Build notes

Working
in the open.

What I’m building, what broke, and what I’d do differently. Mostly automation and the tooling around it.

  • automation
  • llm
  • multi-agent
  • python
  • tooling

Latest4 min read

Why I left Hermes: multi-agent is a protocol problem, not a model problem

Hermes gave every agent a great brain and no reliable way to talk to the others. Here's what broke, what I evaluated instead, and why I still haven't fully committed.

  • automation
  • multi-agent
  • llm
  • tooling
Read →