Jev in production › Document and record classification

Resurface

Jev answers resume-screening questions per quoted resume line, reaching 99% agreement with Claude Sonnet 5.5 (author).

Open on GitHub ↗

99%measured against a baseline, as published by the source
Use
Document and record classification
Industry
Data and analytics
Form
Open-source tool
Stage
In production
Listed
2026-10-01
Found via
discord
Repository
dolevhayut/resurface
Stars
0
Forks
0
Last push
2026-10-03
Language
TypeScript
License
MIT
Resurface screenshot
docs/launch/poster-hook.jpg in the dolevhayut/resurface README, MIT; shown from GitHub.

The README opens with

Language models write. Classification models decide. Turn the candidate database you already have into evidence-backed shortlists. An LLM writes the screening questions once per job. A classification model answers them for every resume, and quotes the exact line that proves each answer.

Zero setup. This repo is the open-source version: the full method, to run yourself. To score real resumes without running anything, use Resurface Cloud, the hosted API: sign up, create a key, send a job description and a resume file. There are no model keys to manage, no database to run and no PDF parsing to fix.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for document and record classification