TryOn AI

TryOn AI lets you virtually try on outfits from e-commerce store with just one click. Simply share a product from your favorite shopping app and instantly see how it looks on you before you buy.

Comment

Inspiration

Shopping for clothes online is convenient but often frustrating. You can’t see how an outfit will actually look on you. I wanted to remove that uncertainty by letting people virtually try on any outfit. With nano banana it was easy then ever to build this kind of apps. There are plenty of try on apps/tools but there the main friction is that if you want to see how it will look on you, you have to download the product image or take screenshot and uplod. Try-On AI goes beyond the typical try on apps availabel in the market. It offers single click try on directly from e-commerce store.

What it does

TryOn AI lets users instantly try on any outfit before they buy.

  • Upload your photo once
  • Share a product image from any shopping app/website or snap a picture in a store.
  • See a realistic try-on preview in seconds.

How I built it

I built a cross-platform mobile app using Flutter for a smooth UI and AWS for backend services and processing store shares. The try-on engine uses Gooogle's latest Gemini 2.5 Flash Image(aka Nano Banana) generation model to generate high-quality outfit overlays while preserving user privacy.

Challenges I ran into

Building a seamless sharing flow across multiple e-commerce and getting the product image was the challenge that I faced to solve this problem I created custom backend services on EC2 to process shared product urls and send back the original product image to the app. currently it supports 15+ popular e-com platforms for single click try on experience.

Accomplishments that I'm proud of

I'm super proud about the experience that I built to do a seamless tryon with single click, directly from e-com store. It was the most difficult thing and no one had done it before but I did it. So I'm proud of this I am also seeing good result from users as some users are already paying for it :)

What I learned

  • Implementing share sheet and actions in android/ios apps.
  • Writing good prompt for accurate image editing result.
  • Using RevenueCat with flutter.

What's next for TryOn AI

  • Adding support for more popular e-com platforms
  • Apart from current subscription plan for monthly token limits offering credit buying system as consumables.

Built With

Share this project:

Updates