Hey! I'm Max, 26 years old, and I'm a total geek. I'm writing this because I wanted to do something other than programming as my creative outlet. Writing is at the top of the list of candidate activities, so I thought a little about the what, when, and where, which resulted in some decisions. The subject will be mostly technology, with a focus on programming, because it's my long-term passion, and I have tons of words that I want to put on neat (web)pages. The "where" is simple. I have a website, so I'll put my thoughts there. I self-host my website, and the prospect of expanding it with some HTML and CSS helped me convince myself to start blogging. I'll write whenever I feel like I have something to say, but writing is hard for me, so getting started requires a lot of effort. However, with the right timing I feel it's going to be much easier. This morning I just sat down, wrote hello world, and there are enough words for a tweet already! I'll limit the lifestyle stuff to the minimum, but I think that I have to introduce myself in my first-ever blog post, so let's get it over with and get to fascinating stuff.

I started getting into programming in middle school, then computer science in high school, then an engineering degree in CS, and then a master's in AI. I got really nerdy in university, constantly reconfiguring/ricing my Linux install, while I should've been paying attention to lectures, and tinkering on a programming language that never fully materialized. I also started getting into machine learning. It was like a new world of possibilities that previously I didn't even know existed. In my spare time I went with my friends to dozens of hackathons, duct-taping stuff, and enthusiastically presenting it at the end. It was really fun and there was free pizza :) Recently, I took down the projects from my GitHub and awards from LinkedIn, because the projects themselves are nothing to write home about, and I think I'm a bit too old to brag about such things anyway!

Anyway, I love going on tangents, which risks the reader getting sleepy, so let's push the intro forward.

I finished university with a thesis on learning concept embeddings from knowledge bases using the ALC description logic. When someone looks at that, the most common thing they say is that they don't get it. I attribute its mind-boggling properties to description logics being a niche topic, these embeddings being a bit awkward, and my not-so-great writing skills. But in the end it got published as an article, so I'm really proud of it!

After university I got a cozy job at Google. I liked it at first for superficial reasons like salary, trips, free lunches, gym, and never-ending lactose-free decaf ice lattes. But the project work quickly turned out to be uninteresting and unrewarding. I did some benchmarking for Borg at first, and invented some side projects helpful to the cause, to make work more interesting. Then I got to do cleanups of Borg code, which is almost as old as Google. It was difficult to work with this code. The tiniest changes broke things in far-away places, and it was really hard to wrap my head around this huge system. I had no patience for it, so I switched teams to something seemingly more aligned with my interest in AI. Gemini for Google Cloud sounded amazing on paper, but in reality it was just plumbing Protocol Buffers from one place to another, in the most frustrating ways possible. I definitely wanted to get away from that. First I burned out hard and was sick for months, then I returned to work and asked around for other projects in the company. But there was nothing interesting for me, so finally I resigned in March, and after almost 3 years of corporate paradise I felt truly free to do whatever I want.

Between April and October I was recovering from burnout. To recover, I set out to make something cool, useful, and challenging. I definitely learned more in those months than during my few years at the supposed dream job, and in contrast to in-house tech, those skills are transferable! And I will transfer them to you through this blog! Long story short, I finished the thing (or at least the proof-of-concept), and I was able to proudly show it to my friends. I think the thing has many interesting aspects, so I have lots of ideas of what to write on this blog :)

During burnout recovery, I remembered that I really liked my university years, so I enrolled in the PhD program at my hometown university. I chose to study neuro-symbolic approaches in reinforcement learning. They're just so happy to let me lead my project, which is exactly what I need to master a tiny slice of the AI field! I'm just starting tomorrow, so I don't have anything concrete right now, but I'll write about research stuff on my blog too.

Actually, I'm thinking to use this domain for a personal blog about my interests, my research, and a bit of life, and a separate domain for a more professional blog for the thing I made, so that internet surfers can discover it and actually use it if they like! The idea of sharing tweet-sized thoughts is also appealing to me, so I'll experiment with that format too.

So that is it for who I am! Next up, I want to tell you about setting up a freaking e-mail server! Self-hosting email is pretty simple if you limit the scope just a tiny bit :)