Build and automate end-to-end MLOps pipelines from data preparation and model training to production deployment.
This skill provides Claude with comprehensive guidance for architecting and implementing production-grade Machine Learning pipelines. It covers the entire lifecycle including data ingestion, validation, feature engineering, model training, and automated deployment strategies like canary or blue-green releases. By integrating best practices for modularity and idempotency, it helps developers build reproducible workflows using popular orchestration tools like Airflow, Dagster, and Kubeflow, ensuring robust model performance through continuous monitoring and experiment tracking.
主な機能
01Robust validation frameworks for performance regression detection
020 GitHub stars
03Automated deployment patterns including canary and blue-green rollouts
04Standardized model training and hyperparameter management
05End-to-end DAG orchestration for complex ML workflows
06Automated data validation and feature engineering pipelines
ユースケース
01Building reproducible ML training pipelines from scratch
02Setting up experiment tracking and model registries for production systems
