Matija Grcic
@matijagrcic • 1,782 subscribers
Father of two. speciality coffee / whisky / hiking / AI
Videos
Agree with DHH that outsourcing your understanding to AI leads to a loss of competence and you then become less capable in the long run. "Whenever I try to use it in vibe coding form, at least when I'm in a new domain I haven't fully internalized, I can feel the competence dripping out of my fingers. And this is, when I started building Omachi, I started writing a lot of bash. ... And I caught myself asking the AI for the oomph time how to do a certain if conditional. You're like, why have I not internalized this? I have not internalized this because I've outsourced it to the AI. Is that better? Am I better off now? Or am I as delusional as the teachers who thought that by the time we got calculators, students no longer need to know their timetables? No. You will actually turn yourself into an idiot if you can't do seven times seven real quick in your head."
Matija Grcic45,680 Aufrufe • vor 8 Monaten
I just listened to a breakdown of Gemini jump from 2.0 to 2.5, and here's the gist. The team did not rely on lab tests alone. They scraped real user feedback from X, turned those "this broke" moments into living evals, and keep appending new edge cases with every release. It's a community driven QA loop: our failures become their benchmarks, which become our better results. If you're a creator or PM, this is a template: public feedback → evals → release → repeat. Great work Logan Kilpatrick Nicole Brichtova and whole team.
Matija Grcic62,249 Aufrufe • vor 1 Jahr
Keine weiteren Inhalte verfügbar