MLOps

Go from first principles to running a production ML platform: data versioning, feature stores, CI/CD/CT, Kubernetes serving, monitoring, drift detection, security, cost, LLMOps, and incident response.