TypeSafe Jev · System One · Vercel AI Gateway

Jev tutorial: call TypeSafe Jev (System One) for typed decisions

typesafe-ai/jev returns Choice, Score, and Boolean answers with probabilities. High confidence → auto-suggest; low confidence → human queue. Suggestions only — never money or refunds.

Quick answers

What is Jev?

TypeSafe Jev is a System One evaluation model: it returns typed Choice, Score, and Boolean answers with probabilities — not free-form prose.

Choice / Score / Noul

Choice picks one named option; Score places state on an ordered rubric; Noul (boolean) estimates P(true). Confidence comes from the option distribution.

When to use a human

Low confidence → human queue with no default action. Money, refunds, and irreversible side effects stay human-owned — suggestions only.

Choice

Pick one named option. Returns the key, full probability map, and confidence.

Score

Place state on an ordered rubric. A level index — not an accuracy promise.

Thresholds

Keep act / confirm / human bands in your code. Calibrate on your labeled data.

Pack templates

Boundary: Gateway + Jev patterns only. No accuracy promises, no money/refund automation, no CMD Guard product. Gumroad link is a placeholder until the live listing is ready.

FAQ

What is TypeSafe Jev / System One?

TypeSafe Jev is a System One evaluation model on Vercel AI Gateway. It returns typed Choice, Score, and Boolean answers with probabilities instead of free-form text.

Do I need a self-built API relay?

No. This Jev tutorial teaches calling AI Gateway directly from your app or serverless function. A custom relay is out of scope and discouraged.

Does Jev make money or refund decisions?

No. Templates and examples only suggest routes or labels. Money and refund decisions stay with humans or other systems.