds, and dataset versions, then enforce automated metric gates comparing against baselines and production models. - Deployment and Operations: Package versioned artifacts with preprocessing, validate serving contracts, and monitor drift, latency, and delayed labels with rollback plans. - Use Case: When converting a churn-prediction notebook into a production service, use this Skill to produce a data contract, baseline scorer, promotion gate tests, a versioned artifact, and a canary rollout plan with rollback triggers. ## Quick Start Use the mle-workflow skill to plan a production pipeline for my classification model, including the data contract, evaluation gates, and rollback strategy.
