AI Chatbot - Middleware Assignment

An intelligent AI chatbot application that learns from conversations and can generate personality profiles based on chat history.

Features

• 💬 Free-form Chatting: Chat naturally with an AI assistant powered by OpenAI
• 🧠 Conversation Memory: All conversations are stored and used as context for future interactions
• 👤 Personality Profiling: After a few messages, ask "Who am I?" or "Tell me about myself" to get a personalized personality profile derived from your chat history
• 🔐 Authentication: Secure user authentication with JWT tokens
• 🎨 Modern UI: Clean, responsive interface built with React + Vite, Tailwind CSS, and shadcn/ui components

Tech Stack

• Frontend: React + Vite, JavaScript, Tailwind CSS
• Backend: Express.js REST API (separate server)
• Database: PostgreSQL with Prisma ORM (local server)
• AI: OpenAI GPT-4o-mini
• Authentication: JWT with bcrypt
• Testing: Vitest

Prerequisites

• Node.js 18+ and npm/yarn
• PostgreSQL database server (local installation)
• OpenAI API key

Project Structure

This project is split into Frontend and Backend:

• Frontend: React + Vite application in the root directory
• Backend: Express.js API server in /backend directory

See SETUP.md for detailed setup instructions.

Quick Setup

Backend Setup

cd backend
npm install

Create backend/.env:

DATABASE_URL="postgresql://roopansh@localhost:5432/chatbot?schema=public"
OPENAI_API_KEY="your-openai-api-key"
JWT_SECRET="your-secret-key"
PORT=5000
FRONTEND_URL="http://localhost:3000"

npm run db:generate
npm run db:push
npm run dev

Backend runs on http://localhost:5000

Frontend Setup

# From project root
npm install

Create .env:

VITE_API_URL="http://localhost:5000"

npm run dev

Frontend runs on http://localhost:3000

See SETUP.md for detailed instructions.

Demo Credentials

For testing purposes, you can:

1. Create an account through the sign-up page on the app
2. Use the demo user script (for local development):

   npm run create-demo-user

   Or:

   node scripts/create-demo-user.js

   This will create a demo user with:
   • Email: demo@example.com
   • Password: demo123

Note: Demo credentials are not hardcoded in the codebase. For production, demo credentials should be provided separately in the submission documentation.

Usage

1. Sign Up/Login: Create an account or log in with existing credentials
2. Start Chatting: Begin a conversation with the AI assistant
3. Build Context: Have a few conversations to build up context
4. Ask About Yourself: Try asking:
   • "Who am I?"
   • "Tell me about myself"
   • "What do you know about me?"
   • "Describe me"
   • "What am I like?"

The AI will analyze your conversation history and generate a personality profile!

Running Tests

# Run tests once
npm test

# Run tests in watch mode
npm run test:watch

Building for Production

npm run build
npm start

Deployment

Option 1: Vercel (Recommended)

1. Push your code to GitHub
2. Import your repository in Vercel
3. Add environment variables in Vercel dashboard
4. Deploy!

Option 2: Railway

1. Create a new project on Railway
2. Connect your GitHub repository
3. Add PostgreSQL service
4. Add environment variables
5. Deploy!

Environment Variables for Production

Make sure to set these in your hosting platform:

• DATABASE_URL: Your PostgreSQL connection string
• OPENAI_API_KEY: Your OpenAI API key
• JWT_SECRET: A strong, random secret key

Project Structure

├── app/
│   ├── api/              # API routes
│   │   ├── auth/         # Authentication endpoints
│   │   ├── chat/         # Chat endpoint
│   │   └── messages/     # Messages retrieval
│   ├── chat/             # Chat page
│   ├── layout.tsx        # Root layout
│   └── page.tsx          # Login/signup page
├── components/
│   └── ui/               # Reusable UI components
├── lib/                  # Utility functions
│   ├── auth.ts          # Authentication helpers
│   ├── openai.ts        # OpenAI integration
│   ├── personality.ts   # Personality profiling
│   └── prisma.ts        # Prisma client
├── prisma/
│   └── schema.prisma    # Database schema
└── test/                # Test files

Core Features Implementation

Conversation Storage

All messages are stored in the database with user association, allowing the AI to maintain context across sessions.

Personality Profiling

When a user asks about themselves, the system:

1. Retrieves the last 50 messages from the conversation history
2. Sends them to OpenAI with a specialized prompt
3. Generates a personality profile based on interests, communication style, preferences, and traits

Context-Aware Responses

The AI uses the last 20 messages as context for generating responses, ensuring continuity in conversations.

License

This project is created for the Middleware assignment.

Contact

For questions or issues, please contact the development team.