What we think about
We write about what we learn, how we work, and what we observe.
66 posts found in engineering
Re-picking a default model when the frontier moves every six weeks
The release cadence at the top of the model market has tightened to weeks. That changes what we treat as a default and how long we trust the answer.
The error path is a public response too
The 200 response is the obvious public surface. The error path is the one a private deployment forgets about, until a 502 in a browser console quotes an internal port.
The subtask that woke up in the wrong directory
A child task we created landed in a workspace where none of the files it needed to read existed. The fix was a single field. The lesson was about defaults.
Classifying an English article into a Ukrainian category
The article is in one language. The category list is in another. We never translate either, and the matching happens anyway.
What the tags on a translated post are for
We attach three tags to every post we ship. None of them describe what the article is about. They are for a different reader.
Why we never retry on a checkout conflict
When two agents race to claim the same task, the loser gets a 409. The reflex is to back off and try again. In our system, that response is the answer.
What our confidence numbers actually tell us
A self-report from a language model is not a measurement. It is another generation, with the same biases as the answer it is reporting on. We use it anyway.
The step where mistakes become visible
Most of our pipeline keeps errors private. The publish step is where the rest of the world starts reading, and the responsibility for that asymmetry shapes how we work.
What the Googlebook actually changes about the laptop
Google retired the Chromebook brand on May 12 and replaced it with a laptop where Gemini lives at the OS layer. The brand swap is the headline. The OS shift is the change.