Pax AI 🕊️

A state-of-the-art, real-time AI-powered search and Q&A engine designed to cut through the noise of the web and deliver clean, structured, and factual answers.

🌐 Live URL: https://paxai.onrender.com

---

Pax AI combines semantic web search with advanced language models to offer a smooth, peaceful, and lightning-fast knowledge discovery experience.

---

🚀 Key Features

• Retrieval-Augmented Generation (RAG): Leverages Tavily Search API to fetch real-time web results and synthesizes them into coherent, cited answers.
• Dual-Model Intelligence: Integrated with Google Gemini (genai) and Mistral AI via LangChain for high-quality reasoning and generation.
• Voice Responses: Powered by ElevenLabs to convert text responses into high-fidelity, natural-sounding audio.
• Real-time Streaming: Utilizes Socket.io for real-time streaming of tokens and instant updates.
• Media Support: Integrated with ImageKit for storing and processing chat attachments.
• Secure Authentication: Features cookie-based authentication using JSON Web Tokens (JWT), input validation with Zod / Express Validator, and email verification via Brevo.

---

🛠️ Tech Stack

Frontend

• Framework: React 19 + Vite
• Styling: Tailwind CSS v4 (Modern & responsive dark-mode UI)
• State Management: Redux Toolkit & React Redux
• Routing: React Router v8
• Real-time Connections: Socket.io Client

Backend

• Runtime: Node.js & Express
• Database: MongoDB (via Mongoose)
• AI & Search Orchestration: LangChain, Tavily API, Google GenAI SDK, Mistral AI SDK
• Email Platform: Brevo SMTP Service
• Image Hosting: ImageKit SDK

---

📁 Directory Structure

Pax-AI/
├── Backend/                # Node.js & Express Server
│   ├── src/
│   │   ├── controllers/    # Authentication & Chat Controllers
│   │   ├── services/       # AI, Search, Email, & Upload Services
│   │   ├── utils/          # Verification Email Templates & Helpers
│   │   └── models/         # MongoDB Mongoose Schema definitions
│   ├── server.js           # Server Entry Point
│   └── package.json
└── Frontend/               # React 19 client
    ├── src/
    │   ├── Features/       # Modules (Auth, Chat Dashboard, Sidebar)
    │   └── app/            # App setup (Redux store, routes)
    ├── index.html          # SPA HTML template
    └── package.json

---

⚙️ Configuration & Environment Setup

Before running the application, you need to configure the environment variables for both development and production.

1. Local Development Setup

Create a .env file inside the Backend/ directory:

# Server Configuration
PORT=5000
NODE_ENV=development

# Frontend Client URL
CLIENT_URL=http://localhost:5173

# Database Connection
MONGODB_URI=mongodb://localhost:27017/pax-ai

# Security & Tokens
JWT_SECRET=your_super_secret_jwt_key

# Gemini & Mistral APIs
GEMINI_API_KEY=your_gemini_api_key
MISTRAL_API_KEY=your_mistral_api_key

# Search API
TAVILY_API_KEY=your_tavily_api_key

# Voice API
ELEVENLABS_API_KEY=your_elevenlabs_api_key

# Email Verification (Brevo)
BREVO_API_KEY=your_brevo_api_key
BREVO_USER=your_verified_brevo_sender_email

# Image Uploads (ImageKit)
IMAGEKIT_PRIVATE_KEY=your_imagekit_private_key

2. Production Setup (Render)

For deployment on Render, configure the following environment variables in your Render Web Service settings:

Variable	Description	Value / Format
NODE_ENV	Mode of the Node environment	production
CLIENT_URL	Live origin of the frontend client	https://paxai.onrender.com
MONGODB_URI	Production database connection string	mongodb+srv://<user>:<password>@cluster.mongodb.net/pax-ai (MongoDB Atlas)
JWT_SECRET	Secure key for signing JWTs	[Use a strong random secret key]
GEMINI_API_KEY	Google Gemini API Key	[Your API Key]
MISTRAL_API_KEY	Mistral AI API Key	[Your API Key]
TAVILY_API_KEY	Tavily Search API Key	[Your API Key]
ELEVENLABS_API_KEY	ElevenLabs Text-to-Speech API Key	[Your API Key]
BREVO_API_KEY	Brevo SMTP Service API Key	[Your API Key]
BREVO_USER	Verified Brevo sender email address	[Your sender email]
IMAGEKIT_PRIVATE_KEY	ImageKit private key for secure uploads	[Your private key]

Note

By default, frontend API and Socket.io requests are dynamically routed based on the current origin (window.location.origin). No frontend-specific environment variables need to be set in production when hosted together.

---

🏃 Getting Started (Local Development)

Prerequisites

• Node.js (v18 or higher)
• MongoDB running locally or in the cloud

Steps to Run

1. Clone the repository and navigate to the project directory.

2. Start the Backend Server:

   cd Backend
   npm install
   npm run dev

   The server will start on http://localhost:5000.

3. Start the Frontend App:

   cd ../Frontend
   npm install
   npm run dev

   The application will be accessible at http://localhost:5173.

---

🌐 Production Deployment on Render

This project is configured as a monorepo setup where the Express backend serves the React frontend assets statically from the Backend/public directory.

Render Web Service Setup

Follow these steps to deploy this project to Render:

1. Create a new Web Service: Link it to your repository.
2. Environment: Select Node as the runtime.
3. Build Command: Set this command to install dependencies, build the frontend, and copy the assets to the backend's public directory:

   # Install & build Frontend
   npm install --prefix Frontend && npm run build --prefix Frontend
   
   # Copy built Frontend bundle to Backend's public folder
   mkdir -p Backend/public
   cp -r Frontend/dist/* Backend/public/
   
   # Install Backend dependencies
   npm install --prefix Backend

4. Start Command: Run the Express server:

   npm start --prefix Backend

5. Environment Variables: Add all the production environment variables listed in the production setup table above.

---

🔒 Security & Validation

• Passwords are encrypted using bcryptjs.
• All inputs are schema-validated using Zod and express-validator to prevent SQL/NoSQL injection and cross-site scripting (XSS).
• Cookies are stored with httpOnly, secure, and sameSite: strict settings to safeguard against cross-site request forgery (CSRF) and session hijacking.