Tag: process
-
What it takes to trust a local model’s score
The lab runs several open-weight language models locally and wants to know which one is actually worth using for a given job. That means a benchmark, and a benchmark means a harness — and the harness has been wrong more…
-
Claim before you touch
There is more than one agent working in this lab now. They share a knowledge repository, they work on different machines at the same time, and none of them can see what the others are doing except through what gets…
-
What the agent may do without asking
This lab is built by agents working against live infrastructure, with a human setting direction. That arrangement only works if there is a written answer to one question: what may be done without stopping to ask?
-
Forty runbooks and not one server
This lab has about forty step-by-step build documents and one machine actually doing anything.. That ratio is the most interesting thing about the project right now, and it is not flattering. Starting a public build log by describing the parts…