Guide MLOps practices for model lifecycle, CI/CD, and production reliability.
npx skills add https://github.com/MayaDispeler/TheOrqestra --skill mlops-patterns
Please help me install this Agent Skill. Skill: mlops-patterns Source: https://github.com/MayaDispeler/TheOrqestra/tree/main/skills/mlops-patterns Command: npx skills add https://github.com/MayaDispeler/TheOrqestra --skill mlops-patterns
This Skill provides a comprehensive reference for MLOps best practices, addressing model lifecycle management, CI/CD for ML, feature stores, training pipelines, serving infrastructure, drift detection, retraining triggers, and production reliability for ML systems.
Access the MLOps Patterns Expert Reference and implement best practices for your ML projects.
High-intent search queries and answers about installing and using this skill.
MLOps manages the model lifecycle by applying CI/CD, training pipelines, and drift detection to ML systems. It provides standards for reproducibility, monitoring, and automation to ensure production reliability from training to serving.
Setting up CI/CD for ML training pipelines requires automating reproducibility, data quality checks, and separation of concerns. This Skill outlines decision rules for pipeline engineering and deployment triggers to maintain production reliability.
The best way to detect data drift involves monitoring production inputs against baseline distributions to trigger automatic retraining. This Skill details decision rules for drift monitoring and setting retraining triggers to prevent model degradation.
A feature store is needed for model serving when you require consistent feature engineering and data access across training and production. This Skill provides guidelines for integrating feature stores within training pipelines and serving infrastructure.
Common MLOps mistakes include poor separation of concerns, lacking reproducibility, and ignoring drift detection. This reference identifies these pitfalls and provides solutions to avoid ineffective workflows and ensure reliable deployment.
Yes, you can use these MLOps patterns for experiment tracking and rollback. The guidelines cover deployment strategies, monitoring, and rollback decision rules to maintain production reliability throughout the model lifecycle.
AradotsoProvision AWS S3 buckets and EC2 instances with Terraform for data engineering.
Tnemo65Automate end-to-end ML deployment, monitoring, and lifecycle management in production.
nahisahoDevelop end-to-end ML models with training, evaluation, deployment, and MLOps.
saeed-vayghanAutomate infrastructure, CI/CD pipelines, and monitoring with Terraform, Ansible, Docker, and Kubernetes.
