Skip to content

Jev directory

Jev's Garage

A workshop of small, inspectable Python experiments for TypeSafe System One models: ten critical bays, from fraud screening to emergency dispatch, each giving Jev a realistic state while deterministic policy keeps authority over the next action.
Stars
1
Language
Python
Reading time
2 min
Published by
Community
Author
jgalego
Last verified
Oct 2026

Highlights

  • Ten critical bays from fraud screening to emergency dispatch, each following state -> Jev -> typed result -> deterministic policy -> action or safe fallback.
  • Jev supplies judgment while the application keeps authority; critical demos never let the model directly perform an irreversible side effect.
  • The gallery has Input, Signals, Visuals and Decision tabs, validates JSON before enabling Run Jev, and revalidates on the server before the live call.
  • Signals move from red through amber to green by certainty; charts support + and - zoom, 1:1 reset, mouse-wheel zoom and drag-to-pan.
  • Runs with uv sync --locked --all-groups, a .env holding TYPESAFE_API_KEY, and just gallery; if port 8000 is busy, use just gallery --no-browser --port 8080.

Quickstart

bash
uv sync --locked --all-groups
cp .env.example .env
just gallery

Watch out

Grounded on the truncated README and GitHub metadata reviewed 2026-10-15; the critical bay table is cut off inside the AI output verification row, so the described list may extend further. The repo lists no license and publishes no accuracy figures; the demos are illustrative and were not run for this entry.

Questions about this entry

Short answers drawn from the linked project or page.

From the Jev guides

Original write-ups on what this entry is for.

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai