What is the Jev Cookbook?

A curriculum of Jupyter notebook recipes that teach typed decisions with Jev, TypeSafe AI's System One model: each recipe covers one bounded decision, starting from a first Choice question.

What counts as a published recipe?

A recipe is published once its notebook.ipynb exists in the repository, after which its catalog row links to the notebook. Rows still in progress are marked coming soon and link to a tracking issue when recorded.

Do the recipes run against the live Jev API?

The banner describes the project as offline first with live calls opt-in, and each published recipe states in its notebook and README which mode it ran in, such as synthetic fixtures, so you can tell how it was produced.

What do I need to run the recipes?

Python 3.10 or newer and a Jupyter environment, plus Jev access for any recipe that makes live calls. The repository is MIT-licensed and the progress badge showed 50 of 100 recipes published.

Who maintains the cookbook and how active is it?

The Jev-Engineering organization on GitHub: the digest shows MIT licensing, 24 stars, 4 forks, creation on 2026-10-06 and a push on 2026-10-15, with the generated badge and table showing what has shipped.

How does this differ from official Jev documentation?

It is a community curriculum, not official documentation: the catalog lists planned use cases, categories and levels that can be revised as recipes are built, and the digest truncates the note on sources and review.

Jev directory

CommunityJev Cookbooks

Jev Cookbook — jev-engineering

A community curriculum of Jupyter notebook recipes that teach typed decisions with Jev, TypeSafe's System One model. Each recipe takes one bounded Choice, Noul or Score decision over text or image state, spanning first classification through closed-loop control.

See highlights Open github.com

Stars: 24
Language: Jupyter Notebook
License: MIT
Reading time: 2 min
Published by: Community
Author: Jev-Engineering
Last verified: Oct 2026

jev-engineering/cookbook

Organization repository on GitHub

View on GitHub

Sixty notebook recipes for typed decisions with Jev: Choice, Noul and Score patterns from a first classification to closed-loop control

Forks: 4
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

This is one of several Jev listings for the same project. The canonical entry is Jev Cookbook — datawhalechina.

Tagscommunity github python cookbook jupyter

Works forJev for Classification Jev for Scoring Jev for Realtime control

Used inJev for Software & DevTools

Languages: Python
Capabilities: Open source

Highlights

Planned as 100 notebook recipes for typed decisions with Jev; the generated badge showed 50 published across 15 categories and 5 levels in the digest.
A row counts as published once its notebook.ipynb is in the repository; other rows are marked coming soon and link to a tracking issue when one exists.
Framed around Jev as TypeSafe AI's System One model: send a state plus typed questions, get structured answers your code can branch on, sort by and route with.
Each published recipe states which mode it ran in, such as synthetic fixtures; the banner describes the project as offline first with live calls opt-in.
MIT-licensed Python 3.10+ notebooks; 24 stars, 4 forks, created 2026-10-06, last pushed 2026-10-15.

Watch out

README reviewed 2026-10-17 and truncated inside its status note, so the full recipe list, sources and per-recipe review states are not in the digest. The badge and recipes table are generated by the project, the catalog can be revised as recipes are built, and most planned recipes are still coming soon.

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.

Retrieval with Decisions: Filter, Rerank, and Verify

Oct 12, 2026

Retrieval is a sequence of bounded judgments. How a decision model slots into passage filtering, semantic find, reranking, and citation checks, and what stays in deterministic code.

Jev Cookbooks#retrieval#reranking#rag

Read article

One Request, Many Questions: Batching, Fan-Out, and Composite Scoring

Sep 12, 2026

Jev bills input tokens and gives output away, and questions in one request run in parallel. The design consequence is simple: ask more per call, then decide in code.

Jev Cookbooks#batching#patterns#search

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai