Dealership Accelerator

An AI-powered platform for automotive lead engagement, qualification, and appointment booking.

Features

• Lead intake and management
• AI-powered lead qualification and response
• Appointment scheduling
• Admin dashboard with analytics
• Cold lead re-engagement
• SMS and email communication

Tech Stack

• Frontend: React.js with TypeScript
• Backend: Node.js + Express
• Database: PostgreSQL
• AI Integration: OpenAI/Claude/LM Studio
• Messaging: Twilio (SMS) & SendGrid (Email)

Project Structure

dealership-accelerator/
├── client/                 # React frontend
├── server/                 # Node.js + Express backend
├── database/              # Database migrations and seeds
└── docs/                  # Additional documentation

Prerequisites

• Node.js >= 18
• PostgreSQL >= 14
• npm or yarn
• Docker Desktop
• ngrok
• LM Studio (for local AI model)

Getting Started

1. Clone the repository

2. Set up environment variables (see .env.example files)

3. Start Docker Desktop and run the project containers:

   # From project root
   docker-compose up -d

4. Install dependencies:

   # Install frontend dependencies
   cd client
   npm install
   
   # Install backend dependencies
   cd ../server
   npm install

5. Start ngrok for Twilio webhook:

   # Start ngrok on port 8000 (or your backend port)
   ngrok http 8000

6. Configure Twilio webhook:

   • Copy your ngrok URL (e.g., https://abc123.ngrok.io)
   • Go to Twilio Console > Phone Numbers > Your Number
   • Under "Messaging", set the webhook URL for when "A Message Comes In" to: your-ngrok-url/api/ai/webhook/sms
   • Save the changes

7. Start LM Studio:

   • Open LM Studio application
   • Start the local server (default port: 1234)
   • Load your preferred model (e.g., llama-2-7b-chat)

8. Start the development servers:

   # Start frontend (from client directory)
   npm run dev
   
   # Start backend (from server directory)
   npm run dev

9. Access the application:

   • Frontend: http://localhost:3000
   • Backend API: http://localhost:8000

Important Development Notes

1. Docker Setup:

   • Ensure Docker Desktop is running before starting the project
   • The docker-compose file includes PostgreSQL and other required services
   • Monitor container logs through Docker Desktop for debugging

2. Twilio Webhook:

   • Keep ngrok running to maintain the webhook URL
   • If ngrok restarts, update the Twilio webhook URL with the new ngrok URL
   • Test webhook connectivity through Twilio's console

3. LM Studio:

   • Ensure LM Studio is running before starting the backend
   • The AI service will fall back to simulated responses if LM Studio is unavailable
   • Monitor LM Studio logs for any model loading or inference issues

Environment Variables

See .env.example files in both client/ and server/ directories for required environment variables.

License

MIT