Venus — AI skincare diagnosis app
Live on the App Store and Google Play — 17 months of solo full-stack delivery.
- Client
- Beauty-tech startup (K-beauty)
- Timeline
- Apr 2025 – Aug 2026 · 17 months
- My role
- Sole engineer — mobile app, backend, AI pipeline, payments, analytics and releases.
- Continuous delivery
- 17 months
- One React Native codebase
- iOS + Android
- Versions shipped to the stores
- 1.4.x
Context
Venus is a K-beauty “glow up” program in an app: scan your face, get a plan from AI and Seoul skincare experts, follow a routine, and track progress day by day. I joined at the MVP stage and stayed on as the only engineer as it grew into a subscription product.
Challenge
Turn a selfie into a skin analysis people trust, wrap it in a daily habit loop, and make it a business — subscriptions, expert consultations and paid acquisition — on iOS and Android at once, without a team.
What I built
AI skin analysis
Guided selfie capture and upload, then a GPT-4V pipeline with engineered prompts and a structured output schema. It scores each concern — acne, pores, eye bags, redness, oiliness — and writes the detail cards users read.
Daily check-in and progress tracking
The habit loop that keeps users coming back: a daily check-in with a weekly score calendar, and before/after face comparison so users can see their skin change over time.
Personalized routines and product intelligence
A LangChain pipeline from diagnosis to a personalized routine, product scanning for ingredient fit, and K-beauty product discovery — with a cache layer in front of the model to keep inference cost down.
Subscriptions and expert consultations
Venus Pro subscriptions through RevenueCat and StoreKit, onboarding and paywall flows built from the Figma designs, and the intro consultation with Korean skincare experts.
Growth and attribution
Amplitude product analytics, Facebook and TikTok SDK attribution, Branch deep links and campaign tracking — so the client could see which channels actually brought paying users.
Backend, infrastructure and releases
FastAPI and Postgres on AWS (EC2 + RDS) with S3 image uploads, Docker and automated CI. Expo/React Native builds shipped continuously through TestFlight and Google Play, version after version.
Tech stack
- App
- React NativeExpoTypeScriptZustand
- Backend
- FastAPIPostgresS3
- AI
- GPT-4VLangChain
- Payments & growth
- RevenueCatStoreKitAmplitudeBranchFacebook SDKTikTok SDK
- Infra
- AWS EC2AWS RDSDockerCI
Result
- Live on both the App Store and Google Play
- Shipped release after release through version 1.4.x
- Subscription business in place: paywall, Venus Pro plan and expert consultations
- Full acquisition and attribution stack wired in for paid growth