HackRX60/Document-Query-System

AI Summary: Implement a semantic search function that takes a user's question as input, generates a vector embedding for it using a suitable model, queries a Pinecone index to retrieve the most relevant text passages from a document, and returns these passages as a list. This function is the core of a document query system and needs to accurately retrieve relevant context based on semantic understanding.

AI Summary: Set up a basic FastAPI application with a POST endpoint (/hackrx/run) that accepts JSON containing documents and questions, and returns a hardcoded dummy JSON response with an 'answers' field. The endpoint should validate the incoming request format.