2026-07-06 15:55:08
Agents can write code faster than we can absorb it. Here are some techniques for doing that efficiently.
2026-07-06 15:53:36
Are activities motivated by an array of reasons much less useful than ones done with a distinct purpose?
2026-07-06 15:49:14
Or how we refactored half the business to fix a janky script.
2026-07-06 15:48:24
Codebases deteriorate quickly when automated agents generate user interfaces without canonical constraints.
2026-07-03 12:09:53
LLMs have broken legibility of effort - our ability to tell, at a glance, whether something took a human real work. What happens next?
