AI Knowledge Copilot

A production-grade Retrieval-Augmented Generation (RAG) system that enables users to query PDFs, documents, and web content with high accuracy. The system implements semantic search, reranking, citation-based answers, and hallucination control to provide reliable, verifiable responses.

Features

• Document Processing: Upload and index PDF documents and web content
• Semantic Search: Find relevant information using embeddings and vector similarity
• Reranking: Improve result relevance with cross-encoder models
• Citations: Get answers with source references for verification
• Hallucination Control: Responses grounded only in your documents
• Flexible LLM Support: Use OpenAI or open-source models (Llama, Mistral)
• Modern UI: Clean, responsive Next.js interface with Tailwind CSS

Architecture

This is a monorepo containing:

• backend/: FastAPI server with RAG pipeline (Python)
• frontend/: Next.js web application (TypeScript + React)

Quick Start

Backend Setup

1. Navigate to backend directory:

cd backend

2. Create and activate virtual environment:

python -m venv venv
# Windows
venv\Scripts\activate
# Linux/Mac
source venv/bin/activate

3. Install dependencies:

pip install -r requirements.txt

4. Configure environment:

copy .env.example .env
# Edit .env with your settings

5. Run the server:

uvicorn src.main:app --reload

Backend will be available at http://localhost:8000

Frontend Setup

1. Navigate to frontend directory:

cd frontend

2. Install dependencies:

npm install

3. Configure environment:

copy .env.local.example .env.local
# Edit .env.local if needed

4. Run the development server:

npm run dev

Frontend will be available at http://localhost:3000

Technology Stack

Backend

• FastAPI: Modern Python web framework
• LangChain: LLM application framework
• FAISS: Vector similarity search
• Sentence Transformers: Text embeddings
• PyPDF2/pdfplumber: PDF processing
• BeautifulSoup4: Web scraping
• pytest + hypothesis: Testing

Frontend

• Next.js 14: React framework
• TypeScript: Type safety
• Tailwind CSS: Styling
• React Query: API state management
• Zustand: Global state
• React Dropzone: File uploads

Project Structure

.
├── backend/
│   ├── src/
│   │   ├── __init__.py
│   │   ├── main.py          # FastAPI application
│   │   └── config.py        # Configuration management
│   ├── tests/               # Backend tests
│   ├── data/                # FAISS index and metadata storage
│   ├── requirements.txt     # Python dependencies
│   ├── config.yaml          # Application configuration
│   └── .env.example         # Environment variables template
│
├── frontend/
│   ├── app/
│   │   ├── layout.tsx       # Root layout
│   │   ├── page.tsx         # Home page
│   │   └── globals.css      # Global styles
│   ├── components/          # React components
│   ├── package.json         # Node dependencies
│   ├── tsconfig.json        # TypeScript config
│   ├── tailwind.config.ts   # Tailwind config
│   └── .env.local.example   # Environment variables template
│
└── .kiro/
    └── specs/               # Feature specifications

Configuration

Backend Configuration (config.yaml)

• Embedding model: Sentence transformer model for embeddings
• Chunking: Chunk size and overlap settings
• Retrieval: Top-k results and similarity threshold
• Reranking: Cross-encoder model and top-n selection
• LLM: Provider, temperature, and token limits
• Storage: Paths for FAISS index and metadata

Environment Variables

Backend (.env):

• OPENAI_API_KEY: OpenAI API key (optional, can be set via UI)
• ENCRYPTION_KEY: Key for encrypting stored API keys
• CORS_ORIGINS: Allowed frontend origins

Frontend (.env.local):

• NEXT_PUBLIC_API_URL: Backend API URL
• NEXT_PUBLIC_APP_NAME: Application name

Testing

Backend Tests

cd backend
pytest                    # Run all tests
pytest --cov=src tests/  # Run with coverage
pytest -v tests/         # Verbose output

Frontend Tests

cd frontend
npm test                 # Run all tests
npm run test:watch      # Watch mode

Development

Backend Development

• API documentation available at http://localhost:8000/docs
• Health check endpoint: http://localhost:8000/health
• Hot reload enabled with --reload flag

Frontend Development

• Hot reload enabled by default
• TypeScript type checking
• ESLint for code quality

API Endpoints

• POST /documents/upload - Upload PDF document
• POST /documents/url - Index web content
• GET /documents - List all documents
• GET /documents/{doc_id} - Get document details
• DELETE /documents/{doc_id} - Delete document
• POST /query - Ask a question
• GET /health - Health check
• POST /settings/api-key - Save API key
• GET /settings/api-key - Get current provider
• POST /settings/test-connection - Test LLM connection

About This Project

This is an educational project developed as part of college coursework to learn about:

• Retrieval-Augmented Generation (RAG) systems
• Vector databases and semantic search
• Modern web development with FastAPI and Next.js
• LLM integration and prompt engineering
• Production-grade software architecture

Note for Students: This code is shared for learning and reference purposes. If you're working on a similar assignment, please use this to understand concepts and approaches, but develop your own implementation. Direct copying violates academic integrity policies.

Note for Recruiters/Employers: This project demonstrates my understanding of AI/ML systems, full-stack development, and software engineering best practices.

Contributing

This project follows a spec-driven development approach. See .kiro/specs/ for detailed requirements, design, and implementation tasks.

All code and documentation © 2026. All rights reserved.