Mastering TypeSafe Jev
Companion code for a hands-on book on building decision engines with hosted Jev and open-source Kev, with two capstone projects.
Overview
Mastering TypeSafe Jev is the companion repository for a short, hands-on book about moving the small judgments inside software, such as routing, flagging, safety checks and next operational steps, out of brittle if statements and slow LLM prompts and into a typed System One decision model. Its thesis is 'code computes, LLMs create, Jev decides, fresh state proves the result'. Every chapter ships working Python and a mini-project.
Two capstones share one decision client. Part I builds an incident and fintech triage engine on hosted Jev, asking whether fraud is suspected, an urgency score from 1 to 5 and a dispatch target, while code handles eligibility, fraud overrides, confidence gates, refund limits, a circuit breaker and fresh-state verification. Part II builds an industrial IoT system on self-hosted Kev, deciding intervention urgency, an operational action (monitor, inspect, schedule maintenance, transfer load or request shutdown), whether a standard workflow applies and severity, with safety trips kept authoritative in code. The book has 13 chapters, two appendices, 60 diagrams, two installable packages and 48 automated tests.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Is fraud suspected? | — |
| Choice | Where should this report be dispatched? | tier1_support · secops_agent · refund_pipeline · manual_escalation |
| Choice | What operational action should be taken for this motor? | MONITOR · INSPECT · SCHEDULE_MAINTENANCE · TRANSFER_LOAD · REQUEST_SHUTDOWN |
| Score | How urgent or severe is this (1-5)? | 1 · 2 · 3 · 4 · 5 |
How it works
- Read a chapter and run its mini-project
- Build a shared decision client
- Capstone 1: Jev triage with code-owned gates and fallbacks
- Capstone 2: Kev industrial decisions with safety partition
- Verify results against fresh state
Strengths
- Working code in every chapter
- Covers both hosted and self-hosted models
- Production-shaped patterns such as circuit breakers and audit trails
- Automated tests
Caveats
- Educational examples, not production systems
- Part I needs a TypeSafe key; Part II needs a Kev server
Getting started
Read the book at mastering-jev-and-kev-book.vercel.app and install the capstone packages from this repository.At a glance
- Type
- Article
- Status
- production
- Language
- Python
- Runtime
- packages and mini-projects
- Has code
- yes
- Has tests
- yes
Use cases and integrations
How curated lists describe it
Book: a hands-on engineering playbook for decision engines whose thesis is that code computes, LLMs create, Jev decides, and fresh state proves the result, with working Python and a mini-project in every chapter.
yibie/awesome-jev
Frequently asked questions
What does the book teach?
How to move small software judgments into typed decision models with code-owned safeguards.
Does it need TypeSafe?
Part I uses hosted Jev; Part II uses open-source Kev.
Is there tested code?
Yes, two packages with 48 automated tests.
Related projects
Gallery of small Python demos where Jev answers typed questions on realistic state and explicit policy decides the action or fallback.
AppprototypePythonAwesome list and guide to System One models (Jev, Laya and others) organized around a five-stage agent lifecycle, with comparison tables.
ProjectproductionOpinionated catalog of 84 Jev projects in 11 categories, each with a hand-written reason it is worth looking at.
ProjectproductionUnofficial collection of Jev state formats, question patterns, cookbook index and case studies, plus a ranked Top 30 of community builds.
ProjectproductionBilingual catalogue of 183 Jev applications in 11 categories, explaining how each works and what its demo actually proves.
ProjectproductionKorean course code for attaching Jev to Claude Code, Codex and OpenClaw: skills, hooks, routers and open-model comparisons.
ProjectproductionJavaScript