Articles tagged “research”
4 articles on research.
How We Verify Agent Work
A builder that reports "tests green" from a runner that collected zero tests is not lying, exactly. It is describing output nobody checked. Here is the layer of our stack that does the checking, and the six times it was wrong.
GPT-6 Astra vs Claude vs Llama: We Tested the New Model on Real Work
GPT-6 Astra is days old and already everywhere. So we gave it three jobs our stack actually does, put Claude and Llama 4 on the same briefs, and had a fourth model score them blind. The result was the same on all three.
We Measured Which Claude Model Should Write Code — the Answer Wasn't About the Model
Everyone picks their coding model by price per token. We spent a day measuring what actually costs money, and price per token turned out to be the least useful number on the table.
Our Full Agent Stack, Explained
One founder. Fourteen terminal sessions. A phone. The whole system we use to design, build, review and ship software with AI agents — and the one idea everything else is built on.