Essays & insights on ai & technology.
Most teams plan the launch of an AI agent and never plan the firing. Here is the agent tenure model, the graded path from probation to a clean decommission.
Founders optimize AI cost and quality but ignore availability. Here is the AI capacity risk, why it is structural, and how to keep serving under real load.
AI agent reliability falls as autonomy rises. Here is the tradeoff most builders miss, why the most reliable agents are barely agents, and how to ship them.
AI recommends brands it recognizes, and recognition is built off your site, not on it. How to get named by ChatGPT and AI search, and measure it.
Multi-model routing sends each request to the cheapest capable model. The framework, the four patterns, and how to cut AI costs 60 to 85 percent.
Robotics raised $18.8B in 2026, past every prior year. The signal: AI moved the durable moat from software to atoms. A founder's map of where to build.
The verification layer is the next AI platform: as agents produce faster than humans can check, machine verification of AI output becomes the product.
AI agent sprawl is a non-human identity crisis. Every agent you deploy is a worker you never offboard. Here is how to govern your ghost workforce.
AI agent orchestration has a hard ceiling. You can spawn unlimited agents, but your speed is capped by the work only you can do. Here is the math and the fix.
A bigger context window is not more usable context. The usable window, the three taxes of a full window, and the context engineering fix every founder needs.
Building something interesting? Looking for an investor who’s been in the trenches? Or just want to say hello? I’d love to hear from you.
One email when I publish. Essays on reskilling, rewiring your mindset, and building your next thing. Free, no spam, unsubscribe anytime.