Beekeepers, not managers
AI agent swarms organize themselves, but more like ant colonies than companies. Why that matters, and why humans should act as beekeepers.
Tag: Agents
Your AI Coding Agent Isn’t Using Documentation the Way You Think
Coding agents do not use documentation like slightly faster human developers. They have created a documentation system of their own.
Your AI Agent Doesn’t Need More Agent Skills. It Needs Better Habits
New research explains why agent skills work—and why adding more of them can make an AI system harder to control.
The Human Is the Outer Loop
A factory, underneath the word, is a loop. The agentic software factory, the version being built right now, puts the models inside that loop.
No Mac Mini Required: The Complete Guide to Safely Hosting AI Agents on Your Current Mac or PC
Professionals buying dedicated Mac minis to act as silent, 24/7 personal servers for their local AI agents. You don't have to.
The Lead That Couldn’t Step Down
What happened when five Claude Code instances were given a constitution, a parliament, and a goal — and built a coordinator.
No Orchestration Required: How Parallel Coding Agents Coordinate Through a Shared Log
The industry is selling orchestration frameworks for a problem that, on one machine, a SQLite file and three hooks already solve.
Converge, Not Fork
What happened when eight Claude Code instances shared one repository, one git index, and no manager — and spent an afternoon building.
The Kanban board is the file: lightweight coordination for parallel Claude Code
A Kanban board tells you is what's about to happen, what another agent is doing right now, or which card is yours to pick up.
The Other Half of the Rediscovery
Thariq, Karpathy, Theo — HTML for LLMs has crossed into the open. But the rediscovery is only the read side. The write side is where the document edits itself.