Prague · Ukrainian 🇺🇦
I lost everything twice. First in 2014, when I had to leave Donetsk. I was 25, managed nearly 100 people in construction. Then the war started. I left with nothing.
Started over at the bottom — rating search queries for Yandex. Then QA. Then built a dev team. Then watched clients promise payment "after the next round" for years. That got old.
Switched to venture. Started as an assistant, ended up a partner. 100+ deals in 3 years. One of them is now worth half a billion. Then 2022 happened. Full-scale invasion. The syndicate had Russian money in it. I couldn't stay.
So I left again.
Spent the next two years at venture studios in Luxembourg and Dubai. Launched dozens of products. Tested over a thousand hypotheses. Watched most of them die.
Meetings about meetings kill companies faster than bad ideas. The best team size is two people who ship. And you learn more from a product that flopped in two weeks than from a deck that got polished for six months.
I build alone. Pick an idea Monday, ship by Friday, see if anyone cares by Sunday. If they don't — next idea.
I'm not a great programmer. Never was. But I understand product, marketing, analytics, sales — all at about 6 out of 10. LLMs turned that into 8 or 9. That's enough to ship.
SkillCanvas is my current bet. I built it because I needed it myself. Everyone I talked to about Claude Code skills said the same thing: powerful but impossible to debug. So I'm fixing that.
Will this be my life's work? No idea. Six months ago Claude Code skills didn't exist. I don't pretend to know what's coming. But this is the product I believe in right now. And I'm shipping it.