AI-Powered Digital Library

An AI-based digital library that leverages advanced AI, RAG (Retrieval-Augmented Generation), and similarity search concepts to enable users to upload documents (PDFs and websites) and interact with AI-driven insights. The application generates summaries and embeddings, making documents easily searchable and interactive.

Features

• Upload and embed PDFs and websites.
• Automatic summary generation and storage.
• Search for similar content using embeddings.
• AI-powered responses and insights from documents.
• User authentication for secure access.
• Cloud-based storage for documents and embeddings.
• Planned: Rate limiting for enhanced performance.

Tech Stack

• Frontend: React (Client)
• Backend: Node.js, Express (Server)
• Database: MongoDB (Atlas)
• Vector Store: Upstash
• File Storage: Cloudinary
• Embeddings: Geko
• AI and RAG: LangChain, Gemini API
• Monorepo: Custom monorepo setup

Project Structure

vinitngr-iolib/
├── client/          # Frontend (React, Vite)
├── server/          # Backend (Node.js, Express, TypeScript)
├── practice-js/     # JS practice and RAG experiments
├── README.md
├── package.json
└── structure.txt

Installation

1. Clone the repository.

    git clone https://github.com/vinitngr/IOLIB.git

2. Install dependencies for both client and server:

   npm install
   cd client && npm install
   cd ../server && npm install

3. Set up environment variables:
   • MongoDB URI (Atlas)
   • Upstash API Key
   • Gemini API Key
   • Cloudinary Credentials
4. Run the application:

   npm run dev

Usage

1. Access the web app at http://localhost:3000.
2. Upload a PDF or website link to generate embeddings and summaries.
3. Interact with the documents via AI-driven responses and similarity search.

Screenshots

Home Page

Chat Area

1. Access the web app at http://localhost:3000.
2. Upload a PDF or website link to generate embeddings and summaries.
3. Interact with the documents via AI-driven responses and similarity search.

Potential Future Improvements

• Consider implementing rate limiting to manage high query volumes.
• Explore enhancing error handling and user feedback during file upload and processing.
• Evaluate the introduction of caching mechanisms to optimize frequent queries.
• Integrate Optical Character Recognition (OCR) to enable text extraction from scanned documents and images.

Contributing

Feel free to open issues and contribute by submitting pull requests.