Solutions / Résumé screening

Résumé screening agent

Every incoming résumé is read against the actual role, scored, and ranked — with a written reason for each score, so the shortlist arrives explained rather than asserted.

HR Document processing Professional services
100 CVs
screened in 20–30 minutes
6–8 hours
of manual reading removed per role
Every score
comes with its reason in writing
Who it is for

Heads of Recruiting handling 50 or more résumés per role, where reading everything properly is impossible and reading only the top of the pile is how good candidates get missed.

Short answer

Each résumé that arrives is read against the role you actually opened, scored on the criteria you set, and ranked. 100 CVs are screened in 20–30 minutes instead of 6–8 hours, and every score carries a written reason a recruiter can agree with or overturn.

The problem

The pile is read in the order it arrived, until someone runs out of time.

01

Volume forces skimming

Fifty résumés at six minutes each is five hours of concentration nobody has. So the first twenty get read and the rest get a glance, and the order of arrival quietly becomes the ranking.

02

Screening standards drift

The same profile scored on Monday and on Friday gets different outcomes, and two recruiters on the same role disagree — not out of carelessness, but because the criteria live in their heads rather than on paper.

03

Good candidates go cold

The strongest people are interviewing in three places. A shortlist that takes a week to assemble is competing for candidates who accepted something else on day four.

How it works

From trigger to result, step by step.

01

Turn the vacancy into criteria

Must-haves, nice-to-haves, disqualifiers, and how much each weighs — written down once with your recruiters, per role. This is the step that makes the rest defensible.

02

Read whatever arrives

PDF, Word, a LinkedIn export, a scan, a two-page CV or a ten-page one. Structure is extracted first — experience, roles, dates, skills, gaps — so scoring works on facts rather than on formatting.

03

Score against the role, with reasons

Each criterion gets a score and a sentence citing the part of the résumé it came from. No opaque overall number: a recruiter sees why a candidate placed where they did.

04

Rank and route

The shortlist lands in your ATS with scores attached. Strong matches are flagged for immediate contact, borderline ones queued for human review, and clear misses kept with their reason rather than deleted.

05

Keep the recruiter in charge

Every score can be overturned in one click, and overrides are logged. Rejections stay a human decision — the system orders the pile, it does not close the door.

Before / after

What changes on the ground.

Today, by hand
×Five hours of reading per role, done in fragments
×The first twenty CVs get real attention, the rest a glance
×Two recruiters rank the same pile differently
×Shortlists take a week; strong candidates are gone
With the automation running
›100 CVs screened in 20–30 minutes
›Every résumé read against the same written criteria
›Each score explained, and overridable in one click
›The shortlist ready the day applications arrive
What you get

Delivered, not demoed.

›A criteria template per role, agreed with your recruiters
›A screening agent reading PDF, Word, LinkedIn exports and scans
›Scores with written reasons, written back into your ATS
›An override log showing where the model and your team disagree
›Documentation and a handover session — the system is yours
Built with

We build in your stack rather than moving you onto ours. The list below is what this solution most often connects to — other systems are a scoping question, not a blocker.

LLM (replaceable) OCR Greenhouse Workable Huntflow Google Drive Gmail / Outlook PostgreSQL n8n
Time to production4–6 weeks
Build priceFixed quote
First stepFree mini-audit
Honest limits

When this is not the right solution.

·If you cannot write down what makes a candidate strong for a role, this cannot score it. That conversation is part of the build, and some teams discover the vacancy itself was never defined.
·Ten résumés per role does not need screening automation — it needs an afternoon. We will tell you that rather than sell you a system.
·If you want automatic rejections without a human in the loop, we are not the right partner. Ranking is defensible; unattended rejection is not.

Questions we get about this one

It scores only the criteria you wrote down and shows its reason for each, which makes bias visible where a fast human read hides it. Name, age, photo and address can be masked before scoring. It ranks; a person still decides, and every override is logged so you can see where the criteria themselves are wrong.

Bring us the process that hurts.

The mini-audit is free: we take your version of this process apart and tell you plainly whether automating it pays. If it does, you get a scope and a fixed price.