This repository contains a small end-to-end example LLM application. The project demonstrates a minimal production-like setup with a FastAPI backend serving a LangChain-based LLM service, a React frontend, PostgreSQL for persistent storage, and Redis for caching/rate-limiting.

Overview

• Purpose: Example app showing how to wire a simple LLM service into a web app with persistent storage and caching.
• Backend: backend_service — FastAPI app exposing API endpoints and the LangChain-powered LLM service.
• Frontend: frontend_service — React app that interacts with the backend.
• Data stores: PostgreSQL for storing application data (users, query logs) and Redis for caching and rate limiting.

Tech Stack

• LLM / Orchestration: LangChain
• API: FastAPI + Uvicorn
• Frontend: React
• DB / Cache: PostgreSQL, Redis
• Containerization: Docker & Docker Compose

Repository Layout

• backend_service/ — backend implementation, configs, and its own README.md.
• frontend_service/ — React frontend and its README.md.
• docker-compose.yml — orchestrates backend, frontend, Postgres, Redis, and other services.

OpenAI Model Configuration

This application uses OpenAI models via LangChain. To use the service, you need:

1. OPENAI_API_KEY
   • Get your OPENAI_API_KEY from https://platform.openai.com
   • supported models: gpt-4, gpt-4-turbo, gpt-3.5-turbo, etc.

Notes & Links

• set .env file for backend_service, refer backend_service/README.md

Start End to End Application To spin up the entire stack quickly

docker-compose up --build

Accessing the Application

After starting the project, you can access the following:

• http://localhost:3000/

This is the main user interface of the application.

• http://localhost:8000/docs

MIT License

For issues and questions, please create an issue in the repository.