Umut Güden

I wrote my first line of code at ten, during the COVID-19 lockdown in March 2020. That same year I started building under a name, HMD Developments, and I have shipped everything since under it. The idea was straightforward: build software properly, charge fairly, and never cut corners. That is still the standard I hold the work to.

HMD Developments is a small studio rather than a registered company. Most of it is me, but not all of it. For mobile work and for security I bring in people I have been building things with for years, because they are better at those than I am, and pretending otherwise would only produce worse software. Some of the work is done for other people and stays private. The rest is mine, and the parts worth reading about end up here.

Most of what I know came from shipping and then living with the result. I taught myself by building things people actually used, which is a faster teacher than any course, mainly because it does not let you skip the part where something breaks at an inconvenient hour and the only person who can fix it is you.

The work itself has moved around: consumer software for iPhone and Apple Watch, messaging infrastructure, moderation and safety tooling for large online communities, and a long stretch of the unglamorous plumbing that sits behind all three. I would rather describe the problems than list the products. The problems are what transfer, and the ones I write about here are the ones that turned out to be harder than they looked.

I have also started businesses, run them, and closed some of them down. The closing taught me more than the launches did. A product can be working, growing, and still be the wrong thing to keep alive, and noticing that early is worth more than any amount of enthusiasm. It is the least romantic skill in this line of work and the one I would most want to be judged on.

So what occupies me now sits between the engineering and the decision. What is worth building at all. Where an interest in artificial intelligence genuinely changes a product, as opposed to where it only changes a pitch. When to stop. I find those questions harder and more interesting than the code, which is a reversal of how I felt at thirteen.

Away from the screen I read a good deal, mostly Turkish and translated literature, and I write essays about it here: etymology, attention, and what we hand over for convenience without noticing. Those essays are the part of this site I would point at first. They are also, I suspect, the more accurate picture of how I think than anything in the engineering write-ups.

I am seventeen, I work from Istanbul, and I am applying to read computer science with a focus on machine learning for 2027 entry. I am not in a hurry to be finished with any of it.