Jev technical blog and early access
The end of Diogo’s launch thread links to the official technical blog and early-access entry point for further reading.
Jev Skill Suggestion benchmarked against Claude models
Benchmarked Jev Skill Suggestion against Claude models with 26 skills, 10 prompts, 3 runs each: Jev 90% accuracy at 330ms median; Haiku 80% at 1.24s; Sonnet/Opus/Fable 100% at 2.2–4.5s. Jev trades a bit of accuracy for 4–14× faster decisions.
I benchmarked Jev Skill Suggestion against Claude models with 26 skills, 10 prompts, and 3 runs each - Jev: 90% accuracy, 330ms median - Haiku: 80%, 1.24s - Sonnet/Opus/Fable: 100%, 2.2–4.5s So Jev trades a bit of accuracy for a 4–14× faster decision, while keeping the skill
· 67 likesOpen on X
More like this
Getting started with Jev: waitlist and Agent Skill
A community guide walks developers through requesting access and installing the TypeSafe Agent Skill.
Reverse-engineered a Jev-like architecture with open-source training repo
The author says they reverse-engineered a Jev-like architecture from its type and shares a GitHub repo for training your own jevlikes. Link
