Jev's API: Typed Answers With Confidence, App-Code Thresholds Like 0.6 for Reads and 0.95 for Writes
schwentker · x · 2026-09-17
The engineer-facing thread explains Jev's interface: many independent questions evaluated in a single call, returning typed answers with calibrated confidence and no text generation. The key design: the model doesn't select branches — the same call can gate reads at 0.6 and high-stakes writes at 0.95, with thresholds owned by application code and thus reviewable.
More from coding & agent
- Agent services: a monitorable sandbox-escape relay whose logs unlock after one week — cis_female · 2026-09-17
- SciML lead stacks Claude, Codex, Devin, Grok subscriptions to maintain libraries—and it's still not enough — ChrisRackauckas · 2026-09-17
- Quick Start: Building Your Own Claude Skills With Just a SKILL.md — technextpreneur · 2026-09-17
- Atlassian design lead built a Swift personal workbench to orchestrate dozens of local AI agents — davidhoang · 2026-09-17
- Nat Friedman's Muse impresses with sub-150ms responses so fast users suspect a glitch — manosaie · 2026-09-17
- Dev Vibecodes Procedural First-Person Hands, Releases Free Game Mod — TAbrodi · 2026-09-17