Posts
All the articles I've posted.
-
Stop Calling LLMs "Junior Developers." They're Monkey Paws.
Calling an LLM a junior developer smuggles in assumptions about intent and shared context. I find the monkey's paw a more useful metaphor because it keeps ambiguity, verification, and the cost of delegation visible.
-
The Meme, the Agent, and the Leverage Gap
Why the real value of AI tools isn't single-step task solving but tackling entire classes of problems at scale, from managing information overload to continuous monitoring, and how to find the leverage worth closing.
-
Are LLMs killing Open Source
AI-generated pull requests are making maintainers more cautious just as familiar hiring signals become easier to fake. I suspect sustained open-source work and deep domain knowledge will matter more, not less.
-
Give Claude in Chrome Your CLI Superpowers
A small discovery about the `--chrome` flag: start Claude from the right repository and its browser session inherits the skills, commands, and instructions you actually need.
-
The year of the Orchestrator
I returned to a live audience to talk about AI orchestrators: how we got here, what they may change for software teams, and the practices we still need to work out.