ckaestne/seai

Covers the full ML lifecycle beyond model training—including deployment pipelines, testing strategies, data quality monitoring, concept drift detection, and MLOps infrastructure (Kafka, Docker, Prometheus). The curriculum emphasizes designing systems resilient to model failures through fault tolerance, safety considerations, and responsible AI practices (fairness, explainability, privacy), while fostering collaboration between software engineers and data scientists through practical case studies and a capstone movie recommendation service project.

446 stars. No commits in the last 6 months.