Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

jev-intro-to-decision-models

Companion code for the ZazenCodes course on decision models: primitives notebook, use-case examples and a Pi model-router build.

Projectproduction
View on GitHub ↗

Overview

Jev: Introduction to Decision Models is the companion repository for a paid ZazenCodes course on using fast, non-autoregressive decision models in AI agents. Lesson 3 is a notebook on Noul, Choice and Score with a latency and cost benchmark against GPT-6 Sol and Claude Opus 5.5. Lesson 5 works through one example per use-case category using public datasets and generated games, including TextWorld. Lesson 6 live-builds a /jev slash command for the Pi coding agent in which Jev picks the model for a prompt, with build prompt and reference solution.

It uses uv and needs a TypeSafe key; the benchmark also needs OpenAI and Anthropic keys, and Lesson 5 an OpenAI key. Links from other lessons are collected in LINKS.md.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich model should handle this prompt?—
OtherNoul, Choice and Score examples across use-case categories—

How it works

  1. Clone and uv sync
  2. Fill .env with keys
  3. Run lesson notebooks
  4. Build the Pi /jev model router

Strengths

  • Structured course material
  • Includes latency and cost benchmark
  • Hands-on Pi integration

Caveats

  • Full course is paid
  • Several provider keys needed for all lessons

Getting started

`git clone https://github.com/zazencodes/jev-intro-to-decision-models.git && cd jev-intro-to-decision-models && uv sync`, then copy .env.example to .env.

At a glance

Type
Project
Status
production
Language
Python
Runtime
Jupyter notebooks and Pi extension
Has code
yes

Use cases and integrations

How curated lists describe it

Frequently asked questions

What is this?

Code for a course on decision models with Jev.

Is the course free?

No, it is a paid course; the code is in this repo.

What will I build?

Notebooks on primitives and use cases, and a Pi model-router slash command.

Related projects

  • jev-handson

    Korean course code for attaching Jev to Claude Code, Codex and OpenClaw: skills, hooks, routers and open-model comparisons.

    ProjectproductionJavaScript
  • Learn Jev end to end

    Free 12-notebook course building 13 agent tools with Jev as the fast decider and an LLM as the slow writer, for under $0.20.

    ProjectproductionPython
  • jev-agent-harness

    Beginner manual, video script and three small examples for building an agent harness with Jev via langchain-typesafe.

    ProjectprototypePython
  • jev-engineering-zh

    Chinese translation of 'Jev Engineering for coding agents', design notes on a harness where Jev makes every per-turn decision.

    Projectdiscussion
  • Jev vs GPT-6 Astra: when to use each

    Vercel's decision guide on when to use a System One model like Jev versus a frontier model like GPT-6 Astra.

    Articlediscussion
  • Ten Levels of Jev

    Thirty worked Jev uses in ten levels, from a smart if statement to a pi coding agent that writes its own Jev questions.

    ProjectbetaTypeScript