RAG basics — embeddings, vector search, chunking

Retrieval-Augmented Generation is the workhorse of the "Select" strategy. Students
learn the pipeline end to end: chunking source material, generating embeddings,
storing them in a vector index, and retrieving the most relevant chunks at query
time to ground the model's response in verifiable data rather than parametric
memory. This is taught as dynamic, just-in-time knowledge integration — a way to
give the model information it never saw in training without retraining it.