Retrieval Augmented Generation

You have 500KB of relevant documentation that exceeds the 200K-token context window of your production model. How should you choose between context stuffing (fit what you can) and RAG retrieval?

Your prompt iteration has plateaued. Walk through 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?