I build AI-native products, apps & automations. Writing about tech, AI and things that industry refuses to admit | Feature your brand: →brandmysetup.lol
A practical deep dive into System One models and Jev, TypeSafe’s new decision-focused model. We’ll break down how Jev works, how it differs from traditional LLMs and structured outputs, why calibrated
I was wondering what if you could take some behaviour of the recent stronger frontier AI model and give it to a smaller model ? I'm not highlighting to copy the whole intelligence and weights, which
What the crawl-to-refer ratio measures and what it misses, why a bot’s request costs more than yours, and why robots.txt was never going to fix it. An explanation of measurement and infrastructure,
Most developers treat it like autocomplete. Underneath, it's an autonomous agent runtime with features 90% of users have never touched. I went through the full source-level leaked architecture and
That's what i used to think. For years, I genuinely believed MacBooks were overpriced status symbols shiny alluminum boxes people bought to feel productive rather than actually be productive. In my
Most people still think of X as a place for hot takes, memes, and arguments. This perception is totally outdated for now. Quietly without a big rebrand campaign or flashy announcements X is reshaping