Chat with your docs, fully offline.

What it is

Local RAG App is a retrieval-augmented generation system that runs entirely on your own machine. You upload PDFs, plain text, or Markdown files and ask questions about them in natural language. Nothing is sent to a third-party API: the language model and the embedding model both run locally through Ollama.

That constraint is the point. Teams with sensitive documents, legal files, medical records, internal engineering docs, often cannot use hosted AI services. A fully local pipeline gives them document question answering without a privacy trade-off.

How it works

Documents are chunked and embedded locally, then stored in ChromaDB for semantic search. At question time the app retrieves the most relevant chunks, feeds them to a local LLM served by Ollama, and returns an answer grounded in the retrieved passages, with source citations pointing back to the original documents.

The backend is Python with FastAPI, and the whole system ships in Docker so it can be stood up with a single command. Grounded answers come back in about half a second.

What I owned

I built the system end to end: the ingestion and chunking pipeline, the embedding and retrieval layer over ChromaDB, the FastAPI service, the Ollama integration, and the Docker packaging.