AugmentED: AI Augmented Learning Library - Backend

Description

WIP - Nothing is finished yet

AI:

• Vector index querying almost working - for pdfs
• custom retrieval and response synthesis next
• ReAct later.. probably

Backend:

• User auth is done
• working on conversations
• need to make file uploading work and save in sql
• SQL schema for everything

Setup

This project uses FastAPI, SQLAlchemy, llama_index and llama_cpp_python, and it requires Python 3.10+.

To start the application, navigate to the project folder and execute the command:

python main.py

The application will be available at http://localhost:5555.

Key Features(TBD)

• User registration and authentication system with OAuth2.
• Secure conversation endpoint that generates a unique, AI-generated response based on the question.
• Uploading books to the library (stored in SQL)
• Making custom notes for topics using AI
• Same for PYQa as well
• SQLite database for user data persistence.

Running the project:

To run this project, you will need to do the following:

1. Clone the project repo or download zip and extract it
   git clone https://github.com/nusaturn/augmentED-backend
2. Create a venv: python -m venv ml-env
3. Install requirements: pip install -r requirements.txt