This skill provides comprehensive guidance for architecting and automating the full machine learning lifecycle, enabling developers to create robust MLOps workflows. It covers every stage from data ingestion and feature engineering to model training, rigorous validation, and scalable deployment patterns. By using this skill, teams can implement reproducible training workflows, manage experiment tracking, and ensure smooth transitions from research to production environments using industry-standard tools like Airflow, Kubeflow, and MLflow.
主な機能
01Model validation frameworks with A/B testing support
02End-to-end DAG orchestration patterns for ML workflows
03Automated data validation and feature engineering pipelines
04Experiment tracking and model registry integration
050 GitHub stars
06Production deployment strategies including canary and blue-green
