The Engineering & AI Systems territory. Long-form thought on AI-assisted development, agent mode, DevOps philosophy, and what changes when the human is three abstraction levels above the code. Some essays lean technical, some lean philosophical; they share the systems-engineering frame.
71 posts • Page 1 of 4
I typed my thirty-two loops, then drew them as a graph, and thought the loop was the top of the system. It turned out to be the roof of a stack eight floors deep, and the floor holding all the weight is the smallest one: the edge.
Typing my automation handed me most of its rules for free; the type carries the contract. One rule came free with nothing, and it was the one my system was breaking: every file gets exactly one job that writes it. On the difference between rules you remember and rules the structure carries.
I gave every job in my automation a receipt, a health check, and a green light, and the system still would not fit in my head. A dashboard shows the state of the parts. The question I actually had was about the shape of the whole, and no amount of state answers a question about shape.
The best thing anyone told me about my own system this year was that I had not invented it. The scripts I grew by hand are a data pipeline, the shape is a DAG, and the discipline is solved. What the tools actually sell, why I am not buying yet, and the difference between wanting an engine and wanting a map.
Thirty-two automated jobs run underneath my days, I wrote every one of them, and I could not hold the system in my head. The problem was never my memory. It was a word. "Loop" was one word doing the work of seven.
Edgar Morin died in May at 104, and I met him through his obituary. He spent eighty years on the question I have been building around: how to think inside the weave without cutting it apart. Two book reviews are coming. This is the marker I am leaving at the trailhead first.
Making complexity visible is the method. Orientation is the objective function. The research program now has an installable runtime, laboratories that keep their data, and its first pre-registered experiment, running on my own life.
The first program you write used to be the machine greeting the world. Hello, world. I built a different first program: my own system greeting me. One warm door instead of four scattered starters, and the greeting sits inside the loop it opens, so it can reach in and improve itself.
A book I wrote about selling options for income, built on one law: survive first. It shows a record most books would hide, and that is the reason it exists. Out now in Kindle and paperback.
I built a method, pointed it at my own thirty-four repositories, and watched the sprawl become a page I could read. Then I tried to prove its deeper claim and the first experiment came back null. Under the practice is a field: the science of complexence, the loop as a recursive equation, meaning velocity, and the cognitive form, an idea held apart from how you say it. A research program, published openly, nulls and all.
Complexence is the capability of staying oriented inside complexity. Complexence OS is the rough machine I built today to run it: you talk into a file, one Chief of Staff agent sorts what you said and hands the rest to a few specialist roles, and you review only the exceptions. Day one of going from AI assisted to agent director. The method is open source; your data stays yours.
The fight over whether AI is good or bad is the wrong fight. AI is a friction-removal engine, and there are two kinds of friction: the kind that is building you, and the kind that is only taxing you. Telling them apart, in the moment, is the whole skill.
Most defenses of AI are about productivity, and they are weak. The strong case is older and bigger: this is the next rung on a seventy-year ladder of removing effort that was never building anyone. Here it is, made at full strength.
AI does not destroy your work in front of you. It moves the cost somewhere you cannot see: to later, to the invisible, to the version of you that stops getting built. The strongest case against leaning on it, made in full.
I had thirty-three git repositories and treated all thirty-three as if they were on fire. The fix was not more discipline or a better tool. It was a census: an inventory, an edge, and a rule for what counts as alive. Domain cartography applied to my own sprawl.
The hardest part of mapping a complex domain is the blank page. Here are the six layers I run every domain through, from primitives to levels, with the question each one answers, how to actually do it, and the failure mode that shows up when you skip it.
Domain cartography is the practice of charting a complex domain you have to live and act inside, until it is legible enough to navigate. Not making it simple. Making it readable. Here is the field, the method, and the six layers every map needs.
Complexity is the condition of the world. Complexence is the quality you bring to it: the capability of standing inside a system larger than yourself, seeing it whole, and still choosing. A word for the thing all my work has been about.
The same tool that can fold a beginner's learning curve shorter can also erase it. AI compresses apprenticeship when it transfers expertise and destroys it when it lets people skip the repetitions that were quietly teaching them.
The subscription is the small line. The real meter is the time you spend feeding context, checking output, catching the confident mistake, and cleaning up the work that looked finished. That hidden labor has a name now: botsitting.