By scipio on Skatehive
Learn AI Series (#34) - ML Engineering - From Notebook to Production What will I learn You will learn the gap between "works in Jupyter" and "works in production" -- and how to bridge it; model serialization with joblib -- saving the entire pipeline so preprocessing and model travel together; batch vs real-time serving -- when precomputed predictions beat on-demand inference; serving a model with FastAPI -- a minimal but real deployment you can build in an afternoon; feature stores and shared feature modules -- the consistency problem that silently kills production ML; model monitoring -- detecting data drift, concept drift, and prediction drift before they cost you; A/B testing and shadow mode for validating new model versions in production. Requirements A working modern computer running macOS, Windows or Ubuntu; An installed Python 3(.11+) distribution; The ambition to learn AI and machine learning. Difficulty Beginner Curriculum (of the Learn AI Series): Learn AI Series (#1) - What