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.
88 posts • Page 4 of 5
Agent mode is not a faster way to type. It moves the unit of thought up from syntax to intent, turns you into a conductor of parallel work, and makes the real bottleneck the clarity you can hold in your head at once.
Microservices used to be a tax a solo builder could not afford. AI agents change that math. Here is why I am drawn to the architecture for what I build next, and the discipline that keeps it from becoming a mess.
This site was rebuilt from scratch by directing an AI agent, then migrated off cPanel onto AWS S3 and CloudFront. Here is what that actually looked like, the parts that worked, and the day a deploy script quietly deleted production.
Autonomous mode is when the agent runs the whole loop without you in each step: plan, act, check, repeat, until done or stuck. How it differs from agent mode, and where the risk lives.
When a system is hard to see, decisions migrate to whoever holds the context. You cannot simplify the ocean, only chart it: why a live view of a system widens the circle of people who can think about it.
I use GitHub Copilot at work and the Claude Code CLI at home, every day. Here is the firsthand comparison for DevOps and infrastructure work, from someone who ships with both, in the context each one actually lives in.
Life Ops is running your life as a deliberate system, designed the way you design software: observe before acting, choose responses over reactions, build systems that make the good path the easy one, and version the whole thing like code.
Agent mode is when you stop typing code and direct an AI that reads your codebase, writes the change, runs the tests and reports back. How it differs from autocomplete and chat, the loop, and when to use it.
Coherent complexity is what you get when a complex system is made legible without being made simple. You do not reduce it. You map it, until you can move through it on purpose. A field guide to the practice, and the idea behind everything else I build.
AWS is math and Kubernetes is physics, the galaxy and the atom. With managed services you watch the cost; with packed containers you watch the performance. When to reach for each, and for both.
Antifragility is not resilience: resilient things survive disorder, antifragile things gain from it. How I apply Taleb to the systems I build, to films and books made from code, and to a ten-day journal cycle.
I make music, animated films, and games from code: ChipForge, Napkin Films, and Pixel Vault. Why I hold to no samples, no GPU, and one small file by default, and what happens when I break the rule.
What changes when an AI agent gets real access to your repository, terminal and tests: the patterns that hold, the failure modes that cost me time, and a four-week way in.
DevOps was never a job title. It is a thesis about building systems that change without breaking: what compounded once the tools became a commodity, what automation never solved, and what agents change.
People of the Stars uses the 36-decan star calendar as a slow, non-human clock for decisions, with options-trading logic: choose a position that holds under several outcomes instead of predicting one.
Decanal journaling runs the year in ten-day cycles, each named for a bright star, after the ancient Egyptian decans. Why ten days, what a cycle looks like on the page, and a table of all 36 stars.
Vibe coding means accepting whatever the model hands back and shipping it. I build with AI every day, from a chiptune synth to an animated rap film, and that is not what I am doing.
I journal with Claude and build with GPT, so if there is a ground zero for AI slop, I am standing in it. What I think separates the work from the waste: purpose, not the absence of tools.
I spent years trying to save time when what I was short of was attention. Handing work to AI changed where mine goes: fewer threads running in my head, more of me at home.
The index to my seven-part series on building with AI: six weeks at machine speed, a method, AWS infrastructure, a handwriting OCR pipeline, a company website, what it cost, and patterns from one game.