Adapting LLMs

Prompting, RAG, fine-tuning, distillation — and how to choose between them

• Prompting as Engineering — The cheapest, most reversible adaptation lever: instructions, examples, and structure — treated like code, not incantations. (4 lessons, 45 min)
• Context Engineering at the Model Level — The window is a budget you allocate: what goes in, where it sits, and how structure interacts with caching and attention. (4 lessons, 40 min)
• RAG Mechanics: Embeddings and Search — The machinery under retrieval: embedding models, chunking, approximate nearest-neighbor search, and hybrid ranking. (4 lessons, 50 min)
• RAG End to End — From ingest to grounded answer: the full pipeline, reranking, the failure modes that actually occur, and how to evaluate any of it. (4 lessons, 50 min)
• Fine-Tuning: Full, LoRA, QLoRA — Moving behavior into the weights: full fine-tuning, low-rank adapters, quantized training — and the recipe that keeps it honest. (4 lessons, 50 min)
• Distillation — Teaching a small model with a big one: synthetic data, filtering, and what actually transfers. (3 lessons, 40 min)
• In Production: Choosing Your Lever — The domain’s production capstone: prompt, retrieval, or weights — the decision framework, the cost shapes, and why real systems compose all three. (3 lessons, 40 min)