What it is
A Python reference harness for production-shaped use cases built around Jev, TypeSafe AI's System One model. It shows how to turn state into typed decisions with confidence-gated logic for developers building automation.
How it uses Jev
Each module assembles state from records, asks several narrow Choice, Score, or Noul questions in one call, reads probabilities, and applies thresholds and exact checks in code. The result is a UseCaseResult with decision, action_band, and actions; the caller performs the action.
Primitives:choicescorenoul
Technique worth stealing
Confidence-gated decision logic: use Jev probabilities plus code-computed checks to choose an action band.
Try it
Run the committed fixtures against the live TypeSafe API; see README for module names.
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-centricyes0.94
- Shows a System One patternyes0.80
- Handles uncertaintyno0.29
- Measuredno0.23
- Runnableno0.21
- Worth recommendingunclear0.59
- Model replicano0.07
- Problem scopescore on a 0–2 scale1.16
- About Jevyes0.97
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.