What it is
A Python sandbox for developers that turns TypeSafe Jev judgments into testable issue-routing proposals, with labeled replay, an offline demo, and Kev support.
How it uses Jev
Jev answers Choice (investigation area), Noul (explicit reproduction steps), and Score (observation specificity). Policy gates confidence and reproduction to propose a specialist queue or review fallback; the Score hint never changes the queue.
Primitives:choicenoulscore
Technique worth stealing
Replay labeled judgments to compare thresholds without another model call.
Try it
python3 -m jevidence demo
Evidence
Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.
- Jev-centricyes0.92
- Shows a System One patternyes0.93
- Handles uncertaintyyes0.80
- Measuredno0.06
- Runnableunclear0.47
- Worth recommendingyes0.61
- Model replicano0.11
- Problem scopescore on a 0–2 scale1.23
- About Jevyes0.91
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 23 Sept 2026.