Inspiration
Ever needed to shop but felt overwhelmed by endless options? Or maybe your closet is full, but you still rotate the same three outfits?
Most people figure out what looks good on them through trial and error—and a lot of wasted money. Fashionista is your AI stylist that changes that.
What it does
By analyzing your color palette, body type, and style inspirations (collected through just 10 quick swipes), Fashionista builds a personalized style profile. From there, it suggests full outfit combinations through a natural, conversational interface—like talking to a stylist who actually gets you.
How we built it
We used modern tools like Cursor and V0 to build the frontend, with a Python backend that fetches live product data from sites like H&M. At the core is a built-in LLM-powered AI agent that interfaces with you directly—like chatting with your own personal stylist.
Challenges we ran into
One of the biggest challenges we faced was sourcing and processing fashion catalog data. Every online retailer structures their product data differently—varying formats, inconsistent metadata, and non-standardized category tagging made it difficult to build a unified catalog. Processing and normalizing these vectors into something usable for our AI agent required a lot of custom parsing, cleaning, and ongoing refinement.
Accomplishments that we're proud of & What we learned
Building this from scratch in just 6 hours was both intense and technically challenging. From stitching together a frontend with Cursor and V0, to building a Python backend that interfaces with an LLM agent and fetches product data across inconsistent catalogs—it pushed us to move fast, make trade-offs, and stay laser-focused on the core experience.
What's next for Fashionista
We’re turning Fashionista into more than just a stylist—it’ll be your wardrobe, shopping assistant, and style profile in one. Soon, users will be able to save outfits, shop seamlessly, and share their personal lookbook with the world.
Built With
- react
- tailwind
- typescript

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