Inspiration
Over 73% of online beauty shoppers return products because they looked different in real life. Existing try-on tools are siloed per brand, low quality, or require app downloads. We asked: what if every shopper had an AI stylist in their pocket one that knows their exact skin tone, face shape, and personal style?
AURA was born from that question.
What it does
AURA (AI Universal Retail & Beauty Assistant) is a web platform that:
- Analyzes your skin tone, texture, hydration, pores, dark spots, and more using Perfect Corp YouCam Skin Analysis API
- AR Makeup Try-On real-time virtual lipstick, eyeshadow, and foundation powered by YouCam AR APIs
- Fashion Try-On virtually wear outfits before buying
- Hair Color Try-On preview any hair color on yourself instantly
- AI Recommendations products ranked by your personal skin compatibility score
- Sustainability Scoring surfaces eco-friendly, cruelty-free, and vegan alternatives
- AI Beauty Advisor chat with a personalized AI expert powered by GPT-4o
How we built it
Frontend: Next.js 15, TypeScript, Tailwind CSS, Framer Motion
Backend: Next.js API Routes (Node.js), REST APIs
AI/ML: Perfect Corp YouCam API (Skin Analysis, AR Makeup, Hair Color, Fashion Try-On, Image Enhancement), OpenAI GPT-4o-mini for conversational recommendations
Database: Firebase Firestore for user profiles and skin history
Deployment: Vercel (frontend + API), Firebase (auth + storage)
Key innovation: A custom skin compatibility scoring algorithm that maps 14 skin parameters against product attributes to generate a personalized compatibility percentage for every product.
Challenges we ran into
- Integrating the YouCam V2 API required understanding the correct authentication pattern (direct
X-API-KEYheader, no OAuth token exchange) - Building a compatibility scoring engine that meaningfully maps skin analysis output to product recommendations
- Achieving fast AR rendering in a web browser without native app installation
- Managing image size limits for base64 encoded photo uploads through API routes
Accomplishments we're proud of
- Full AR try-on experience running in a web browser no app download required
- Skin analysis returning 14 parameters in under 2 seconds
- AI recommendation engine scoring every product against the user's unique beauty fingerprint
- Clean, luxury-grade UI that feels like a real consumer product
- Graceful fallback to mock data so the app never crashes without API keys
What we learned
Deep integration with the YouCam API ecosystem showed us how powerful AR becomes when embedded directly into the commerce decision flow not as a gimmick, but as the core tool for making purchase decisions. The skin analysis data creates a beauty fingerprint that makes every recommendation feel genuinely personal.
What's next for AURA
- Native iOS/Android app with persistent beauty profile
- Live video AR try-on (real-time camera, not just photos)
- B2B white-label SDK for retailers (Sephora, Ulta, ASOS)
- Dermatologist partnership for medical-grade skin analysis
- Social sharing "Share my AURA look"
- Skincare progress tracking with monthly photo comparisons
Log in or sign up for Devpost to join the conversation.