// the blog
notes on running agents
Field notes from building vord and from running a lot of agents at once. Practical, opinionated, written by people who do this all day. No thought-leadership, no funnels — just what we've learned about keeping parallel agents from turning into chaos.
-
multi-agent
2026-06-23
Most of the discussion about AI coding agents assumes you pick one. In practice, if you run agents all day, you don't. Here's what running both on the same codebase actually looks like — and why vendor lock-in bites in ways people don't usually mean.
read →
-
workflow
2026-06-16
You can run five agents in tmux right now. It works — until you try to answer the question that actually matters: which one needs me? tmux can't tell you. The reason why is worth understanding, because it's the whole argument for a different tool.
read →
-
attention
2026-06-09
The first agent is a joy. Around the fourth, something quietly breaks — and it isn't the code. It's you. The bottleneck moved from typing to attention, and nobody warned you what that looks like in practice.
read →