BMAD, Loops, and the Initiative Problem
Picking a back-burner project back up with the BMAD-method, running into agents that won't take initiative, and wondering whether that's the framework or my own closing prompts.
Posts coming weekly, usually Wednesdays.
Welcome to my corner of the internet. I document things here bugs, ideas, architecture wins, AI ramblings, time audits, and the occasional philosophical spiral.
If you really want something to work... keep it simple.
Picking a back-burner project back up with the BMAD-method, running into agents that won't take initiative, and wondering whether that's the framework or my own closing prompts.
A reverse prompt that flipped how I work with AI, why I use it as a memory-keeper and not a coach, and the small differences between models that actually matter.
On portable applications, admin rights, and the quiet difference between the computer you use every day and the one you actually own.
Six months after publishing LLMR, the project has a proper home, a generator, and real numbers. Here's what changed and why it matters for anyone building with AI.
Everyone says modern frameworks are fast. But is a Svelte page actually faster than a hand-written static HTML file? I ran the tests. The answer is more nuanced than the benchmarks suggest.
Twenty years in IT and I was still manually checking SEO dashboards like it was 2008. I don't have an SEO problem. I have a follow-through problem. So I automated the follow-through: GitHub Actions + Google Search Console API + Claude = 5 actionable issues in my repo every Friday at noon.
When exactly does this hook fire, how do you read submission data, and how do you avoid the most common traps? Three real copy-paste examples: Slack notifications, custom DB writes, and on-the-fly entity creation.
LLMR v2.0 is live; a JSON site index that gives AI systems your entire website in one fetch. Now with a formal spec, open-llmr.org, and a full ecosystem plan.
Two days, a dead SD card, an abandoned Acer BIOS, 83 million Mistral tokens, and a bot that only knew how to say "Task Completed". Building a personal AI assistant is not like the YouTube tutorials make it look.
From sneaking out of bed to watch MS-DOS boot, to building wind turbines in Haiti and architecting systems for Fortune 500 companies. This is who I am and how I got here.
December 25, 2025
I'm building Constituant, a platform where French citizens can vote on real legislation from the french National Assembly. Yes, anonymously. Yes, using AI summaries. And yes, I know exactly how broken that sounds.
Wanna know more?! Click the link and read on!
December 10, 2025
Some of my thoughts, and experience, with artificial intelligence in enhancing productivity and creativity.
December 07, 2025
How pausing one game project led to launching two new ones : a civic engagement platform and an open-source social MMO.
October 30, 2025
Jibberish mode but for LLMs when browsing websites.
October 08, 2025
What do you do when your team of two becomes a team of one, and the projects are on fire?
September 22, 2025
But you can get the right ones.
Sometimes I wish days were 72 hours long, so I could get more things done.
September 06, 2025
Hey Claude. I wanna bounce ideas about my game project.
Sure thing. Here's your... entire project. Just coded from top to buttom. That doesn't work... ^^
September 04, 2025
See this pile of trash ? It's an unfinished website, make it work the way we want but we wont tell you how!
Here's some of the bugs I fixed just recently after inheriting the unfinished "legacy" code base
August 30, 2025
A real debugging story: how AI assistants (GPT, Claude, Gemini) helped fix character collisions and movement in a Godot 3D project, turning a buggy scene into a functional game prototype.
August 30, 2025
Another legacy system integration nightmare featuring PowerPoint "documentation," Drupal hooks that fire too early, and web services that return completely different data than promised. When you can't test locally and debugging happens through client screenshots, every assumption becomes a liability. Spoiler: the eligibility check wasn't returning "Yes" or "No" like the specs claimed.
After using GPT for a while, I found that Claude's approach to coding tasks was more aligned with my workflow. Except now it doesn't anymore, and I'm not sure why.
What started as a "quick bugfix" spiraled into a multi-day nightmare>
Online scams are everywhere
I've been working on side projects a lot
Working on automatically updating my blog with git push. This launches a workflow on GitHub, then pushes everything to FTP via a script.
I've not been able to set my head straight to actually be productive. I've just not been able to produce the things on my list.
Posted on July 6, 2025
This week was chaos and cleanup.
// no JS frameworks were harmed in the making of this log
I believe the web should be simple. I don't like frameworks. They bloat, obfuscate, and complicate things that HTML already did well decades ago.