me-agent

A simple RAG agent to chat with your personal documents. This is a simple and reproducible example of a RAG agent that can be used for example to integrate in a personal portfolio website.

This project is based on the LangChain RAG Tutorial.

Installation

This project is managed with uv.

uv sync

Usage

1. Add Data

Place your documents (PDF, DOCX, or Markdown) in the data/ directory.

2. Ingest Data

Run the ingestion script to create the vector store:

uv run ingestion/ingest.py

3. Run Agent

Start the chat interface:

uv run me-agent/agent.py

4. Run API Server

Start the FastAPI server:

uv run server.py

The server will be available at http://localhost:8000. You can test it with:

curl -N -X POST http://localhost:8000/chat \
     -H "Content-Type: application/json" \
     -d '{"message": "What projects has Duarte done?"}'