GlowFit AI

Beauty shoppers guess which products and looks will suit them. GlowFit scans the skin, renders three looks on your own face and scores them — then lets you change one thing at a time.

A hackathon build on Perfect Corp’s YouCam APIs. It turns skin analysis and virtual try-on into a decision flow: understand your skin, compare looks on your own face, adjust one variable, and leave with a look and a product list.

The problem

Try-on tools show you a look, but don’t help you decide between looks — or explain why one suits your skin better.

The approach

A five-step Decision Lab that combines skin metrics, face analysis and try-on renders into a comparable score, plus budget mode and single-variable previews.

What it includes

• Skin IntelligenceYouCam Skin AI score, eight metrics and a GlowMap.
• Decision LabThree looks rendered on your face, each with a match score.
• Change one thingSwap a single product with a before/after slider.
• Hairstyles & budget modeAI hairstyle try-on and budget-aware product picks.
• AssistantRule-based answers; an optional OpenAI-compatible model only rewrites the wording.

Role

• Solo build: product flow, UI, API integration and deployment.

How it is put together

1. SelfieBrowser capture
2. Next.js API routesKeys stay server-side
3. YouCam APIsSkin, face, makeup, hair
4. ScoringLooks compared on metrics
5. Decision UIZustand + IndexedDB cache

Simplified. No database or accounts — results are cached in the browser.