Home / Products / CheriHire

AI product, Hiring

CheriHire

Hire the real thing. Every resume gets an authenticity score (is this fabricated?) and a fit score for the role, each opening into the evidence behind it. Screen hundreds of resumes in minutes, then run the rest of the hire in the same place.

Book a Demo Visit cherihire.com

Every candidate, two scores

Authenticity Is this resume fabricated?

Fit Suited to this specific role?

Evidence Strengths, gaps, flags

What it actually is

Screening is where hiring quietly fails. A recruiter with 400 resumes and a day to read them is guessing, and the resumes that read best are not always the ones that are true. CheriHire handles the first pass: resumes come in from uploads, folders, an email inbox or your careers page, get normalised, and are scored on two separate questions.

Authenticity asks whether the resume is fabricated, unsupported claims, keyword stuffing, impossible timelines, and no way to contact anyone who could confirm it. Fit asks whether this person suits this specific role. Both open into the evidence, so you can disagree with the score and see exactly why it said what it said.

Nothing is decided automatically. The ranking is decision support. A person still makes the call, and the tool is built to show its reasoning rather than hide it.

Three kinds of user

Recruiters, screen in minutes, run the whole hire in one place
Hiring companies, see exactly who your agency submitted, and where each candidate stands
Job seekers, get a resume health score, apply, and see a real match and shortlist chance

How the screening works

We would rather explain this than call it magic.

Ingest

Resumes arrive from uploads, a folder, an email inbox or your careers page, and are normalised into a consistent structure.

Score

Deterministic matching plus semantic comparison against the role, producing calibrated authenticity and fit probabilities.

Explain

Each score opens into its evidence, strengths, gaps and specific flags, so a human can agree or overrule it.

Hire

Interview rounds, offers, joining and referrals run in the same system, so the decision trail stays intact.

On the honest limits

The authenticity score blends explicit rule signals with an anomaly detector, and the rules carry most of the weight precisely because they are explainable. It flags resumes worth a closer look, it does not prove anyone lied, and it is not a background check. Treat it as a reason to ask better questions, not as a verdict.

Is it right for you?

A good fit if

You get far more applicants than anyone can properly read
You have been burned by resumes that did not match the person
You are an agency and clients want visibility into submissions
You want screening decisions you can justify afterwards

Probably not, if

You hire two or three people a year, the overhead will not pay for itself
You want fully automatic rejection with no human review; that is deliberately not how it works
You need formal background verification, that is a different kind of service

Common questions

No. It ranks and explains; a person decides. That is a deliberate design choice, not a missing feature. Automated rejection without review is both a hiring risk and, increasingly, a legal one.

It is a signal, not proof. Most of its weight comes from explicit, explainable rules (unsupported claims, keyword stuffing, impossible timelines, no contactability) blended with an anomaly detector. It tells you which resumes deserve a harder look. Anyone claiming certainty about this is overselling.

It can run with a local model as the primary engine, which keeps candidate data inside your environment. Deployment options and data handling are part of the first conversation. Ask us directly and we will be specific about your setup.

Screening keeps working. The system falls back to its built-in heuristics rather than stopping, so a model outage slows the quality of scoring. It does not halt your hiring.

Yes. Job seekers can upload a resume for a health score, apply to open roles, and see their real match and shortlist chance. That is the same scoring the recruiter sees, rather than a black box.

Try it on a role you are hiring for

Bring a real job description and a batch of real resumes. That tells you more in twenty minutes than any demo dataset will.