@stolinski Were the problems the extra PRs merged, commits or new repos? Or were they just about how AI has impacted the whole development process and the new culture of shipping more even at a quality cost?
@mitchellh is pointing out uptime, no degradation or slowness.
@darraghcurran Before AI: May do layoffs to reduce cost per PR
After AI: Add claude code to every single person in the company and cost per PR will go down dramatically. Love it!
Our run-rate revenue has surpassed $30 billion, up from $9 billion at the end of 2025, as demand for Claude continues to accelerate. This partnership gives us the compute to keep pace.
Read more: https://t.co/XgSjL0And7
@big_duca Someone has to prompt the Claudes, talk to customers, coordinate with other teams, decide what to build next. Engineering is changing and great engineers are more important than ever.
@Avaltenea No creo que sea el trafico de los bot de AI, si puede ser codigo de AI que no esten revisando con la misma calidad. Si hay mucho mas PRs (que los bots son mas rapidos) los que hacen las reviews tienen mas para revisar y puede que bajen la vara. Igual #hablemosSinSaber
The Adolescence of Technology: an essay on the risks posed by powerful AI to national security, economies and democracy—and how we can defend against them: https://t.co/0phIiJjrmz
@mattpocockuk I tested Ralph without the progress.txt and it wasn't that good, the experience was much worse. But this approach I like it mainly because of the need to manually cull progress.txt