3 post(s)
One Jev call picks a skill from 141 candidates, or says none applies: 90.8% first-answer correct against a strong generative model's 93.8%, at 1/9 the cost and 3.6× the speed. Two rules came out of the measurement: give the Choice a none option, and judge models are a good fit for routing.
The question came up while we were wiring the desktop to the agent: why are we hand-rolling a gRPC protocol when the Agent Client Protocol (ACP) already exists for exactly the "UI talks to an agent" problem? It was a…
A chat loses context. future-loop turns "keep an eye on this for a week" into a durable goal — a todo graph, human gates, per-step evidence, and a verifiable definition of done that survives sessions, restarts, and parallel workers. This is how the control plane works, and how it carried the Matilda tiling run.