Glow Up AI

The brand neutral beauty advisor: one selfie returns a scored skin report with exact fixes for every skin tone. No brand money, no bias. Built solo, live as web app and Android APK.

Comment

Beauty advice online is either generic or bought. Routines ignore skin tone, climate, and budget, and product picks follow sponsor money. Mainstream beauty AI is trained mostly on light skin, so it misreads the faces of most of the world. I wanted an advisor that earns nothing from the products it judges, so it can say the one thing no brand can say: skip this one.

What it does

Glow Up AI is a brand neutral beauty advisor. One selfie returns a scored skin report: every issue graded, explained, and matched to its exact fix across routines, DIY face packs, and nutrition. It also runs hair fall scans with severity tracking, an ingredient scanner that judges products against your skin before you buy and finds cheaper dupes, color and wardrobe palettes, a bridal studio, 90 day event ready plans, cycle synced skincare, and an always on AI advisor chat.

How I built it

Solo, no team, no funding. Nine months of dermatology and cosmetics research first, then the build: Base44 for the web app and Android APK, Google Gemini vision for skin and hair analysis, and a recommendation engine that matches every output to the user's skin tone, climate, cycle, and budget. Every feature shipped from real user feedback, including hair 2.0 and the product scanner.

Challenges I ran into

Building alone means every role is mine: research, design, code, support, marketing. API costs forced hard choices about what stays free. And training data bias meant deeper skin tones needed extra care, so I tested and tuned until the scans tell the truth on every face.

What I learned

Trust is the product. Users stay when the advisor has no brand to protect. Nineteen strangers signed up with zero paid marketing, and one founder written post reached 56,000 organic impressions, which taught me that honesty scales cheaper than ads.

What's next

Skin age tracking with a shareable progress curve, white label licensing for clinics, and entry into the US and Gulf markets where deeper skin tones are ignored the loudest.

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Glow Up AI

Built With

Share this project:

Updates

Submission history