Dmitry

The machine that never stops

There's something in my head that doesn't stop. It scans everything — looking for patterns, for the logic behind how things work and why people do what they do. When it finds a real problem, it doesn't let go.

I've gone days without sleep because of this. I've forgotten to eat. I've lost track of time completely, and come out the other side with something working that wasn't working before. That feeling when a hard problem finally breaks — I don't know anything quite like it. Then it gets quiet again. And I start looking for the next thing.

It's my best quality. It's also what makes normal life harder than it should be.

Where it started

It didn't start in a classroom. It started in a game server.

I was a teenager who liked games, but more than playing — I liked taking them apart. Setting up servers, deploying to the cloud, watching the numbers, chasing errors that showed up at 2am. Nobody taught me this. It just needed to work, so I figured it out.

School was okay. Technical college was worse: old textbooks, abstract explanations, no real examples. A 30-second video sometimes taught me more than a full semester. I still finished on scholarship, above average. Figure that out.

Into the deep

Ukraine isn't the kind of country where fresh graduates walk into their field. You need connections or real talent. I had connections — my stepfather worked at sea, put in a word, and suddenly I was a motorman: up to my elbows in engine grease, cleaning parts of machines the size of small apartments.

It wasn't glamorous. But there's something very clear about a diesel engine that won't start when you're far from any port. You fix it or you don't. Nothing else matters in that moment.

I stayed. I got better. I worked through the early years and ended up at Stolt Tankers — a global chemical tanker company, proper standards, real accountability. Five years now. The work is honest. The pay is honest. And four months at a time, I go away.

The year I tried to change everything

April 2025. I came ashore with a plan — not to try things, but to actually do them. I learned Python, C#, Kotlin, Java. I touched Spring Boot, PostgreSQL, Docker, AWS, Firebase, React.

I shipped things. PotionCalculator hit Google Play — a crafting optimizer for Albion Online with live market data. P2Poker: a peer-to-peer LAN poker game for six players, no server, tested with real people. Glazd went live. BizRadar reached MVP stage in Poland — market analytics for small business owners who couldn't afford the enterprise version.

By May 2026: no income. A relationship under a lot of pressure. A list of mistakes I now understand clearly.

Wrong tech choices. Too scattered. Timelines I underestimated badly.

I learned more in thirteen months than in the four years before. Just not fast enough.

Tactical retreat

I'm going back to sea.

Four months. Same routes. Same distance from the people I care about.

I've thought about whether this is failure. I don't think it is — but it's not a win either. It's a break. Time to think clearly, study properly, and recover what a year of scattered effort used up.

I know what went wrong. I'm building the next version.

There will be another attempt. Then another.