Cinemagine | Devpost

Inspiration

The Citrus Hack 2025 Theme of Lights, Code, Action! (Cinema) inspired us to make a mood-based movie recommendation platform that is immersive, interactive, and personal.

What it does

Our mood-based movie recommendation platform asks the user for their movie mood and the genre of the movie they want. Based on these inputs, our AI-ML model queries the TMDb database and fetches out the top movie picks relevant to the search.

How we built it

We used Next.js and Tailwind CSS for our frontend (UI /UX). We used Python, Flask API for our backend. We incorporated the TMDb dataset and the Gemini API for our requests.

Challenges we ran into

We had trouble linking the APIs. Executing the machine learning model perfectly was challenging. The integration of the frontend and the backend took significant amount of time and effort.

Accomplishments that we're proud of

We are proud of getting the API to work and the frontend to match our prototype design pretty close.

What we learned

We learned Convex and Clerk for our authentication purposes. We learnt how to use randomness for our database query to get random movie picks for various inputs.

What's next for Cinemagine

We are looking to implement the feature of clicking on the movie recommendations to get further details and also links to the platforms where users can watch it.