Typesafe AI on Jev's sweet spot: semantic judgment in the model, numbers on the CPU
schwentker · x · 2026-09-27
Part 2 of the Jevathon field notes: @allietheicon of Typesafe AI on where Jev fits.
- "If it's vibes, now we're getting somewhere": pure number crunching belongs on a CPU; semantic judgment is what Jev is for.
- Key clarification: Jev is not agentic — it does nothing until code puts it in the loop.
- A useful boundary-setting take on how to deploy cheap judgment models in engineering.
More from coding & agent
- François Fleuret asks: are there programming languages tailored for LLMs? — francoisfleuret · 2026-09-27
- AI Can't 'Read the Room': The Hard Problem of Selective Info Sharing in Enterprise AI — devanshmehta · 2026-09-27
- antirez Builds a Real VM to Run Another World Intro on ZX Spectrum 48k — antirez · 2026-09-27
- A 'new type of mind' trained without natural language writes kernels at up to 66x Triton speed — repligate · 2026-09-27
- ThePrimeagen joins dhh's Omarchy Core to lead Agentic QA with new Oligarchy harness — nicolascraske · 2026-09-27
- ChatGPT Astra auto-creates a proxy email to connect Google Docs — 0xsachi · 2026-09-27