Alyza - AI-Powered Conflict Resolution

Alyza is a revolutionary AI-powered conflict resolution application that runs completely on-device using Google Gemma 3N. This privacy-first solution provides intelligent, empathetic guidance for resolving conflicts in personal relationships, workplace situations, and professional settings.

🌟 Features

• 100% On-Device AI: All processing happens locally on your device using Google Gemma 3N
• Zero Latency: No network requests mean instant responses
• Complete Privacy: Your sensitive conflict data never leaves your device
• Intelligent Analysis: Real-time sentiment analysis and underlying need identification
• Empathetic Responses: AI-powered guidance that helps navigate difficult conversations
• Cross-Platform: Works on Android, iOS, and Web
• Beautiful UI: Modern, intuitive interface with interactive ocean animations

🏆 Google Gemma 3N Hackathon Entry

This project was developed for the Google Gemma 3N Hackathon, showcasing the revolutionary capabilities of on-device AI for sensitive applications like conflict resolution.

🚀 Quick Start

Prerequisites

• Node.js 18+ and npm
• Android Studio (for Android development)
• Xcode (for iOS development, macOS only)
• Expo CLI: npm install -g @expo/cli

Installation

1. Clone the repository

   git clone https://github.com/yourusername/alyza.git
   cd alyza

2. Install dependencies

   npm install

3. Install Expo CLI globally (if not already installed)

   npm install -g @expo/cli

Setting Up the AI Model

For Android Development

1. Download the Gemma 3N model

   # Create models directory
   mkdir -p android/app/src/main/assets/models
   
   # Download the model (you'll need to obtain this from Google's model hub)
   # Place the model file in android/app/src/main/assets/models/

2. Push model to device using ADB

   # Connect your device and enable USB debugging
   adb devices
   
   # Push the model to the device
   adb push android/app/src/main/assets/models/gemma-3n-model.bin /sdcard/Android/data/com.omijimo.realalyza/files/models/

3. Verify model installation

   adb shell ls /sdcard/Android/data/com.omijimo.realalyza/files/models/

For iOS Development

1. Add model to Xcode project

   • Open ios/realalyza.xcworkspace in Xcode
   • Drag your Gemma 3N model file into the project
   • Ensure it's added to the app target

2. Update model path in code

   • The model will be bundled with the app
   • Update the model path in your native module if needed

Running the Application

Development Mode

1. Start the development server

   npx expo start

2. Choose your platform

   • Press a for Android
   • Press i for iOS
   • Press w for web
   • Scan QR code with Expo Go app

Production Build

1. Build for Android

   npx expo build:android

2. Build for iOS

   npx expo build:ios

3. Build for Web

   npx expo build:web

🏗️ Project Structure

alyza/
├── app/                    # Main React Native app
│   ├── _layout.tsx        # Root layout
│   ├── index.tsx          # Main app screen
│   └── +not-found.tsx     # 404 page
├── components/            # React components
│   ├── Chat.tsx          # Chat interface
│   ├── OceanScene.tsx    # 3D ocean background
│   ├── WebChat.tsx       # Web chat component
│   └── ...
├── android/              # Android native code
│   └── app/src/main/java/com/omijimo/realalyza/
│       ├── LlmModule.kt  # Native LLM integration
│       └── ...
├── ios/                  # iOS native code
├── website/              # Web version
│   ├── app/             # Next.js pages
│   ├── components/      # Web components
│   └── ...
└── assets/              # Images, fonts, etc.

🔧 Technical Architecture

On-Device AI Processing

• Model: Google Gemma 3N (2B parameters)
• Processing: 100% local inference
• Response Time: < 200ms
• Memory Usage: Optimized for mobile devices
• Privacy: Zero data transmission

Frontend Technologies

• Mobile: React Native with Expo
• Web: Next.js 14 with TypeScript
• 3D Graphics: Three.js with React Three Fiber
• Styling: Tailwind CSS
• Animations: Custom CSS animations and Framer Motion

Backend Integration

• API: RESTful endpoints for model inference
• Authentication: Optional user accounts
• Data Storage: Local SQLite database
• Sync: Optional cloud sync (user-controlled)

📱 Usage

Getting Started

1. Launch the app and grant necessary permissions
2. Tap "Try Alyza Now" to start a conversation
3. Describe your conflict in natural language
4. Receive AI guidance with sentiment analysis and empathetic responses
5. Follow the suggestions to resolve your conflict

Features

• Real-time Analysis: Get instant feedback on emotional states
• Need Identification: Understand underlying concerns
• Guided Resolution: Step-by-step conflict resolution strategies
• Conversation History: Review past interactions
• Export Options: Save important conversations

🤝 Contributing

We welcome contributions! Please see our Contributing Guidelines for details.

Development Setup

1. Fork the repository
2. Create a feature branch

   git checkout -b feature/amazing-feature

3. Make your changes
4. Test thoroughly
5. Submit a pull request

Code Style

• Follow the existing code style
• Use TypeScript for all new code
• Add tests for new features
• Update documentation as needed

📄 License

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

🙏 Acknowledgments

• Google for the Gemma 3N model and hackathon opportunity
• Expo for the excellent development platform
• Three.js for the beautiful 3D graphics
• React Native community for the amazing ecosystem

📞 Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Email: support@alyza.ai

🔮 Roadmap

• Multi-language support
• Voice input/output
• Group conflict resolution
• Professional mediation tools
• API for third-party integrations
• Advanced analytics dashboard

---

Made with ❤️ for the Google Gemma 3N Hackathon