Realtor AI Assistant

Realtor AI Assistant is a full-stack LangGraph real estate agent with a Python backend and a React chat widget. It coordinates lead intake, property-search criteria, database lookup, Google Calendar scheduling, and chat, SMS, and voice entry points, with human approval before sensitive appointment actions.

Backend: LangGraph Real Estate Agent

Key Features

1. Property Information Inquiry

   • Provides detailed property information, including square footage, pricing, condition, and amenities.

2. Appointment Scheduling

   • Integrates with Google Calendar to facilitate the scheduling of property viewings and consultations.

3. Multi-Platform Access

   • Accessible through chat-widgets, text messaging, and phone calls, ensuring real-time communication with clients across various channels.

4. 24/7 Availability

   • Operates around the clock, ensuring no inquiries go unanswered, reducing missed opportunities, and enhancing client satisfaction.

Application Workflow

1. Initialization: Starts at the __start__ node.
2. Main Interaction Hub: The main_agent directs users to specific functionalities.
3. Property Inquiry Process: Routes users to search_criteria_agent and query_database.
4. Appointment Management: Guides users through the appointment_agent and appointment_tools.
5. Human-in-the-Loop: Ensures accuracy for complex tasks.
6. Conclusion of Interaction: Returns to main_agent and ends at __end__ node.

Backend Setup

1. Install dependencies

   pip install -r requirements.txt

2. Set up environment variables

   • Create a .env file in the root directory of the project.
   • Add the necessary environment variables.

3. Download and set up data

   • Download the real estate dataset from Kaggle and place it in the /data folder.
   • Run csv_to_sql.py to convert the CSV file into an SQL database.

4. Run the main application

   • Terminal: python main.py
   • Local server:
     1. Run the ngrok.exe file.
     2. Start the server: uvicorn app-retell.server:app --reload

Frontend: Realtor AI Chat Widget

Features

• Real-time chat interface with an AI assistant
• Responsive design for various screen sizes
• WebSocket integration for live communication
• Tailwind CSS for styling
• TypeScript for type safety

Frontend Setup

1. Install dependencies:

   npm install

   or

   yarn install

2. Create a .env file in the root directory and add any necessary environment variables.

Development

To run the development server:

npm run dev

or

yarn dev

Open http://localhost:5173 to view it in the browser.

Building for Production

To create a production build:

npm run build

or

yarn build

Usage

To use the ChatbotWidget in your React application:

import ChatbotWidget from './components/chatbot-widget'

function App() {
  return (
    <div className="App">
      <ChatbotWidget websiteId="your-website-id" />
    </div>
  )
}

Replace "your-website-id" with the appropriate identifier for your website.

Configuration

• The WebSocket connection URL is currently set to ws://127.0.0.1:8000/ws/${websiteId}/${threadId}. Update this in chatbot-widget.tsx if your backend is hosted elsewhere.
• Tailwind CSS configuration can be modified in tailwind.config.js.
• TypeScript configuration is split between tsconfig.json, tsconfig.app.json, and tsconfig.node.json.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request for either the backend or frontend components.

License

This project is licensed under the GNU General Public License v3.0. See the LICENSE file in the backend directory for details.