Mle Workflow

This is the skill you want when a notebook needs to become a production ML system. It walks you through data contracts, reproducible training, evaluation gates, deployment, monitoring, and rollback without forcing one architecture onto every problem. The scope calibration is refreshingly honest: it tells you not to add heavyweight MLOps machinery when a baseline, eval script, and rollback note would be enough. What makes it useful is the table mapping standard SWE workflows to ML contexts, so you reuse existing surfaces like TDD, code review, database migrations, and canary rollout instead of building a parallel stack. Good for ranking systems, classifiers, recommenders, forecasting pipelines, and anything that needs to survive contact with production data drift.

skills.sh installs: 5.6K
GitHub stars: 300.3K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add affaan-m/ecc --skill mle-workflow --agent claude-code

Installs into .claude/skills of the current project.

Files

Machine Learning Engineering Workflow

Use this skill to turn model work into a production ML system with clear data contracts, repeatable training, measurable quality gates, deployable artifacts, and operational monitoring.

When to Activate

Planning or reviewing a production ML feature, model refresh, ranking system, recommender, classifier, embedding workflow, or forecasting pipeline
Converting notebook code into a reusable training, evaluation, batch inference, or online inference pipeline
Designing model promotion criteria, offline/online evals, experiment tracking, or rollback paths
Debugging failures caused by data drift, label leakage, stale features, artifact mismatch, or inconsistent training and serving logic
Adding model monitoring, canary rollout, shadow traffic, or post-deploy quality checks

Scope Calibration

Use only the lanes that fit the system in front of you. This skill is useful for ranking, search, recommendations, classifiers, forecasting, embeddings, LLM workflows, anomaly detection, and batch analytics, but it should not force one architecture onto all of them.

SKILL.md on GitHub