MediaVoyager is an intelligent recommendation engine that provides personalized movie and TV show suggestions using artificial intelligence. The system analyzes user preferences, favorite content, and watch history to deliver tailored recommendations that match individual tastes.

• 🤖 AI-Powered Recommendations - Leverages Google Gemini AI to understand user taste and generate personalized recommendations
• 🎬 Movie Recommendations - Get movie suggestions based on your favorite films
• 📺 TV Show Recommendations - Discover new TV series tailored to your preferences
• 👤 User Management - Secure user authentication with Google Sign-In support
• 📋 Watchlist Management - Maintain separate watchlists for movies and TV shows
• 📊 Watch History Tracking - Track what you've watched to avoid duplicate recommendations
• ⭐ IMDb Ratings - Displays IMDb ratings for movies
• ⚡ Smart Caching - Efficient caching system for movies and TV shows data
• 🔒 Secure Authentication - JWT-based authentication and authorization
• 🚦 Rate Limiting - Built-in rate limiting to ensure service stability

MediaVoyager/
├── Clients/                    # External API clients
│   └── GeminiRecommendationClient.cs
├── Controllers/                # API Controllers
├── Entities/                   # Data entities
├── Handlers/                   # Request handlers
│   └── SignInHandler.cs
├── Middleware/                 # Custom middleware
├── Models/                     # Request/Response models
├── Repositories/               # Data access layer
│   ├── UserRepository.cs
│   ├── UserMoviesRepository.cs
│   ├── UserTvRepository.cs
│   ├── MovieCacheRepository.cs
│   └── TvShowCacheRepository. cs
├── Services/                   # Business logic
│   ├── MediaRecommendationService.cs
│   ├── UserMediaService.cs
│   └── TmdbCacheService.cs
└── Program.cs                  # Application entry point

MediaVoyager uses Google's Gemini AI to provide intelligent recommendations:

1. Taste Analysis - The AI analyzes your list of favorite movies/TV shows to understand your preferences
2. Context Awareness - Watch history is considered to avoid recommending content you've already seen
3. Personalized Output - The AI generates a single, highly relevant recommendation based on your unique taste profile
4. Adaptive Temperature - If initial recommendations don't yield results, the system retries with adjusted parameters for better results

User Favorites → Gemini AI Analysis → Taste Understanding → Personalized Recommendation

The core service that orchestrates the recommendation flow:

• Fetches user preferences and watch history
• Communicates with the Gemini AI client
• Validates and processes AI responses
• Returns enriched movie/TV show data from TMDb

Handles all communication with Google's Gemini API:

• Rate-limited to ensure API compliance
• Supports configurable temperature for response variation
• Provides separate recommendation logic for movies and TV shows

Integrated with The Movie Database (TMDb) for rich media information:

• Movie details, posters, and metadata
• TV show information and episode details
• Cast and crew information
• Trailers and videos

The application is deployed on Azure Container Apps using source-based deployment with buildpacks:

• Health Endpoint: /health
• Graceful Shutdown: 30-second timeout
• Logging: Console logging optimized for containers

For detailed deployment instructions, refer to [DEPLOYMENT.md](DEPLOYMENT. md).

The application requires the following secrets (managed via Azure Key Vault or similar):

• NewHorizonLib - Shared library for common services
• TMDbLib - . NET wrapper for TMDb API
• Newtonsoft.Json - JSON serialization

Contributions are welcome! Please feel free to submit a Pull Request.

This project is open source and available under the MIT License.