I’ve been working on Ragpi, an open-source AI assistant that builds knowledge bases from docs, GitHub Issues, and READMEs. It uses Redis Stack as a vector DB and leverages RAG to answer technical questions through an API.

Some things it does:

- Creates knowledge bases from documentation websites, GitHub Issues, and READMEs

- Uses hybrid search (semantic + keyword) for retrieval

- Uses tool calling to dynamically search and retrieve relevant information during conversations - Works with OpenAI or Ollama

- Provides a simple REST API for querying and managing sources

Built with: FastAPI, Redis Stack, and Celery.

It’s still a work in progress, but I’d love some feedback!

Repo: https://github.com/ragpi/ragpi

API Reference: https://docs.ragpi.io