Recipe Generator App

🍽️ AI-powered recipe generation app that creates personalized recipes from photos of ingredients using Google Gemini AI.

📱 Features

• 📷 Smart Ingredient Recognition: Take photos of ingredients and get AI-powered identification
• 🤖 AI Recipe Generation: Generate 5 personalized recipes instantly
• 📖 Detailed Recipes: Complete recipes with ingredients, instructions, cooking time, and difficulty
• 🔄 Recipe Refresh: Generate new recipes using the same ingredients
• 📱 Cross-Platform: React Native app for iOS and Android
• ⚡ Fast API: Optimized FastAPI backend with Google Gemini integration

🏗 Architecture

This is a full-stack application with two main components:

Frontend (/client)

• React Native with Expo
• TypeScript for type safety
• Expo Camera for photo capture
• Native mobile app for iOS and Android

Backend (/core)

• FastAPI Python web framework
• Google Gemini AI for vision and text generation
• Pydantic for data validation
• RESTful API with automatic documentation

🚀 Quick Start

Prerequisites

• Node.js 16+
• Python 3.11+
• Google API key (Get one here)
• iOS Simulator or Android Emulator (for mobile development)

1. Backend Setup

cd core

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
cp .env.example .env
# Edit .env and add your GOOGLE_API_KEY

# Run backend server
python run.py

The API will be available at http://localhost:8000

2. Frontend Setup

cd client

# Install dependencies
npm install

# Start development server
npx expo start

# Run on iOS/Android
npx expo run:ios
npx expo run:android

📡 API Endpoints

Health Check

GET http://localhost:8000/health

Generate Recipes

POST http://localhost:8000/api/generate-recipes-json
Content-Type: application/json

{
  "image": "base64_encoded_image_data",
  "exclude_recipe_ids": ["recipe_1"]
}

Response:

{
  "recipes": [
    {
      "id": "recipe_1",
      "title": "Sesame Ginger Tofu Stir-Fry",
      "summary": "Quick and savory stir-fry...",
      "ingredients": ["14 oz tofu", "8 oz mushrooms"],
      "instructions": ["Heat oil", "Add ingredients"],
      "cookingTime": 20,
      "servings": 2,
      "difficulty": "Easy",
      "tags": ["Asian", "Quick"]
    }
  ],
  "identifiedIngredients": ["tofu", "mushrooms"]
}

🔧 Configuration

Backend Configuration

Edit /core/.env:

GOOGLE_API_KEY=your_google_api_key_here
LLM_MODEL=gemini-3-flash-preview
API_HOST=0.0.0.0
API_PORT=8000
DEBUG=True
ALLOWED_ORIGINS=http://localhost:3000,exp://192.168.1.100:8081

Frontend Configuration

Edit /client/services/config.ts:

export const apiConfig = {
  baseURL: 'http://localhost:8000', // Backend URL
  timeout: 120000, // 2 minutes for image processing
};

📱 Mobile App Features

Camera Integration

• Built-in camera interface
• Gallery photo selection
• Automatic image optimization
• Base64 encoding for API transmission

Recipe Display

• Grid layout with recipe cards
• Detailed recipe views with ingredients and instructions
• Cooking time, servings, and difficulty indicators
• Tag-based categorization

Navigation

• File-based routing with Expo Router
• Smooth transitions between screens
• Back button and navigation controls

🤖 AI Integration

Ingredient Identification

• Gemini Vision API for image analysis
• Confidence scoring for ingredient recognition
• Filtering of non-food items
• Specific ingredient naming (e.g., "cremini mushrooms" not "mushrooms")

Recipe Generation

• Exactly 5 recipes per request
• Difficulty variety (2 Easy, 2 Medium, 1 Hard)
• Diverse cuisines (Italian, Asian, American, etc.)
• Complete recipes with measured ingredients
• Step-by-step cooking instructions

Prompt Engineering

• Optimized prompts for token efficiency
• Structured JSON outputs
• Example-driven responses
• Error handling with fallback recipes

🎨 User Interface

Design System

• Clean, modern interface
• Consistent typography and spacing
• Platform-specific styling (iOS/Android)
• Responsive layouts for different screen sizes

User Experience

• Intuitive camera-first workflow
• Loading states with progress indicators
• Error handling with helpful messages
• Smooth animations and transitions

🔒 Security Features

Backend Security

• Input validation with Pydantic
• CORS configuration for mobile apps
• Environment variable management
• Error sanitization

Data Privacy

• No recipe data storage
• Temporary image processing
• No user tracking
• API key protection

📊 Performance

Optimizations

• ~30% token reduction in AI prompts
• Base64 image encoding for mobile compatibility
• Efficient JSON API design
• Optimized image processing (RGBA→RGB conversion)

Response Times

• Ingredient identification: ~2-5 seconds
• Recipe generation: ~5-15 seconds
• Total processing: ~10-20 seconds
• API timeout: 120 seconds

🧪 Testing

Backend Testing

cd core

# Health check
curl http://localhost:8000/health

# API documentation
open http://localhost:8000/docs

Frontend Testing

• Test on iOS Simulator and Android Emulator
• Real device testing for camera functionality
• Network connectivity testing
• Error scenario testing

🚀 Deployment

Backend Deployment

• Docker containerization support
• Gunicorn for production WSGI
• Environment-specific configuration
• Health monitoring endpoints

Frontend Deployment

• Expo EAS Build for app stores
• Over-the-air updates
• Platform-specific builds
• Development and production variants

📚 Documentation

• Frontend: /client/README.md - React Native app documentation
• Backend: /core/README.md - FastAPI server documentation
• API Docs: http://localhost:8000/docs - Interactive API documentation

🐛 Troubleshooting

Common Issues

"Model not found" Error

• Check Google API key in /core/.env
• Verify model name: gemini-3-flash-preview
• Ensure API key has Gemini access

Camera Not Working

• Check device permissions for camera access
• Test on physical device (camera doesn't work in all simulators)
• Verify Expo Camera installation

API Connection Failed

• Ensure backend is running on localhost:8000
• Check CORS configuration for mobile app URLs
• Verify network connectivity

Image Processing Errors

• Check image format (JPEG, PNG supported)
• Verify image size (under 10MB)
• Test with different images

📈 Roadmap

• User accounts and recipe saving
• Recipe sharing and social features
• Dietary restrictions and preferences
• Nutritional information
• Shopping list generation
• Recipe rating and reviews
• Offline recipe storage
• Multi-language support

🤝 Contributing

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

📄 License

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

🙏 Acknowledgments

• Google Gemini AI for vision and text generation capabilities
• Expo for React Native development tools
• FastAPI for the excellent Python web framework
• React Native community for mobile development resources

---

Built with ❤️ using React Native, FastAPI, and Google Gemini AI