work / f-try · Prototype
F-Try
Photograph your wardrobe once, get outfits with reasons.
Two problems, one app. Most people cannot recall what is actually in their wardrobe, so the same few items get worn and the rest sits idle: catalogue it properly and the app can surface combinations you would never have reached for. The second is buying online, where whatever the size chart says, you still have no idea how it looks on you or whether it works with what you already own. The AI mirror puts the piece you have not bought onto you, next to the clothes you have.
A full wardrobe and the daily feeling of nothing to wear.
AI reads each garment from one photo, then composes outfits from colour theory, the Korean rule of three and 12-season personal colour analysis.
- Supabase
- Claude vision
- PWA
Biometric privacy is a design input, not an afterthought.
A pure-frontend replica with mock data: nothing is saved, nothing leaves the page. The real app is richer; this is the feel of it.
loading the demo…
The thinking
user · pains · gap · tools
People with a full wardrobe and daily outfit paralysis, who want a stylist’s logic without a stylist’s bill.
Target user
- Deciding what to wear costs willpower every morning
- Clothes get forgotten because no one can see their whole wardrobe at once
- Style advice online is generic, not about your actual clothes
Pain points
Photograph each garment once and get outfits composed from real colour science against your own wardrobe, with biometric data treated as sensitive from day one.
Gap in the market
- Claude vision (garment reading)
- Claude Code (build)
- Supabase (RLS)
AI + tooling
- 12-season personal colour analysis
- Colour theory pairing + the Korean rule of three
- Choice architecture: three good outfits beat an infinite wardrobe scroll
- Privacy by design under the Australian Privacy Act (biometric data is sensitive)
Frameworks
- TBD: prototype stage, outfit acceptance rate once piloted
Success metrics
A JARVIS-style assistant that ran my day, by voice.