RAG (Retrieval-Augmented Generation)
Injecting relevant code or docs into a model's context at query time.
Retrieval-Augmented Generation is a pattern where, instead of relying on a model's training knowledge, the system retrieves relevant text at query time and pastes it into the context. In AI coding tools, RAG is how a chat feature "knows" about your codebase without fine-tuning a model on it.