2h 20m total
A hands-on journey to building production applications on top of foundation models — Claude, GPT, and open models — rather than training them. You'll go from your first API call to a shipped, evaluated RAG-plus-agent system: prompting, structured outputs, tool use, embeddings, retrieval-augmented generation, agent loops, evals, and the cost, latency, caching, and safety concerns that separate a demo from production. Mixes prose, runnable Python, MCQ checks, prediction drills, ordering and matching activities, a prompt playground, an embedded REPL, engineering checklists, and mentor-chat prompts so every skill is rehearsed in the format that fits it best. Assumes Python basics; teaches the application layer, not model internals.
