SkinTwin AI

A private, photo-based skincare progress companion that helps people track cosmetic skin insights week by week.

Comment

What is SkinTwin AI?

SkinTwin AI is a photo-based skincare progress companion designed to help people understand how their skin changes over time. Instead of treating skin analysis as a one-time snapshot, SkinTwin organizes check-ins into a private, week-by-week timeline.

The problem

People often struggle to tell whether their skincare routine is making a difference. They rely on memory, inconsistent photos, and scattered notes, making it difficult to compare changes over time.

Our solution

SkinTwin gives users a structured way to capture a close-up photo, keep check-ins organized, review skin-related indicators, and revisit their progress over time.

Key features shown in the current demo

  • A landing page explaining the product and its privacy-first approach.
  • A guided photo-capture and photo-quality guide.
  • A weekly check-in timeline.
  • A check-in detail screen with a score and individual skin indicators.
  • Comparison, routine, and product-catalogue areas in the app navigation.
  • Clear cosmetic-guidance positioning: SkinTwin is not a medical diagnosis tool.

How it works

  1. Capture or upload a consistent close-up photo.
  2. Store the photo privately under the user's account.
  3. Review the check-in and its skin-related indicators.
  4. Return for future check-ins and compare progress over time.

AI/API transparency

The current demo recording visibly labels the displayed score as a Demo result. This is sample/demo content, not proof of a live YouCam Skin AI API response. The live YouCam API integration and a genuine end-to-end API result must be verified before this project is represented as fully integrated for judging. No synthetic score is presented here as a verified YouCam result.

Privacy and safety

SkinTwin is positioned for cosmetic tracking and general skincare guidance, not diagnosis or treatment. The product experience emphasizes private photo handling and clear user-facing privacy information.

Demo and source

Current status

This Devpost entry is a draft. Feature claims above are limited to what is visible in the supplied demo screenshots and video; backend security, data isolation, deletion behavior, and live YouCam API calls still require end-to-end verification.

Built With

Share this project:

Updates

posted an update —

Project Name: Skin Twin AI

Elevator Pitch / Tagline (1 line): A private digital skin twin that tracks your weekly skin changes with real YouCam AI analysis and age-aware routine insights.

Inspiration: Most skincare apps rely on guesswork, generic filters, or push expensive products without proof. We wanted to build a true "digital skin twin" — a private, clinical-grade companion where users take a weekly photo, receive genuine YouCam AI skin metrics, and clearly see whether their routine is actually improving their skin over time, without misleading claims or privacy compromises.

What It Does: Weekly Skin AI Check-ins: Takes a verified face photo and runs it through YouCam Skin AI, returning an overall score plus 11 distinct metrics (Clarity, Pores, Texture, Fine Lines, Firmness, Hydration, Redness, Oil Control, Dark Circles, Radiance, Even Tone). Progress & Before/After Comparison: An interactive trend chart and side-by-side comparison tracking how metrics shift between check-ins. Age-Tailored & Safety-Checked Guidance: Adapts tips and active ingredients based on age bands (e.g., strictly filtering out harsh actives like retinol for younger skin). AI Skin Explanations: Non-diagnostic, plain-language insights generated via AI to help users understand what their scores mean. Privacy & Photo Erasure: Private encrypted storage, 60-second signed photo URLs, and guaranteed complete photo-byte deletion when an account or scan is deleted. Community Showcase: An opt-in only public slider where users can voluntarily share real skin twin journeys. Global Accessibility: Full support for 9 languages, including native right-to-left (RTL) Urdu. How We Built It: Built on a modern full-stack architecture using TanStack Start (React 19), Tailwind CSS, and Lovable Cloud with Row-Level Security (RLS). Server-side functions securely communicate with the Perfect Corp YouCam Skin AI API (featuring automatic credential fallback) and the Lovable AI Gateway for personalized analysis.

Challenges We Ran Into: Ensuring strict privacy compliance while handling biometric imagery was our top priority. We architected a private storage pipeline with tight RLS policies, short-lived signed URLs, and an idempotent photo-byte deletion queue. We also engineered a resilient API gateway fallback so scan tasks never fail if an individual quota is exhausted.

Accomplishments That We're Proud Of: Delivering a live, working integration with YouCam Skin AI that delivers real, actionable skin metrics in seconds, paired with a complete bilingual UI and a non-diagnostic medical safety standard.

Built With (Tags): youcam-api, skin-ai, react-19, tanstack-start, tailwind-css, typescript, lovable-cloud, ai-gateway

Live Demo URL: https://yur-skin-story.lovable.app

Log in or sign up for Devpost to join the conversation.

posted an update —

My Contribution: "I led the concept, end-to-end product design, and implementation of Skin Twin AI. I integrated the YouCam Skin AI API to process real face check-ins across 11 dermatological metrics, built the privacy-first architecture with strict data erasure, designed the age-adaptive routine and ingredient guidance system, and created full multi-language support (including Urdu RTL)."

Log in or sign up for Devpost to join the conversation.

Submission history