The purpose of this trip was (ostensibly) to attend the FOSDEM 2026 conference. Since I’m no longer working at Beeper, there is not really any work-related reason for me to attend the conference. But I wanted to do it to give myself an excuse to go to Europe, and as an opportunity to catch up with former Beeper coworkers, and meet up with a few other friends as well.

I woke up very late today (like 14h). I guess I needed some sleep after my crazy train ordeal yesterday. It was kinda sleeting anyway, so I didn’t feel that bad about missing out on the morning.

I was hungry, so I started walking to find somewhere to eat. I ran across a restaurant that looked like it served local fare, and I got their beef goulash. I ordered with “dumplings” which were the bread slices. I think I should have tried with the wild rice instead.

This morning, I left Vienna for Prague, and on the way I went to Brno. Brno is very near the Austerlitz, Napoleon’s best battlefield masterpiece. After visiting Waterloo (the site of Napoleon’s ultimate defeat) the last two years I figured that it was only fair to go see the site of his greatest victory.

I woke up in time to get to the Vienna Main Station in time for the REX 1 at 09:07 towards Breclav for my connection to Brno. Unfortunately, the train was cancelled! I had to wait an hour until the 10:07 train. The train was a regional train, and was very slow. It stopped so often. We arrived late in Breclav, and I ended up getting on the wrong train. I ended up on a Regiojet instead of whatever rail line I was supposed to be on. When the conductor came around, I had to pay for the Regiojet train segment. It was luckily not that expensive. The Regiojet was also a pretty good train and it was direct from Breclav to Brno. When I arrived in Brno, it was already nearly 13h and the public transit to the Austerlitz battlefield museum was off-phase, so I took a Bolt.

This morning, I met Bodie and Georgia at around 08:30 to head to Bratislava. There were a few others who where planning on coming as well, but they slept in too long.

We walked to the Vienna Main Station and took the REX 8 to Bratislava. An hour later, we arrived in Bratislava and started walking towards the old town. The old town seems like it used to be a walled city, and it is now a really nice pedestrian area.

This morning, I woke up in time for the walking tour. There were about ten of us from the hostel in the group. Our guide was a woman named Katerina (not sure about the spelling). She took us from the hostel, through the Naschmarkt (an extensive outdoor market area), and towards the city centre. Along the way, she pointed out a variety of the shops along the market including various restaurants, wineries, and beer houses.

My plane from Dulles arrived close to on time, but the immigration queue was very long. The new ETIAS requirements seem to be making the border check process somewhat longer than before. In addition to telling the border agent my business and when I was leaving, I had to provide a fingerprint scan. I really do not know what the purpose is, but hopefully it helps them track and prevent illegal immigration (although being Europe, I’m doubtful).

I’m once again heading across the pond to attend FOSDEM as I have the last three years. I’m flying to Brussels via Dulles and then I’m going to fly on to Vienna for a lightning tour of Vienna and Prague, with possible day-trips to Bratislava and Brno. I’m writing this about two hours out from Brussels, and already the trip has been quite eventful, but you’re going to have to read the whole post to learn what happened :)

Over the last month or so, I’ve been using Claude Code for assisting with development tasks¹. One of the things that has confused me the most about the LLM-assisted programming ecosystem is the Model Context Protocol (MCP) landscape. In theory, MCP is a way to provide the LLM with data from an external system through a structured protocol. However, it sucks. It turns out that it is more efficient to have the AI just write code to analyse and manipulate the data from the external system directly. Rather than piping the data from the external system into the LLM’s context window and letting the LLM muddle through it, the LLM-written code can just do the necessary data processing. Computers have been good at running instructions on large datasets for a very long time and AI is not going to be a substitute for that.

A few weeks ago, Typst came back on my radar because of Sylvan Franklin’s videos on the topic came into my YouTube feed. In particular, Rewriting my resume in Typst in which he rewrites his resume from \(\LaTeX\) to Typst piqued my interest. I have maintained my resume in \(\LaTeX\) since I started at Mines in 2016, and so I figured this would be an opportunity for me to learn something new.

In this article, I am going to discuss how to build a career in the software industry. We will explore career tracks, and discuss the attributes needed to become a senior software engineer. Then I will provide my view of how LLMs will affect the software engineering industry and job market.