--- title: "MLOps Fundamentals on STACKIT" description: "Essential MLOps practices on STACKIT: CI/CD for ML models, experiment tracking and model versioning, deployment strategies, and ML observability." scfAsset: managed: false category: "guide" maintainers: - user: "can.celik1" external: true tags: ["STACKIT University", "Learning", "MLOps", "AI"] source_url: "https://framework.stackit.cloud/data-and-ai/assetcontainer/stackit/mlops-fundamentals-on-stackit/" source_file: "docs/data-and-ai/assetcontainer/stackit/mlops-fundamentals-on-stackit.mdx" --- ## Course Overview MLOps bridges the gap between model development and production deployment by applying DevOps principles to machine learning workflows. This course teaches how to build robust ML pipelines, manage model lifecycles, and keep AI applications reliable in production — all on STACKIT's sovereign cloud infrastructure. Across five modules it covers MLOps principles and the maturity model, CI/CD pipelines for ML models, experiment tracking and model versioning with MLflow, model registries and deployment strategies (blue-green, canary, A/B), and monitoring, drift detection, and observability for AI systems in production. ### What You'll Learn - Understand MLOps principles and how they bridge development and production - Design and implement CI/CD pipelines for machine learning models - Set up experiment tracking and model versioning with MLflow on STACKIT - Configure model registries and deployment strategies including blue-green, canary, and A/B patterns - Implement monitoring, drift detection, and observability for AI applications in production ### Modules - Introduction to MLOps and STACKIT Infrastructure - CI/CD Pipelines for ML Models - Experiment Tracking and Model Versioning - Model Registry and Deployment Strategies - Monitoring and Observability