Retrieval Augmented Generation

You have 500KB of relevant documentation and a production model with a 200K-token context window. How should you choose between context stuffing (send it all) and RAG retrieval?

Your prompt iteration has plateaued. Trace the decision of whether to fine-tune or invest in further prompt engineering, with explicit signals for each path.

When should you fine-tune the model instead of iterating on the prompt?