jev-for-engineers
Eight zero-dependency examples applying Jev to mechanical and electrical engineering: CAD routing, FEM triage, DFM, BOM alignment.
Overview
Jev for engineers offers eight minimal working examples of Jev for mechanical and electrical engineering work: CAD, CAE and CAM task routing, finite element result triage, design-for-manufacturing screening, bill-of-materials alignment, drawings and change control, and extraction that cannot hallucinate. Each example ends by taking a decision in ordinary Python, making the point that Jev supplies calibrated inputs while policy stays in reviewable code.
Every example was run against the live API on 17 September 2026. The README highlights a BOM case where two descriptions of the same screw in different languages scored 1.91 of 2 with 0.86 confidence (merge), while M6x20 and M6x25, one character apart, scored 0.01 (keep separate), something string similarity gets backwards; part numbers are invented. It also explains fanning out many questions in one call, citing TypeSafe's measurement of a 13-question briefing as 12.2 times cheaper and 10 times faster batched.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which tool (CAD, CAE, CAM) should handle this engineering task? | — |
| Score | Do these two BOM lines describe the same part? | — |
| Yes / no | Does this FEM result or drawing raise a concern?(inferred) | — |
How it works
- Prepare engineering text or JSON as state
- Ask named typed questions in one call
- Read probabilities and confidence
- Make the final decision in plain Python
Strengths
- Zero dependencies
- All examples run against the live API
- Shows cases where string matching fails
Caveats
- Sample data with invented part numbers
- Minimal examples, not a product
Getting started
Set TYPESAFE_API_KEY and run any example script with Python.At a glance
- Type
- Project
- Status
- prototype
- Language
- Python
- Runtime
- Standalone scripts
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
Eight mechanical and electrical engineering experiments using Jev for task routing, log checks and component selection.
logicrw/awesome-jev-projectsEight minimal working examples of TypeSafe's Jev (a System One model) applied to mechanical and electrical engineering: CAD/CAE/CAM routing, FEM result triage, DFM screening, BOM alignment, hallucination-proof extraction. Zero dependencies.
hellogumbo/awesome-jev
Frequently asked questions
What engineering tasks are covered?
CAD/CAE/CAM routing, FEM triage, DFM screening, BOM alignment, drawings, change control and extraction.
Were the examples run live?
Yes, against jev-latest on 17 September 2026.
Who makes the final decision?
Ordinary Python code using Jev's calibrated answers.
Related projects
Record linkage for researchers: write the match rule in plain English and Jev judges each candidate pair with a probability.
ProjectbetaPythonSmall dependency-free library that maps uploaded CSV columns to your schema with exact-name matching, then one Jev Noul per remaining pair.
ProjectprototypeJavaScriptCLI that normalises messy Japanese addresses to postal codes with rules and Japan Post data, asking Jev only for ambiguous steps.
LibraryprototypeTypeScriptPython re-style library that finds, replaces, splits and extracts text by meaning, with Jev judging each candidate span.
DemobetaPythonNext.js community playground extending the 110-example TypeSafe library with A/B comparisons, routing, extraction, games and simulations.
ServicebetaTypeScriptOfficial TypeSafe cookbook pattern: extract dates with typed questions and do validation and arithmetic in code.
Articleproduction