A team says it ships 30x more code without reading it
poteto previewed slides from an upcoming AI Engineer talk.

poteto (@poteto) shared slides from an upcoming talk at AI Engineer about how their team ships far more code than before.
They say output is up 30x without reading the code, and that performance is as good as or better than at launch, while noting there is a lot left to improve.
The post drew about 1,200 likes.



