Small decision models

yibie/

awesome-jev

Curated index of public projects using Jev, TypeSafe AI's typed decision model, organized by application domain.

What’s new here

This list covers one model’s adoption surface. Every entry must use Jev, or a documented port or derivative, for a concrete decision task, not just resemble the pattern. That inclusion rule makes the list a field guide to which decision patterns transfer across industries rather than a generic AI tools directory.

What it does

The repo aggregates public projects, integrations, and discussions built on Jev, TypeSafe AI’s System One model for typed decisions. Jev takes unstructured state plus a typed question and returns a typed answer (Choice, Score, Boolean, or Noul) with a confidence value. The list organizes entries into 15 categories: Classification & Routing, Verification & Guardrails, Agent Decisions, Evaluation & Benchmarking, Infra/SDKs/Integrations, and ten others. A separate Related Practices category captures X threads, Reddit discussions, and interviews that describe real usage without a standalone artifact.

The README is auto-generated from per-category markdown files via a build script. Contributors add entries to category files and rebuild. The repo also tags entries by the coding agent they target: Claude Code, Pi, Codex, Multi, and others, with counts per agent.

A notable section is the curation warning: inclusion means only that an entry satisfies the rules. The README explicitly flags bulk same-day submissions sharing a scaffold as unproven, and gives readers a five-item checklist (does the code call the Jev API, is there a runnable check, do the numbers have a source, how much is code vs. prose, is there a license) to run before adopting anything.

Who it’s for

Engineers evaluating Jev for routing, classification, guardrails, or scoring who want to see how others have structured the decision loop in practice. Also useful for people tracking which confidence thresholds and question types transfer across domains.

Try it

Browse the category files directly or read the generated README. Submissions go through pull requests against the category markdown files, then a rebuild of the README via scripts/build-readme.py.

How mature is it

Created September 2026, last pushed October 2026. 2,260 stars, 334 forks, 4 open issues and 3 open pull requests. At least 100 contributors and 100 commits in the last 90 days (both capped by the API). No releases, no license.