Inspiration
As college students, we find it difficult to put together stylish outfits that match weather conditions that constantly change.
What it does
We created an app that catalogs your entire wardrobe and generates personalized outfits that match the weather while keeping you looking stylish.
How we built it
We designed a React frontend using Vite and Tailwind (hosted on Vercel), supported by a Node.js backend (hosted on Railway). Our stack uses Supabase for authentication and database management, the OpenWeather API for real-time data, and Modal to run inference through Gemini.
Challenges we ran into
Our biggest hurdles included integrating Gemini API calls through Modal, managing our API credits during heavy testing, and orchestrating the deployment between Vercel and Railway.
Accomplishments that we're proud of
We are proud of building a functional web application that generates truly wearable outfits. The system is dynamic; it learns from user preferences and feedback to improve future suggestions.
What we learned
We gained hands-on experience with full-stack architecture and the integration of containerized AI tools like Modal. We also improved our team workflow by effectively splitting development tasks and learning how to communicate technical ideas clearly.
What's next for Fitted
We plan to transition Fitted into a mobile application featuring a daily notification system that sends you a generated outfit every morning. We are also looking into monetization strategies to scale the app and support more users without hitting API budget constraints.
Log in or sign up for Devpost to join the conversation.