Closed ctmartinez1992 closed 2 months ago
Since we are hitting the max token limit when calling the LLM, because we are passing the entire codebase on the prompt, we are going to use a RAG to greatly reduce the prompt size. We will only send relevant code in the prompt.
https://revs.runtime-revolution.com/leveraging-your-own-data-with-llms-experiments-limitations-and-future-improvements-8f763f3806a8
https://revs.runtime-revolution.com/langchain-langchain-rb-and-rag-based-querying-801b83ca6cc1
Since we are hitting the max token limit when calling the LLM, because we are passing the entire codebase on the prompt, we are going to use a RAG to greatly reduce the prompt size. We will only send relevant code in the prompt.
https://revs.runtime-revolution.com/leveraging-your-own-data-with-llms-experiments-limitations-and-future-improvements-8f763f3806a8
https://revs.runtime-revolution.com/langchain-langchain-rb-and-rag-based-querying-801b83ca6cc1