Evally
Recruiting

Resume screener

Check each resume against the bar you define, with the supporting line quoted.

You write the requirements; each resume is checked for evidence against them, not keywords. A skill counts only if the resume shows it used in real work. Every verdict quotes the line that backs it up, and borderline candidates go to a person.

Hiring decisions affect people's livelihoods and are regulated in many places (for example New York City's automated employment decision rules). Use this to organize evidence for a person who makes the decision, never to reject candidates automatically, and never add questions about age, gender, ethnicity, religion, disability, family status or other protected traits.

Try it live

Real calls to Jev. Pick an example or paste your own.

Examples
200
290
No verdict yet

Pick an example or paste your own input, then run it.

Under the hood

Every question and rule, in the open. Fork it to change any of them.

What it asks

Reads requirements and resume, then asks 6 questions in one call.

  1. Q1Meets every must-haveYes / no quotes evidence

    Does the resume show every must-have in the requirements? (Look for evidence, not keywords: a skill counts only if the resume shows it used in real work.)

  2. Q2Experience levelPick one

    How does the candidate's experience compare with what the requirements ask for?

    BelowMatchesAboveUnclear
  3. Q3Relevant recent workScore

    How relevant is the candidate's most recent work to the role described in the requirements?

    1. 0Unrelated
    2. 1Adjacent
    3. 2Relevant
    4. 3The same kind of work
  4. Q4Shows resultsYes / no quotes evidence

    Does the resume show measurable results (numbers, shipped projects, outcomes), not just responsibilities?

  5. Q5Nice-to-havesScore

    How many of the nice-to-haves in the requirements does the resume show?

    1. 0None
    2. 1Some
    3. 2Most or all
  6. Q6InconsistenciesYes / no

    Does the resume contain claims that contradict each other (for example overlapping full-time roles, or dates that don't add up)? (Career gaps are not inconsistencies.)

How it decides

Rules run in code, top to bottom. The first one that matches gives the verdict.

  1. 1
    If Inconsistencies ≥ 0.7Human reviewreview
  2. 2
    If Meets every must-have < 0.3Below the barreview
  3. 3
    If Fit ≥ 70Advance
  4. 4
    If Fit ≥ 45Human reviewreview
  5. ·
    DefaultBelow the bar
Fit

A 0-100 score weighted from Meets every must-have ×4, Experience level ×1.5, Experience level ×1.5, Relevant recent work ×2, Shows results ×1.5, Nice-to-haves ×1.

Call it from your code

Every template runs as-is with an API key, no setup: POST /api/v1/judges/resume-screener/run.

curl -X POST https://tryevally.com/api/v1/judges/resume-screener/run \  -H "Authorization: Bearer $EVALLY_API_KEY" \  -H "Content-Type: application/json" \  -d '{  "input": {    "requirements": "Senior backend engineer. Must have: 5+ years building production services in Go or Rust; owned a system handling 1k+ requests per second; on-call experience. Nice to have: Kafka, Kubernetes, payments.",    "resume": "Staff Engineer, Paystream (2021-now). Built the Go ledger service handling 4,000 transactions per second; cut p99 latency from 900ms to 120ms. Led migration to Kubernetes. Primary on-call for payments for 3 years.\nSoftware Engineer, Cartly (2017-2021). Built order pipelines on Kafka in Go."  }}'

Seen in the wild

Community builds on Jev that use this pattern. Not affiliated with Evally.

More in Recruiting