Start here: what Jev is for
If you came from the video, this is the JEV guide: the two workflows from the video and 18 more, each with a prompt you can paste into Claude Code.
Jev is TypeSafe AI's first "System One" model. It does not write text. It answers narrow questions about a situation you describe, in three shapes: pick one option from a list, give a score on a rubric, or say how likely a statement is to be true. Every answer comes back with a confidence number your code can check.
That makes it a good gatekeeper. Most of the calls inside a Claude Code workflow are small: which model should do this, is this email worth answering, does this test failure look flaky. You do not need your strongest, most expensive model for those. Let Jev make the small calls and spend Claude on the work that needs it.
Honest note:TypeSafe publishes response times of 70 to 500 milliseconds and input pricing of $0.042 per million tokens, with output tokens free. The "faster" and "cheaper" comparisons are TypeSafe's launch claims on their own workflow tests. How much Claude usage you save depends on how many of your tasks are small, so measure it on your own work.
Free, for your email
Get the full setup steps + the copy-paste setup prompt
The rest of this guide (5 more sections, 4 copy-paste prompts) opens right here. The setup prompt, one paste into Claude that sets it all up, comes by email only. Use your real address.
Unlocks: Set up Jev once · Workflow 1 from the video: route every task · Workflow 2 from the video: triage incoming leads · 18 more workflows · Guardrails
50,000+ follow where these guides come from. No spam, reply “unsubscribe” any time.