orakzai-io/Rednote

Document Assistant (RAG Notebook)

A full-stack RAG (Retrieval-Augmented Generation) document assistant application built to upload, index, and query documents using PostgreSQL (with pgvector), FastAPI, and a React + TypeScript frontend.

Project Structure

├── backend/               # FastAPI Python Backend
├── frontend/              # React (Vite + TypeScript) Frontend
├── docker-compose.yml     # Multi-container orchestration
├── nginx.conf             # Nginx reverse proxy configuration
└── .env.example           # Example environment variables

Quick Start (Docker Compose)

The easiest way to run the entire stack is using Docker Compose.

Prerequisites

• —Docker Desktop (with WSL 2 integration enabled if on Windows)

Steps

1. 1.Configure Environment Variables Copy the .env template from the backend directory to the root (or create a .env in the root containing your keys):

   cp backend/.env .env

Ensure your .env contains a valid GROQ_API_KEY and a secure SECRET_KEY (for JWT authentication).

1. 1.Start the Application Run the following command to build the Docker images and start the services:

   docker compose up -d --build

1. 1.Access the App
2. 2.Frontend Application: http://localhost (runs on port 80 served by Nginx)
3. 3.Backend API Docs: http://localhost:8000/docs
4. 4.PostgreSQL Database: Port 5432 (accessible from host using database settings in .env)

1. 1.Shutdown To stop the services while preserving your data:

   docker compose down

To stop and wipe database data:

   docker compose down -v

Development

For instructions on running backend or frontend individually outside of Docker:

• —Refer to the Backend README
• —Refer to the Frontend README