CaveJev: forcing the Jev judgment model to speak word by word like an LLM
_rchaves_ · x · 2026-09-18
Developer Rogerio Chaves introduced CaveJev: he gave Typesafe's judgment model Jev—a model that returns typed verdicts rather than text—a caveman's vocabulary and forced it to choose the next word like an LLM, letting it "talk." His punchline: "why waste time, say lot word when few jev do trick?" Measured results and benchmarks live in his follow-up thread and the jev-experiments repo.
Related event: Dev Forces Judge Model Jev to Talk at 3 tok/s(3 posts)→
More from Fun
- New academic spam: single-authored papers cold-pitching ARR service contributions — anmarasovic · 2026-09-18
- User on running a Grok bot as 'chief of staff': love it, but it has a lot to learn — RachelVT42 · 2026-09-18
- Restaurant ads are now all AI-generated, and reality can't match — Yamapama · 2026-09-18
- One prompt with Astra turns a Mac desktop into an interactive aquarium that reacts to your mouse — chaseleantj · 2026-09-18
- One-prompt WaveRunner game demo shows off Astra's game generation chops — Dimillian · 2026-09-18
- "Stop sending me Slacks obviously written by Claude" — AI-flavored messages become a workplace pet peeve — alexgoughcooper · 2026-09-18