https://technischepartij.nl Notes on software engineering and building with AI agents. Wed, 28 Oct 2026 07:23:27 GMT https://technischepartij.nl/posts/absorbing-the-change/index.html https://technischepartij.nl/posts/absorbing-the-change/index.html Tue, 27 Oct 2026 00:00:00 GMT Looking back at one year of agentic development at STX, I'm starting to arrive at the conclusion that once we were able to reduce the friction of shipping code to production, our new bottleneck is the ability of the business to absorb the change. Now that we can automate any process, getting stakeholders involved in shaping the automation requires time on their calendar. We no longer need to negotiate the launch date of a new capability because it takes time to build it and we have to trade off features against time. https://technischepartij.nl/posts/productivity-in-the-agentic-age/index.html https://technischepartij.nl/posts/productivity-in-the-agentic-age/index.html Fri, 25 Sep 2026 00:00:00 GMT With everything going on in AI and in the industry right now, it feels like we're all getting a little bit crazy, and obviously this is an industry and life-changing period in time. And while many of us are being 10x productive because we can produce more code than ever, we're treating these new capabilities as just faster horses, keeping our existing ways of working, ceremonies and tools. I want to share my observations on common pressures that I hear people mention when we're discussing what's going on right now. https://technischepartij.nl/posts/ode-to-the-hobby-project/index.html https://technischepartij.nl/posts/ode-to-the-hobby-project/index.html Thu, 30 Jul 2026 00:00:00 GMT So this is going to be a bit of a rant and a bit of an ode to the hobby project. I think it's fair to say that hobby projects have been instrumental in my professional development as a software engineer and without them, I would not have made it to the position where I am: leading and architecting complex software together with awesome people. https://technischepartij.nl/posts/tiny-token-suckers-part-two/index.html https://technischepartij.nl/posts/tiny-token-suckers-part-two/index.html Fri, 10 Jul 2026 00:00:00 GMT While researching the little tiny token suckers that sapped my tokens after switching to Fable — even though I was doing all the best practices — I did a lot of research into how to optimize token usage without having it destroy or debilitate my workflow. And since the research I did all came from blog posts and other people's writing, I think it's only fair if I summarize what worked for me. I am not saying that my way of working is the way of working everybody should be using. I'm just describing what works for me and maybe it will benefit others. https://technischepartij.nl/posts/radical-shift-left/index.html https://technischepartij.nl/posts/radical-shift-left/index.html Thu, 09 Jul 2026 00:00:00 GMT So with Markdown rapidly becoming the industry standard programming language, I think we aren't talking about a very important concept that should be leveraged to reduce the slop and complexity that we are creating together using LLMs. https://technischepartij.nl/posts/tiny-token-vampires/index.html https://technischepartij.nl/posts/tiny-token-vampires/index.html Wed, 08 Jul 2026 00:00:00 GMT Okay, so my Claude code setup is pretty simple. I run Claude in a TMUX and I run a few of them. I've reached a point where I'm comfortable creating a high level project plan using an LLM. And by rigorously reviewing the project plan, removing ambiguity, I trust that when I hand it off to create a set of issues in an epic on GitHub, that it does more or less the right thing. I give it a cursory overview and usually then simply hand it off to a swarm of agents. Our setup at this point has been working quite well and we have been delivering several greenfield projects with near perfect accuracy. How we do that at this point is a little bit besides the point, but what I want to really remark is that my way of working hasn't changed significantly for a couple of months now, but my token usage went up crazy around the same time that Fable got released.