Case study
WearAgain
A mobile-first wardrobe app that uses AI to identify clothing from outfit photos and build a searchable catalog from real wear history.
Purpose
A mobile-first wardrobe app that uses AI to identify clothing from outfit photos and build a searchable catalog from real wear history.
What this is
WearAgain is a wardrobe app. You upload a photo of what you are wearing, the AI identifies the pieces, and a catalog builds itself from your real outfit history rather than a manual inventory.
The AI piece
Clothing extraction runs server-side on Claude Haiku. When you upload an outfit photo, the image goes to an API route that calls Claude, pulls out the identified items with their category and attributes, and returns structured results for review before anything lands in the catalog. The client never touches the API key.
Why this is in the portfolio
This is practical consumer-level AI product design. It takes friction out of a real task, handles multi-user data isolation correctly, includes a demo mode that runs without backend credentials, and makes deliberate choices about where AI runs and where it does not. The clothing domain is personal and the design patterns are portable.
Related