Mobile Shopping Chat Agent

An AI-powered conversational agent that helps users find and compare mobile phones based on their preferences.

• Natural language understanding for mobile phone queries
• Intent classification to understand user requests
• Mobile phone database with filtering capabilities
• Conversational interface for easy interaction
• RESTful API for integration with frontend applications

1. Clone the repository
2. Install dependencies:

   pip install -r requirements.txt

3. Create a .env file based on .env.example and fill in your configuration
4. Run the application:

   python app.py

• Python 3.8+
• Git
• Supabase
• Gemini API Key

1. Prepare your code

   • Make sure all your changes are committed to a Git repository
   • Push your code to GitHub, GitLab, or Bitbucket

2. Deploy to Render

   • Sign up at Render if you haven't already
   • Click "New" and select "Web Service"
   • Connect your repository
   • Configure the deployment:
     • Name: mobile-shopping-agent
     • Region: Choose the one closest to your users
     • Branch: main or your preferred branch
     • Build Command: pip install -r requirements.txt
   • Add environment variables from your .env file
   • Click "Create Web Service"

3. Set up custom domain (optional)

   • Go to your service settings
   • Click "Add Custom Domain" and follow the instructions

Create a .env file in the root directory with the following variables:

FLASK_ENV=production
SECRET_KEY=your-secret-key-here

.
├── app.py                # Main application file
├── requirements.txt      # Python dependencies
├── .env.example         # Example environment variables
├── models/              # ML models for intent classification
├── utils/               # Utility functions
├──────── agent.py              # Chat agent implementation
├──────── data_loader.py        # Data loading and querying
└── data/                # Sample mobile phone dataset

POST /api/chat
{
    "message": "I'm looking for an iPhone under $1000"
}

{
    "response": "Here are some iPhones under $1000...",
    "suggestions": ["Show more", "Filter by storage", "Compare models"],
    "products": [...]
}

• Application not starting: Check the logs in your hosting provider's dashboard
• Database connection issues: Verify your MONGODB_URI is correct and the database is accessible
• Dependency issues: Make sure all dependencies are installed with the correct versions