Kozma Hunor
Kozma Hunor

CineStream AI

An interactive movie trailer streaming application with a real-time AI chat assistant.

CineStream AI demo

The Challenge

The challenge was to create a modern, engaging, and centralized platform for watching movie trailers, enriched with contextual information from an AI, avoiding the scattered and dry experience of traditional movie databases and video platforms.

My Solution

I built a full-stack application with a Node.js backend and a vanilla JavaScript frontend. It fetches movie data from TMDB, streams trailers from an S3 bucket, and integrates a Gemini-powered AI chat that provides witty, contextual commentary on the movies.

Tech Stack

  • Node.js: Used for the server-side environment, handling API routing and serving the frontend.
  • Express: A minimal and flexible Node.js web application framework that provides a robust set of features for web and mobile applications.
  • Gemini AI: The generative AI model from Google, used to power the contextual chat feature and provide witty commentary.

Technical Deep-Dive

The core of the project is the real-time AI chat integration. The backend constructs a dynamic 'system prompt' for the Gemini AI, feeding it factual data about the movie (title, overview, cast, rating). This grounds the AI, allowing it to act as a knowledgeable and witty film critic that can answer user questions in context, directly within the screening room.