| Github | https://github.com/mgratzer |
| Home | https://mgratzer.com |
| Work | https://www.topmind.eu |
| Bsky | https://bsky.app/profile/mgratzer.com |
Weekend project that got a bit meta: I build an agent skill that coaches you how to build a coding agent from scratch.
Eight steps and you understand what's happening when these tools work.
I finally wrote something longer than a toot. First blog post, about building software with agents and the workflow I developed around it.
I built an app for my kids over the holidays. What surprised me wasn't how much the agent could do, but how much the result improved when I treated it like onboarding a new team member instead of prompting a tool.
The post covers the workflow, what worked, where agents fall short, and why code review matters more now, not less.
There’s another Software Architecture Humble Bundle, this time with books from Pearson. (17 more days)
https://www.humblebundle.com/books/software-architecture-pearson-books
Finally, full disclosure: I am in now way, shape, or form oposed to the use of LLMs in general. I use them a lot myself, for tasks I know they can acomplish. And I always keep myself as the human in the loop. In fact, the code to render the charts above was generated by an LLM, and then manually reviewed and valdiated by myself, to save time.
What I am oposed to is the mindless use of LLMs to generate slop, or worse, supposed journalistic content. Especially on tax payer money.
We keep repeating the same hubris. We keep praying to the oracle of prediction, hoping we'll be spared the storm.
But the storm always comes.
And it never looks like the forecast.
Here's why I'm done future-proofing my life.
https://www.joanwestenberg.com/p/i-m-done-future-proofing-my-life
Armin Ronacher
Ruth — of systems & design
Mario Zechner
rands
JA Westenberg
Mina