Empowering Departments with AI-Driven Knowledge.
Effortlessly query lecture notes, assignments, and exam papers using the power of RAG and Google Gemini.
- 🧠 Advanced RAG Engine: Retrieval-Augmented Generation using ChromaDB for pinpoint accurate context retrieval from uploaded PDFs.
- ⚡ Google Gemini Integration: Leverages Gemini 1.5 Flash for intelligent, low-latency responses and document analysis.
- 🔐 Enterprise-Grade Auth: Custom authentication system using bcrypt with SHA-256 pre-hashing, ensuring unlimited password length support and top-tier security.
- 🖼️ Modern Aesthetics: A premium UI built with Next.js 15+ and Tailwind CSS v4, featuring glassmorphism effects and seamless navigation.
- 🌐 Cloud-Ready Networking: Integrated ngrok multi-tunnel configuration for instant, secure public access to both frontend and backend.
- Framework: FastAPI (Asynchronous Python)
- AI Orchestration: LangChain
- Vector Search: ChromaDB
- Database: SQLModel (SQLAlchemy + Pydantic)
- Security: Custom Bcrypt + JWT
- Framework: Next.js 15+ (App Router)
- Styling: Tailwind CSS v4
- API Client: Standardized Fetch with ngrok-bypass headers
- Icons: Lucide React
Navigate to the backend directory and install dependencies:
cd backend
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txtConfigure your .env file:
DATABASE_URL=sqlite:///./cis.db
SECRET_KEY=yoursecretkey
GOOGLE_API_KEY=your-gemini-keyRun the server:
uvicorn app.main:app --reloadNavigate to the frontend directory and install packages:
cd frontend
npm install
npm run devShare your project instantly using the multi-tunnel configuration:
ngrok start --all --config ngrok-cis.ymlbackend/app/routers: API endpoints for Auth, Document Processing, and Chat.backend/app/services: Core logic for RAG (VectorStore) and Document Analysis (DocumentProcessor).frontend/src/app: Modern Next.js pages including the Faculty Dashboard and Student Chat.frontend/src/lib: CentralizedapiFetchwith built-in ngrok warning bypass.
- Faculty: Upload PDF documents via the Dashboard to build the knowledge base.
- Student: Navigate to AI Chat to ask complex questions based on the uploaded material.
- Admin: Monitor system health and user activity.
Distributed under the MIT License. See LICENSE for more information.
Developed with ❤️ for Advanced Academic Intelligence.