typesafe-jev CV screener
Local CV screening workbench that asks Jev fixed typed questions per CV and scores them with editable, free-to-rerun policy arithmetic.
Overview
This repository holds experiments in building software around a decision model. Its main project, cv-screen, is a local workbench for screening a single CV or a folder of hundreds. It asks Jev a small fixed set of typed questions about each CV, scores the answers with readable arithmetic, and shows a sortable shortlist with one verdict per candidate.
The role being hired for is a policy you edit in the app. Because judgments are stored separately from the scoring arithmetic, changing a weight, cap or question re-scores every stored candidate in about 20 ms at no API cost. The project README reports what it tests about the model and the measured results, including two bugs caught by its test data. It is MIT licensed with tests. Hiring decisions carry fairness and legal considerations that a tool like this does not resolve.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Fixed typed questions about each CV (e.g. does the candidate meet a requirement?)(inferred) | yes · no |
How it works
- Point the app at a CV or folder.
- Jev answers a fixed set of typed questions per CV.
- Plain arithmetic scores answers against the editable policy.
- Review the shortlist; edit the policy to re-score instantly.
Strengths
- Policy changes re-score for free in about 20 ms.
- Scoring logic is readable arithmetic.
- Runs locally; includes tests and measured results.
Caveats
- Exact questions are in the sub-project README, not summarized here.
- CV text is sent to the hosted model.
- Automated screening raises fairness and compliance concerns.
At a glance
- Type
- Project
- Status
- prototype
- Runtime
- local app
- Has code
- yes
- Has tests
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
Recruiting: screens a folder of CVs with Jev typed judgments against an editable policy, re-scoring candidates for free when the policy changes.
yibie/awesome-jevScreen a folder of CVs with the TypeSafe Jev decision model: typed judgments, an editable policy, free re-scoring.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev decide?
Answers to a fixed set of typed questions about each CV.
Does changing the policy cost money?
No, stored judgments are re-scored locally in about 20 ms.
Is it production hiring software?
No, it is an experiment.
Related projects
Resume review app and MCP server where a Cloudflare Worker asks Jev typed questions and composes a JevScore, generally or per job.
AppbetaTypeScriptDaily job-search pipeline that triages and scores public postings against a resume with Jev, then applies via a Grok Bot webhook.
ServiceprototypeTypeScriptCV review app where Jev makes traceable document-level judgments on clarity, contribution and results, optionally against a job posting.
ProjectprototypePythonResume-to-JD screening study (Chinese) with 5 Noul gates, 4 Score dimensions and a routing Choice per request, hardened over three versions.
ProjectresearchSmall demo that scores resumes for first-pass fit and interview recommendation with TypeSafe Jev.
DemoprototypeNext.js resume review app where Jev returns typed scores for overall fit, individual criteria and each experience bullet.
AppprototypeTypeScript