~/work/stylesyncs
StyleSyncs
AI-powered virtual try-on for fashion
A virtual try-on platform I co-founded that shows shoppers how clothes look on them. I built its AI features, including static try-on and digital avatars, and I'm currently working on real-time video try-on. It has 100+ registered users and has raised $9K+ through pitch competitions.
React
TypeScript
SwiftUI
Supabase
Gemini API
GCP Cloud Run
FastAPI100+
registered users
$9K+
raised in pitch competitions
3
generative features: static try-on and digital avatars, with video try-on in development
stylesyncs.com
01 · the problem
When you shop for clothes online, it's hard to tell how something will look on you until it arrives, and that uncertainty leads to returns. StyleSyncs lets you see how clothes will look on you before you buy them.
02 · what I built
- Co-founded StyleSyncs and built its AI features, including the image generation pipeline.
- Currently developing the flagship real-time video try-on feature, which overlays clothes on live camera feeds.
- Engineered a generative pipeline based on Gemini 3 Pro Image to power the platform's static virtual try-on and digital avatar creation features.
- Built the product on React + TypeScript with Supabase for auth and data, deployed on GCP Cloud Run.
- Building an iOS app for StyleSyncs in SwiftUI.
03 · how it works
01 upload
User uploads a photo and selects a garment through the React/TypeScript frontend, with Supabase handling auth and storage.
ReactTypeScriptSupabase02 garment analysis
A multimodal Gemini model analyzes the clothing item's category, fit, and texture, and that analysis guides how the try-on image is generated.
Gemini (multimodal)03 avatar
Gemini 3 Pro Image creates a clean digital avatar from the user's photo.
Gemini 3 Pro Image04 try-on image
Gemini 3 Pro Image renders the garment onto the avatar.
Gemini 3 Pro Image05 result
Results are served from GCP Cloud Run back to the user's gallery.
Cloud Run01 upload
User uploads a photo and selects a garment through the React/TypeScript frontend, with Supabase handling auth and storage.
ReactTypeScriptSupabase02 garment analysis
A multimodal Gemini model analyzes the clothing item's category, fit, and texture, and that analysis guides how the try-on image is generated.
Gemini (multimodal)03 avatar
Gemini 3 Pro Image creates a clean digital avatar from the user's photo.
Gemini 3 Pro Image04 try-on image
Gemini 3 Pro Image renders the garment onto the avatar.
Gemini 3 Pro Image05 result
Results are served from GCP Cloud Run back to the user's gallery.
Cloud Run04 · key decisions
One image model family across the pipeline
Consolidating avatar creation and try-on rendering on Gemini 3 Pro Image simplified the pipeline and kept output style consistent between stages.
Ship to real users early
We launched before the pipeline was perfect and run pilots with real users to help decide what to work on next.
05 · where it stands
- ✓100+ registered users on the live platform at stylesyncs.com.
- ✓$9K+ raised through pitch competitions.
- ✓Real-time video try-on in active development.
- ✓iOS app in development.
06 · technologies used
ReactFrontendTypeSwiftSupabaseDatabaseGemini APIAIGCP Cloud RunHostingFast