🤖 RAG Chatbot System - Intelligent Domain-Specific Assistant

A complete full-stack Retrieval-Augmented Generation (RAG) system with modern web interface and intelligent backend, delivering specialized AI assistance across multiple domains.

📸 Project Screenshots

Chat Interface

Modern, responsive chat interface with real-time messaging and typing indicators

Backend Architecture Complete RAG pipeline architecture diagram

Authentication System Secure user authentication with Supabase integration

Source Citations Panel Source citations and references for transparent AI responses

🚀 Project Highlights

Backend System

• 🎯 Business Impact: Intelligent knowledge assistant across 5 critical domains (Agriculture, Education, Environment, Finance, Healthcare)
• ⚡ Performance: Sub-second response times with real-time performance tracking
• 💰 Cost Management: Smart credit-based usage control preventing API abuse
• 🔒 Production-Ready: Secure API key management, CORS support, and scalable serverless architecture

Frontend Application

• 🤖 Multi-Domain AI Assistant: 5 specialized categories with domain-specific knowledge bases
• 💬 Advanced Chat Interface: Real-time chat with typing indicators and message history
• 🔐 User Authentication: Supabase integration with credit-based usage system
• 🎨 Modern UI/UX: Built with Next.js 15, React 19, and Tailwind CSS
• 📱 Responsive Design: Works seamlessly across desktop and mobile devices

---

🏗️ System Architecture

Core RAG Pipeline

1. Query Embedding → Cohere's embed-english-v3.0 converts queries to 1024-dim vectors
2. Semantic Search → Pinecone retrieves top-3 relevant documents with domain filtering
3. Context Augmentation → Retrieved documents provide contextual knowledge
4. Response Generation → Google Gemini 1.5 Flash generates human-like responses
5. Metadata Enrichment → Returns sources, confidence scores, and performance metrics

---

🛠️ Technology Stack

Frontend Technologies

• Next.js 15 - React framework with App Router and dynamic routing
• React 19 - Latest React with enhanced performance
• Tailwind CSS - Utility-first CSS framework with dark/light theme support
• Supabase - User authentication and management
• LocalStorage - Message history persistence

Backend & API

• FastAPI - High-performance async web framework
• Python - Latest Python with enhanced performance
• Pydantic - Data validation and serialization

AI & Machine Learning

• Google Gemini 1.5 Flash - Large Language Model for response generation
• Cohere v3.0 - State-of-the-art text embeddings
• Custom System Instructions - Optimized prompts for consistent output

Data & Storage

• Pinecone - Serverless vector database (AWS us-east-1)
• Redis - Managed Redis for credit tracking
• JSON Data Sources - Curated domain-specific knowledge bases
• Supabase Database - User data and chat history storage

---

📊 Business Domains & Use Cases

Domain	Use Cases	Sample Query
🌾 Agriculture	Farming practices, crop management, agricultural policies	"What are the best drought-resistant crops for Indian farmers?"
🎓 Education	Educational policies, curriculum, learning resources	"What are the key features of India's National Education Policy 2020?"
🌍 Environment	Climate policies, renewable energy, sustainability	"What are India's commitments under COP26?"
💼 Finance	Economic policies, financial regulations, market insights	"What are the latest RBI monetary policy changes?"
🏥 Healthcare	Health policies, medical guidelines, public health	"What are India's vaccination strategies for rural areas?"

---

🔧 Installation & Setup

Prerequisites

• Node.js 18+ and npm/yarn
• Python 3.12+
• API Keys: Google Gemini, Cohere, Pinecone, Upstash Redis
• Supabase project for authentication

Backend Setup

# Clone the repository
git clone <repository-url>
cd Rag_ChatBot_Backend

# Install dependencies
pip install -r requirements.txt

# Set environment variables
cp .env.example .env
# Add your API keys to .env

# Run the backend
uvicorn main:app --reload

Frontend Setup

# Navigate to frontend directory
cd rag-frontend

# Install dependencies
npm install

# Set environment variables
cp .env.local.example .env.local
# Add your Supabase keys

# Run the development server
npm run dev

Environment Variables

Backend (.env)

GEMINI_API_KEY=your_gemini_api_key
COHERE_API_KEY=your_cohere_api_key
PINECONE_API_KEY=your_pinecone_api_key
REDIS_URL=your_upstash_redis_url
REDIS_TOKEN=your_upstash_redis_token

Frontend (.env.local)

NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_key
NEXT_PUBLIC_BACKEND_URL=http://localhost:8000

---

📁 Project Structure

ChatBot/
├── Rag_ChatBot_Backend/        # Backend API
│   ├── main.py                 # FastAPI application entry point
│   ├── requirements.txt        # Python dependencies
│   ├── config/
│   ├── embeddings/
│   ├── llms/
│   ├── response/
│   └── sample_data/
└── rag-frontend/               # Next.js Frontend
    ├── src/
    │   ├── app/                # Next.js App Router
    │   │   ├── layout.js       # Root layout
    │   │   ├── page.js         # Home page
    │   │   └── chat/           # Chat routes
    │   ├── components/         # Reusable components
    │   ├── context/            # React context providers
    │   ├── lib/
    │   └── utils/
    └── public/                 # Static assets

---

Developer Profile

Ready to discuss how this RAG system can solve your business challenges?

• 📧 Email: himadrikaran516@gmail.com | karanhimadri1234@gmail.com
• 💼 LinkedIn: linkedin.com/in/himadrikaran
• 🐱 GitHub: github.com/karanhimadri

---

Built with ❤️ for intelligent information retrieval and enhanced user experiences. A complete full-stack solution combining cutting-edge AI with modern web technologies.