Jev by TypeSafe AI skips token generation entirely, scores decisions in one parallel query
js_craft_hq · x · 2026-09-22
Jev from TypeSafe AI is drawing attention as a departure from text prediction. Instead of generating tokens, it only makes fast decisions.
- How it works: you define a schema of questions and possible answers, send in your app state, and it returns one answer per question with an attached probability.
- Parallel scoring: every question is scored in a single query, finishing in a fraction of the time an LLM needs to write out its response token by token.
- Use cases: replacing brittle if-statements over messy input — support ticket routing, lead scoring, field extraction, validating LLM answers.
It's pitched as a decision-only layer alongside LLMs, trading generation ability for low latency and calibrated outputs.
Related event: Ex-OpenAI researcher launches decision model Jev with bold speed claims(36 posts)→
More from coding & agent
- Sosumi Eval: Apple Docs Access Lifts Coding Agent Pass Rates from ~33% to 94-98% — JordanMorgan10 · 2026-11-12
- CypherGuard fixes structurally wrong LLM Cypher queries, hitting 88.5% on CypherBench — JeremyCMorgan · 2026-11-12
- Building an End-to-End Hedge Fund in Python with Claude Code: A 5-Step Workflow — mdancho84 · 2026-11-12
- Jeff Dean: comfortable with engineers spending up to $10M/year on AI tools — deedydas · 2026-11-12
- Developer builds VladOS, a personal life dashboard, by just describing it to an AI agent — haltakov · 2026-11-12
- Hyperbench launches AI agents for hardware/software co-development built on a unified system graph — ycombinator · 2026-11-11