🤖 AI LAM - The Autonomous University Application Agent

AI LAM is a sophisticated SaaS platform designed to automate the tedious and time-consuming process of filling out university applications. Leveraging a powerful Large Action Model (LAM), this system can intelligently parse user-provided documents and autonomously navigate complex web forms, providing a seamless and efficient experience for students and educational consultants.

LEGAL DISCLAIMER: This project is for educational and research purposes. Automated submissions may violate the Terms of Service of university application portals like UCAS. Users must ensure compliance before use.

✨ Core Features

• Autonomous Web Automation: The LAM backend uses Playwright to intelligently identify and fill form fields, handle multi-step processes, and manage complex UI interactions.
• Multi-Source Data Entry: Users can provide their information in various formats (CSV, PDF, DOC, TXT, MD), and the system will parse and utilize it for form filling.
• Real-time Browser Streaming: Customers can watch the Playwright browser session live in their dashboard, seeing the automation happen in real-time.
• SaaS-Ready Architecture: Built with a multi-tenant architecture, ready for subscription-based access, with user authentication and role-based permissions.
• Modern & Sleek UI: The frontend is designed to be beautiful, futuristic, and intuitive, inspired by modern platforms like OpenAI, Adaline, and Reflect.

🏗️ System Architecture

AI LAM is built with a modern, scalable, and secure architecture designed for a production-ready SaaS application.

• Frontend: A sleek React 18 application built with Tailwind CSS for a futuristic UI. Deployed on Vercel for a fast, global CDN.
• Backend: A high-performance FastAPI (Python 3.11+) server that orchestrates the automation tasks. Can be deployed on Koyeb, Railway, or Cloud Run.
• Database & Authentication: Supabase (PostgreSQL) provides a robust database, user authentication, and real-time capabilities.
• Automation Engine: The core of the system is a powerful Playwright engine that runs headless browser sessions in a custom VM environment.
• Real-time Streaming: WebSockets are used to stream the live browser session from the backend to the customer's dashboard.

📋 Getting Started

1. Database Setup (Supabase)

• Create a Supabase Project: Go to the Supabase Dashboard and create a new project.
• Run the Schema: In the SQL Editor, copy and paste the contents of backend/database/setup.sql to create all the necessary tables for users, subscriptions, and applications.

2. Environment Setup

• Create a .env file in the backend directory (use ENV_SETUP.md as a template).
• Create a .env file in the frontend directory (use ENV_SETUP.md as a template).
• Add your Supabase keys, generate JWT secrets, and configure any other services.

3. Running Locally

Start the Backend:

cd backend
pip install -r requirements.txt
uvicorn server:app --reload

Start the Frontend:

cd frontend
npm install
npm start

Your application will be available at http://localhost:3000.

4. Deployment

• One-container (Koyeb): Build the included multi-stage Dockerfile in backend/. This serves the React build via FastAPI and runs the API.
  • Set secrets in Koyeb (e.g., SUPABASE_URL, SUPABASE_KEY, GOOGLE_API_KEY, JWT_SECRET).
  • Exposed port is 8080.
• Split deploy: Backend to Railway/Cloud Run using backend/Dockerfile, frontend to Vercel using frontend/vercel.json.

🔧 Technologies Used

• React: For building a modern and interactive user interface.
• FastAPI: A high-performance Python framework for the backend API.
• Supabase: For a scalable PostgreSQL database and user authentication.
• Playwright: For robust and intelligent browser automation.
• Stripe: Ready for integration for subscription management.
• Docker: For containerizing the backend for consistent deployments.
• GitHub Actions: For a full CI/CD pipeline.

🎯 What The System Can Do

• For Customers:

  • Sign up for a subscription plan.
  • Upload their personal and academic information in various document formats.
  • Specify the universities they want to apply to.
  • Watch in real-time as the AI LAM fills out their applications.
  • Track the status of all their applications in a beautiful dashboard.

• For Admins:

  • A comprehensive admin panel to manage users, subscriptions, and monitor system health.
  • Analytics on application success rates and system performance.
  • The ability to intervene or assist with customer applications if needed.

🚀 Future Vision

• Enhanced AI Capabilities: Integrate more advanced AI models for parsing complex documents and handling dynamic web forms.
• Stripe Integration: Complete the subscription and payment system with Stripe.
• Expanded University Support: Add more university application portals to the automation engine.
• White-Label Solution: Offer a white-label version for educational consultants and agencies.

🤝 Contributing

1. Fork the repository.
2. Create a new feature branch.
3. Make your changes and submit a pull request.

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.

🙏 Acknowledgments

This project was built with the assistance of an AI pair programmer and inspired by the vision of a truly autonomous web agent. It utilizes a powerful stack of open-source technologies.