Discover your next favorite movie with AI-powered summaries and seamless trailers.

We wanted to revolutionize how people decide what to watch next. Instead of endlessly scrolling through pages and vague descriptions, Cinemaker delivers an immersive, instant-viewing taste testing experience by combining YouTube trailers with AI-generated summaries.

Cinemaker brings you a continuous stream of curated movie and TV trailers, all within your browser. Features include:

• Instant trailers with no reloads
• 👍 Like / 👎 Dislike system to refine personal taste
• ⏭ Skip to the next title at any time
• AI-generated summaries of plot, characters, and genre

All in real time—so you spend less time searching, and more time watching.

Frontend:

• Built with React (functional components, hooks like useEffect and useState)
• Styled using Tailwind CSS for a modern, responsive UI

Backend:

• Trailers were integrated with TMDB's API
• Flask server with endpoints:
  • /api/skip for skipping to the next trailer
  • /api/describe for AI-generated summaries
• Integrated with Google GenAI's Gemini API

Data Flow:

1. User skips or selects a trailer
2. A request gets sent to TMDB for the next movie
3. React sends a request to Flask
4. Flask queries Gemini for a summary
5. React receives and updates UI in real-time

Deployment Tools:

• Flask-CORS for smooth client-server communication
• dotenv for secure key management
• Local dev on localhost:3000 (React) + localhost:5000 (Flask)

• React
• Tailwind CSS
• Vite
• Flask
• Python
• Google GenAI – Gemini
• TMDB API
• HTML / CSS

• CORS Configuration: Tuning headers for clean React-Flask communication
• Async State Management: Handling rapid user input and API latency
• Prompt Engineering: Iterating on instructions for meaningful summaries from Gemini

• Seamless UX with no reloads or interruptions
• First-time integration of APIs
• Implemented Like/Dislike logic to adapt trailer suggestions

• Mastered React hooks and dependency arrays
• Gained experience in Flask REST API design
• Navigated real-world CORS and async fetch patterns
• Practiced prompt engineering for better AI output

• User Accounts: Save history, preferences, and resume where you left off
• Recommendation Engine: Smarter AI-backed trailer suggestions
• Mobile Version: React Native app for on-the-go discovery
• Advanced Filtering: Search, tags, and custom playlists