Inspiration
My family runs a secondhand clothing store, which made me think a lot about how little we actually know about what we wear. I wanted to build something like Spotify for your closet.
What it does
Winnie is a live AI wardrobe assistant. You can talk to it to browse your closet, add clothes seamlessly, get outfit and shopping suggestions, virtually try on pieces, and save things to your wishlist.
How we built it
Winnie is built with React Native and Expo, with Supabase for data and RevenueCat powering Winnie Pro subscriptions and paywalls.
Challenges we ran into
Technically, this was probably the hardest project I've ever worked on. Building a live assistant meant constantly balancing deterministic behavior with the spontaneity of an AI conversation. I also learned a huge amount about phone hardware, audio and video encoding, and real-time media.
Accomplishments that we're proud of
Building a working conversational wardrobe assistant that can actually interact with your closet, not just chat about it.
What we learned
I learned a lot about building real-time AI experiences, especially how much complexity sits behind making something feel simple and natural to the user.
What's next for Winnie
More personalized insights, smarter outfit recommendations, and an even more capable wardrobe assistant.


Log in or sign up for Devpost to join the conversation.