A few years ago, artificial intelligence mostly meant one thing. You asked a question, and the system gave you an answer. It was…Continue reading on Medium »
Continue reading on Medium »
We have lots of benchmarks for AI agent correctness and capability. We have far fewer tools for measuring something subtler: when an agent reads two competent, well-argued positions on a hard topic and picks one — what pattern is driving those picks?Th…
How language becomes geometry — without losing meaningIn the last post, we built the first foundation:Text → Tokens → Numbers → (lots of math) → Tokens → TextWe said:tokens are the piecesembeddings are the numbersThat is the right starting point. But i…
Claude Certified Architect Multi-Agent Research: Hub-and-Spoke Design, Context Isolation, and Why the Super Agent Anti-Pattern Fails Every…Continue reading on Towards AI »
Your APIs were designed for humans. That’s about to be a problem.When Facebook’s engineering team designed GraphQL in 2012, they were solving a mobile problem: REST endpoints were returning too much data over slow networks, and iOS clients were paying …
The Three Generations of Programmers — How Has Artificial Intelligence Affected Our Programming Abilities?There is a sharp paradox lived by the programming community today. At a time when a developer with five years of experience completes a programmin…
IntroductionContinue reading on Medium »
Introduction: This Is a Strategic Shift, Not Just a Product LaunchContinue reading on Towards AI »
Over the past two years I’ve tested dozens of AI tools across writing, image generation, audio, video, and code.
Some were genuinely great…Continue reading on Medium »