Roman Koropets

Roman Koropets

Mobile dev, 15 years /Swapify/Sunny Seed

About

I've been building mobile apps for 15 years, currently two products. Swapify - a face-swapping app, 600K users and 4M generations. Sunny Seed is the personal one: a baby journal with trackers, WHO growth charts and an AI assistant. My wife and I built it for ourselves after our daughter was born, then realised we weren't the only ones drowning in notes across three apps and a paper notebook. She's a children's book illustrator - every illustration in the app is hers, none of it is stock. I write both platforms natively (SwiftUI on iOS, Kotlin and Compose on Android) plus the backend, and ship in six languages including Arabic with full RTL. Happy to talk about mobile subscriptions, on-device AI, or what shipping a consumer app solo actually looks like.

Badges

Tastemaker
Tastemaker
Gone streaking
Gone streaking
Gone streaking 5
Gone streaking 5

Maker History

  • Retrolife
    RetrolifeLife Simulator Through Real Decades
    Nov 2026
  • Sunny Seed
    Sunny SeedBaby tracker with on-device AI and WHO growth charts
    Sep 2026
  • 🎉
    Joined Product HuntAugust 20th, 2026

Forums

If you could be born in any year from 1950 to 2010, which one would you pick?

Hi everyone, I'm Roman, the maker of Retrolife, launching here on November 9.

Retrolife is a life sim set in real history: you're born in the USA, West or East Germany or the USSR and live a whole life one year at a time, with that year's news and prices.

So here's a question I'd love your answers to: if you could be born in any year between 1950 and 2010, which would you pick, and which moment would you want to live through, again or for the first time?

•

10d ago

Retrolife - Life Simulator Through Real Decades

Live a whole life year by year in the real 1950s–2010s of the USA, Germany or the USSR. Real news, era prices, family and heirs. A text life sim for iPhone.

What we run on the phone, and the one thing we didn't

Building Sunny Seed we had to decide this feature by feature, and the answers turned out not to be consistent.

On the phone: WHO growth percentiles - the LMS tables ship inside the app, it's arithmetic, and a server round-trip to divide two numbers would be absurd. Photo tagging for smart albums, via Apple Vision on iOS and on-device ML Kit on Android. And dictation on iPhone, which is transcribed locally and never becomes a file.

In the cloud: the AI assistant. Not because of model quality - because it needs months of a child's history in the context window to say anything useful, and that's the part a phone can't hold at a speed anyone would tolerate. Decode speed gets all the attention in these discussions, but prefill on tens of thousands of tokens is where local inference quietly stops being practical.

The pattern that emerged for us: narrow, purpose-built models go local. Anything that has to read your whole history stays on a server.

View more