Each course is written for engineers, and every chapter ends with a short section for new grads and one for managers. The opening chapter of each course is free to read.
Twelve chapters on measuring LLM systems: datasets, scorers, model judges, human review, RAG and agent evals, safety, and monitoring once the system is live. Each chapter closes with notes for new grads and for the people managing the work.
Twelve chapters on retrieval: chunking, embeddings, hybrid search, reranking, query rewriting, citations and keeping a live system fresh. Each chapter closes with notes for new grads and for the people managing the work.
Twelve chapters on building agents: tools, stopping rules, planning, memory, failure handling, evaluation and human approval for risky actions. Each chapter closes with notes for new grads and for the people managing the work.