My Loop Engineering Workflow
Three nested loops run my agent harness: objective, execution, feedback, refined objective. Fork-join in the execution, a gate at every join.
Product Generalist, Design Thinker, and AI Hacker building at the intersection of technology, design, and entrepreneurship.
Three nested loops run my agent harness: objective, execution, feedback, refined objective. Fork-join in the execution, a gate at every join.
AI is a brilliant butler in a house you're only renting. He cooks everything. Here's why every engineer still needs one dish they make by hand.
My pens are still full and my notebooks untouched. I never decided to stop writing by hand. AI just quietly became where my ideas live now.
For decades more scope meant more time. AI broke that math. The new limit isn't how fast you build, it's how fast you can responsibly approve.
Everyone treats the company brain as a knowledge problem. It isn't. The hard part is interpretation, and AI moved that problem rather than solving it.
Enterprises are pouring billions into AI agents. Almost none can prove they earned a dollar. The measurement layer doesn't exist yet, and that gap is a company.
You used to hide behind other people's work. Now AI hands you an answer, you ship it, and the whole team can see the gap you never closed.
We celebrate the founder who keeps blooming through the drought. Nobody asks what it costs them to stay green right where they are.
One PM. One engineer. One face. The new Holy Trinity of company building, and why the product manager has to change first.
Your team still has questions. They just ask a machine now. And that is a problem you cannot manage your way out of.
If you asked AI to validate your idea and so did your cofounder, your board, and your team, you did not get four opinions. You got one.
The most valuable thing you produce with AI is not the output. It is the conversation that delivers that output, what you asked, how you asked.
From Columbus to AI - how leverage has evolved through geography, infrastructure, code, attention, and now intelligence. The question isn't whether a new era is forming, but whether your ambition is pointed at it.
In AI native workplaces, the tension is no longer who writes better code. It's who extracts better intelligence. When AI writes the code, what's left to compete over?
After studying past technological revolutions, here is what most people still get wrong about AI and what it means for their careers.
A subtle shift from AI coding: reading code faster, and why tiny diffs suddenly feel like a slowdown worth unpacking.
Andrew Miracle chronicles migrating his site from WordPress to Jekyll, and why AI-native workflows fit a digital home for modern publishing.
What if your business could run without you? Not passively, not automatically — perpetually. Exploring the economics of AI-sustained systems.
Open models are planting forests. Closed models are selling lumber. If you have seen this pattern before, it is because you have.
I kept choosing partners who thought like me. It felt like flow. It was a trap. Why the best partnerships need friction, not harmony.