mle-workflow
Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.
npx skills add mturac/everything-openai-codex --skill mle-workflow --agent codex
Same command for any agent — swap --agent for claude-code, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Turns model work into a production ML system with clear data contracts, repeatable training, measurable quality gates, deployable artifacts, and operational monitoring.
How it works
Provides activation prompts for planning or reviewing production ML features, converting notebook code into reusable training/evaluation/batch or online inference pipelines, designing promotion criteria, offline/online evaluations, experiment tracking, and rollback paths. It emphasizes licensing safe, explicit data contracts and engineering-focused gates, and suggests integrating monitoring, canary/shadow deployments, and post-deploy quality checks. It also outlines a workflow structure across data contracts, evaluation, serving, and rollback artifacts, and includes guidance on iteration compactness and decision-driven processes (Decision Brain, Metric/Mistake Economics, Error Analysis Loop, Observation Ledger).
When to use it
Use when planning or reviewing production ML features, converting notebook code to reusable pipelines, designing promotion criteria and rollback paths, debugging data-related failures, or adding model monitoring and canary rollouts.
What it can touch
It references integration with artifacts, evaluation harnesses, deployment patterns, and monitoring dashboards, and mentions lanes like data contracts, eval gates, serving, and rollback, but does not enumerate specific touchable tools beyond general workflow concepts.
Caveats
No explicit license caveat beyond general best-practice tone; emphasizes when to apply lanes and notes about not forcing architectures. Contains guidance to replace project-specific details with appropriate equivalents and explicit assumptions if lacking labels, traffic, or monitoring ownership.
# 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. - Do not assume every model has supervised labels, online serving, a feature store
- When to Activate
- Scope Calibration
- Related Skills
- Reuse the SWE Surface
- Ten MLE Task Simulations
- Iteration Compact
- Decision Brain
- Metric and Mistake Economics
- Data and Feature Hypotheses
- Error Analysis Loop
- Observation Ledger
- Core Workflow
- 1. Define the Prediction Contract
- 2. Lock the Data Contract
What does the mle-workflow skill do?
Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.
How do I install it?
Run `npx skills add mturac/everything-openai-codex --skill mle-workflow --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From mturac/everything-openai-codex, a repository with 84 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.