# heyjev.ai

> Small public experiments with Jev, TypeSafe's System One model. Powered by TypeSafe & nullplatform. An independent experiment. Not an official TypeSafe product.

**Notice.** Entertainment only. Not advice, not a deployment gate. Do not use this to decide, approve, block or automate a real release. No SLA: the API and its format can change or disappear without notice. See /terms.

## What is here

- [Hey Jev, should I deploy?](https://heyjev.ai/shouldideploytoday) — describe a deploy in plain English; Jev answers 18 typed questions about it in one request; the site turns the probabilities into a verdict (SHIP IT, MAYBE MONDAY, GO TOUCH GRASS, …), three reasons, the real latency, and every question's probability distribution. Markdown version: [https://heyjev.ai/shouldideploytoday.md](https://heyjev.ai/shouldideploytoday.md)
- [Skill for agents](https://heyjev.ai/skill) — how to call `POST https://heyjev.ai/api/ask` from an LLM or coding agent.
- [Terms](https://heyjev.ai/terms) — personal, non-commercial use; not advice, not a deployment gate; no SLA. Markdown: [https://heyjev.ai/terms.md](https://heyjev.ai/terms.md)
- [llms.txt](https://heyjev.ai/llms.txt) · [llms-full.txt](https://heyjev.ai/llms-full.txt) · [Health](https://heyjev.ai/health)

## What Jev is

Jev is TypeSafe AI's System One model. Instead of generating text, it takes a `state` (any text or JSON) and a set of typed questions (noul = yes/no probability, choice = one option with the full distribution, score = ordered levels) and returns calibrated probabilities for all of them in a single request, typically in a few hundred milliseconds. Code decides what to do with the numbers. Docs: https://docs.typesafe.ai

## Why this site exists

We use Jev in a real deployment approval gate at nullplatform. This site gives it a less serious job, so people can see what typed AI decisions look like: the exact request, the raw probabilities, and the rules that turn them into a verdict.