🏡 Real Estate AI Qualification Bot

An intelligent voice-based AI assistant that automates lead qualification in real estate

Live Demo : https://realestateaiagentjayprakash.vercel.app/

📋 Table of Contents

Problem Statement
Solution Overview
Features
Tech Stack
Architecture
Installation
Usage
API Documentation
Project Structure
Environment Variables
Contributing
License
Contact

🚨 Problem Statement

Real estate agents today face several critical challenges:

Wasted Time: Up to 60-70% of their time is spent engaging with leads who are not ready or qualified to buy.

Repetitive Tasks: Agents must repeatedly ask the same basic qualifying questions, which drains productivity.

Missed Opportunities: Without 24/7 availability, many potential leads slip away when no one is available to respond.

Manual Data Handling: Agents manually convert conversations into structured data, increasing the chance of errors and delays.

Limited Scalability: Human availability restricts how many leads can be handled simultaneously, capping growth.

Additionally, many customers visiting real estate websites find no live person available to chat, which leads to frustration and lost business.

💡 Our Solution

Our AI-powered voice agent transforms the lead qualification process by:

Engaging Customers Naturally: The AI holds real-time, natural phone conversations with potential buyers, just like a human agent.

Asking Smart Questions: It intelligently gathers all necessary qualification details without boring the lead with repetitive or irrelevant queries.

Automating Data Capture: Spoken answers are converted instantly into structured data, eliminating manual entry and mistakes.

Instant Qualification: Leads are automatically scored and tagged as qualified or unqualified based on preset criteria.

24/7 Availability: The AI works around the clock, never missing a lead regardless of time or volume.

Seamless Admin Integration: Qualified leads are sent directly to the admin dashboard, exported to Excel, and notifications are emailed to administrators in real time.

🎯 Features

🎙️ Voice Interaction

Natural Conversations: AI-powered voice calls using Vapi
Real-time Processing: Instant speech-to-text and text-to-speech
Dynamic Responses: Context-aware conversation flow

🧠 Intelligent Qualification

Smart Questions: AI asks relevant follow-up questions
Budget Analysis: Automatic budget parsing and validation
Location Matching: Property location preference detection
Timeline Assessment: Purchase timeline evaluation

📊 Data Management

Structured Storage: Automatic lead data organization
MongoDB Integration: Scalable database storage
Real-time Updates: Live lead status tracking
Export Options: CSV, JSON data export capabilities

🔄 Workflow Automation

n8n Integration: Visual workflow automation
CRM Sync: Automatic lead forwarding to CRM systems
Notification System: Real-time alerts for qualified leads
Analytics Dashboard: Lead conversion tracking

🛠️ Tech Stack

Frontend

React 18 - Modern UI framework
Tailwind CSS - Utility-first styling
Vite - Fast build tool
Vapi Web SDK - Voice interaction

Backend

Node.js - JavaScript runtime
Express.js - Web framework
MongoDB - NoSQL database
Mongoose - ODM for MongoDB

AI & Automation

OpenAI GPT-4 - Natural language processing
Vapi - Voice AI platform
n8n - Workflow automation
Webhook Integration - Real-time data flow

Development Tools

ESLint - Code linting
Prettier - Code formatting
ngrok - Local tunnel for webhooks
Git - Version control

🏗️ Architecture

graph TD
    A[Customer Call] --> B[Vapi Voice AI]
    B --> C[OpenAI GPT-4]
    C --> D[Natural Language Processing]
    D --> E[n8n Workflow]
    E --> F[Data Processing]
    F --> G[MongoDB Database]
    G --> H[Lead Dashboard]
    
    I[Real Estate Agent] --> H
    H --> J[Qualified Leads]
    H --> K[Analytics & Reports]
    
    E --> L[CRM Integration]
    E --> M[Email Notifications]
    E --> N[SMS Alerts]

🚀 Installation

Prerequisites

Before you begin, ensure you have the following installed:

Node.js (v18 or higher)
npm or yarn
MongoDB (local or cloud)
Git

Quick Start

Clone the Repository

git clone https://github.com/JaYRaNa213/Real-Estate-AI-Sales-Qualification-Bot
cd real-estate-ai-bot

Install Dependencies

# Install frontend dependencies
npm install

# Install backend dependencies (if separate)
cd real-estate-backend
npm install
cd ..

Environment Setup

# Copy environment template
cp .env.example .env

Start Development Servers

npm run css:watch
# Start frontend (port 5173)
npm run dev

# Start backend (port 5000) - in separate terminal
npm run server

# Start n8n workflow (port 5678) - in separate terminal
n8n

Expose Webhooks (Development)

# Install ngrok globally
npm install -g ngrok

# Expose n8n webhook
npx ngrok http 5678

Production Deployment on Replit

Fork this Repl
Set Environment Variables
Configure MongoDB connection string
Deploy

💻 Usage

Starting a Voice Call

Navigate to the Demo Page
```
http://localhost:5173/demo
```
Click "Talk to Agent"
- The AI will answer and start the qualification process
- Speak naturally about your property requirements
Answer Questions
- Budget range
- Preferred location
- Property type (apartment, house, commercial)
- Timeline for purchase
View Results
- Check the dashboard for qualified leads
- Export data or integrate with your CRM

Dashboard Features

Lead Overview: Total, qualified, and unqualified leads
Real-time Status: Active calls and processing
Search & Filter: Find specific leads quickly
Export Options: Download lead data
Analytics: Conversion rates and trends

📡 API Documentation

Endpoints

Leads Management

GET /api/leads

Get all leads with optional filtering

Query Parameters:

status - Filter by qualification status
List - Search by name, phone, or email

Response:

{
  "success": true,
  "data": [
    {
      "_id": "64a7b8c9d12e3f4g5h6i7j8k",
      "name": "John Doe",
      "phone": "+1234567890",
      "email": "john@example.com",
      "budget": "500000-800000",
      "location": "Mumbai",
      "propertyType": "apartment",
      "qualified": true,
      "createdAt": "2024-01-15T10:30:00Z"
    }
  ],
}

Create New Lead

POST /api/leads

Request Body:

{
  "name": "Jane Smith",
  "phone": "+1987654321",
  "email": "jane@example.com",
  "budget": "300000-500000",
  "location": "Bangalore",
}

Webhook Endpoint (n8n)

POST /webhook/qualified

Receives data from Vapi calls and processes lead qualification.

📁 Project Structure

real-estate-ai-bot/
├── 📁 public/                    # Static assets
├── 📁 src/
│   ├── 📁 components/            # React components
│   │   ├── AssistantCallButton.jsx
│   │   ├── LeadCard.jsx
│   │   ├── PropertyCard.jsx
│   │   └── Navbar.jsx
│   ├── 📁 pages/                 # Page components
│   │   ├── Home.jsx
│   │   ├── PropertyDemo.jsx
│   │   └── DemoSplitView.jsx
│   ├── 📁 api/                   # API utilities
│   │   └── api.js
│   ├── 📁 utils/                 # Helper functions
│   ├── App.jsx                   # Main app component
│   ├── index.jsx                 # Entry point
│   └── vapiClient.js            # Vapi configuration
├── 📁 backend/                   # Backend server (if separate)
│   ├── 📁 controllers/
│   ├── 📁 models/
│   ├── 📁 routes/
│   └── server.js
├── 📁 n8n-workflows/            # Automation workflows
├── 📄 .env.example              # Environment template
├── 📄 package.json              # Dependencies
├── 📄 tailwind.config.js        # Tailwind configuration
├── 📄 vite.config.js            # Vite configuration
└── 📄 README.md                 # This file

🔐 Environment Variables

Create a .env file in the root directory:

# Vapi Configuration
VAPI_PUBLIC_KEY=your_vapi_public_key_here
VAPI_PRIVATE_KEY=your_vapi_private_key_here
VAPI_ASSISTANT_ID=your_assistant_id_here

# OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key_here

# Database
MONGODB_URI=mongodb://localhost:27017/real-estate-ai
# Or for MongoDB Atlas:
# MONGODB_URI=mongodb+srv://username:password@cluster.mongodb.net/dbname

# n8n Configuration
N8N_WEBHOOK_URL=https://your-ngrok-url.ngrok.io/webhook
N8N_RUNNERS_ENABLED=true

# Server Configuration
PORT=5000
NODE_ENV=development

# Frontend URL (for CORS)
FRONTEND_URL=http://localhost:5173

Environment Variables Description

Variable	Description	Required
`VAPI_PUBLIC_KEY`	Vapi public API key for frontend	✅
`VAPI_PRIVATE_KEY`	Vapi private API key for backend	✅
`VAPI_ASSISTANT_ID`	Your configured Vapi assistant ID	✅
`OPENAI_API_KEY`	OpenAI API key for GPT integration	✅
`MONGODB_URI`	MongoDB connection string	✅
`N8N_WEBHOOK_URL`	n8n webhook endpoint URL	✅
`PORT`	Server port (default: 5000)	❌
`NODE_ENV`	Environment (development/production)	❌

🧪 Testing

Run Tests

# Run all tests
npm test

# Run tests with coverage
npm run test:coverage

# Run tests in watch mode
npm run test:watch

Manual Testing

Voice Call Testing
- Test different budget ranges
- Try various property types
- Test edge cases (unclear responses)
API Testing
- Use Postman or curl to test endpoints
- Verify webhook functionality
- Test database operations
Integration Testing
- End-to-end call flow
- Data persistence verification
- n8n workflow testing

📊 Performance Metrics

Key Performance Indicators

Response Time: < 2 seconds for voice responses
Accuracy: 95%+ intent recognition accuracy
Uptime: 99.9% availability target
Conversion: Track qualified lead percentage

Monitoring

# Check system status
npm run health-check

# View performance logs
npm run logs

# Monitor database performance
npm run db:status

🔧 Troubleshooting

Common Issues

1. Voice Call Not Starting

# Check Vapi configuration
console.log('Vapi Key:', process.env.VAPI_PUBLIC_KEY)

# Verify assistant ID
console.log('Assistant ID:', process.env.VAPI_ASSISTANT_ID)

2. Database Connection Issues

# Test MongoDB connection
npm run db:test

# Check connection string
echo $MONGODB_URI

3. Webhook Not Receiving Data

# Check ngrok status
ngrok status

# Verify webhook URL in Vapi dashboard
curl -X POST your-webhook-url/test

4. Frontend Build Issues

# Clear cache and reinstall
rm -rf node_modules package-lock.json
npm install

# Check for version conflicts
npm audit

Debug Mode

Enable debug logging:

DEBUG=true
LOG_LEVEL=debug

🚀 Deployment

Replit Deployment (Recommended)

Prepare for Deployment

# Build the project
npm run build

# Test production build
npm run preview

Configure Replit Secrets
- Add all environment variables to Replit Secrets
- Ensure database URLs are accessible from Replit
Deploy
- Use Replit's built-in deployment feature
- Configure custom domain if needed

Manual Deployment

# Build and deploy
npm run build
npm run deploy

# Or deploy to specific environment
npm run deploy:production

Development Workflow

Fork the repository
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

Code Standards

Use ESLint and Prettier for code formatting
Write unit tests for new features
Follow conventional commits for commit messages
Update documentation for API changes

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📞 Contact & Support

Developer

Jay Prakash Rana

📧 Email: jayrana0909@gmail.com
🐙 GitHub: @your-github-handle
💼 LinkedIn: Your LinkedIn Profile

Support

📖 Documentation: Wiki

🎥 Demo Video

Live Demo: https://realestateaiagentjayprakash.vercel.app/

🏆 Hackathon Information

Event Details

Detail	Information
Hackathon	Swafinix AI Hackathon 2025
Category	Real Estate – AI Sales Qualification Bot
Participant	Jay Prakash Rana
Status	✅ Working Prototype Complete

Important Dates

Event	Date & Time (IST)
Registration Deadline	01 Aug 2025, 12:00 AM IST
Hackathon Start	01 Aug 2025
Submission Deadline	10 Aug 2025, 01:21 AM IST
Winners Announcement	15 Aug 2025

Prizes

💼 PPO Opportunity: 12 LPA with Swafinix Technologies

🌟 Acknowledgments

Vapi for providing excellent voice AI platform
OpenAI for powerful GPT models
n8n for flexible workflow automation
MongoDB for reliable database services
Replit for seamless development and deployment

📈 Roadmap

Phase (Current)

✅ Voice call qualification
✅ Basic lead management
✅ MongoDB integration
✅ n8n workflows

⭐ Star this repo if you find it helpful!

Made with ❤️ by Jay Prakash Rana

Name		Name	Last commit message	Last commit date
Latest commit History 61 Commits
images/i.img		images/i.img
real-estate-backend		real-estate-backend
real-estate-ui		real-estate-ui
DOCUMENTATION.md		DOCUMENTATION.md
Readme.md		Readme.md
n8n_workFlow.json		n8n_workFlow.json

JaYRaNa213/Real-Estate-AI-Sales-Qualification-Bot

Folders and files

Latest commit

History

Repository files navigation