MLOps & Model Deployment
Take models from notebook to production — CI/CD, monitoring, and infrastructure. Learn how to deploy, monitor, and maintain ML models in production.
Level: Advanced · Category: MLOps · Estimated time: 5 hours
Prerequisites
- PyTorch Mastery
Lessons
- Experiment Tracking — MLflow, Weights & Biases, and systematic experiment management.
- Model Serving with FastAPI — Building REST APIs for model inference with FastAPI.
- Containerization with Docker — Dockerizing ML models, multi-stage builds, and Docker Compose.
- Data & Model Versioning — DVC, model registries, and reproducible ML pipelines.
- CI/CD for ML — Automated testing, training pipelines, and continuous deployment.
- Monitoring & Drift Detection — Monitoring model performance, data drift, and alerting in production.
Topics covered
mlops, deployment, docker, mlflow, fastapi, monitoring, ci-cd