Skip to content

← Back to projects

AI Documentation RAG Agent

Overview

A system that ingests selected documentation and makes it possible to ask questions about its contents.

Documents are indexed in a vector database, after which a Telegram interface retrieves relevant passages and generates an answer grounded in the connected sources.

Focus

How to ground answers in a defined knowledge corpus and make documentation search convenient in conversation.

Technologies

  • Python
  • RAG
  • Pinecone
  • Telegram
  • LLM API