Notes from building and deploying AI systems Main Homepage: Subscribe via email or RSS feed...
François Chollet was talking about his children. "Are we just going to outsource our minds? Our thoughts, our cognition, and all our abilities to the machine? What are we going to retain? What's it going to mean to be human, not in ten years, but fifty years from now? My son, my daughter, they're going to live to see the 22nd century. That's something I think about a lot." That was the ARC-AGI 3…
Martin Casado , GP at a16z, put it plainly last week: It's only a matter of time before only the model creators have access to the most powerful models. The rest get access to smaller, distilled versions. Or access the models through first party apps and services that don't provide direct access to the token path. Raghu Raghuram followed with the economic logic: If the investment needs of training…
Origin Story This project started as a sequel. Late last year I published Agentic Patterns in OpenAI's Codex — a project that took the 21 agentic design patterns from Antonio Gulli's textbook and located them inside OpenAI's Codex CLI. That project proved a method: take a real agent runtime, read the source, extract the patterns the textbook didn't anticipate. Then, on March 31 2026, the Claude…
Query camouflage and using frontier AI on your own terms I use frontier models every day. For code, for research, for reasoning tasks my local models can't touch. And every time I do, I'm handing over my problem decomposition, my domain framing, my novel question — as training data. I've been thinking about what to do about that. The Sovereign AI Problem Every nation, every enterprise, every…
The debate over whether language models can genuinely reason has been burning for a while now. Gary Marcus put it this way: "LLM 'reasoning' is so cooked they turned my name into a verb." Karpathy has a word for what Marcus is reacting to. He calls it jagged intelligence : "The word I came up with to describe the (strange, unintuitive) fact that state-of-the-art LLMs can both perform extremely…
Introduction Something is shifting in public health—and most people building AI are looking at the wrong layer. I've been building deep learning systems for public health in India for seven years. Before that, a decade in the US—deep learning since the AlexNet days, across academia, IBM, and startups. I've watched a phone-based baby weight tool reach 100,000 vulnerable newborns, a pest advisory…
Mapping agentic design patterns to real production code in OpenAI’s Codex.
A repo-editing coding agent is not a chatbot with tools stapled on. It is a system that reads a codebase, decides what to change, executes commands, edits files, and produces artifacts intended for human review and CI. That combination produces a predictable failure mode: the agent sounds right until it does something surprising, expensive, or irreversible. This tutorial describes how to build a…