Four AI actions. One workflow. Reasoning you can defend.
CertAIn doesn't give you a score. It gives you a paragraph a recruiter can take into a hiring-manager meeting, a candidate interview, or a compliance review — and stand behind.
Every candidate gets a position — and a plain-English why.
What it does
You upload a JD and a stack of resumes. CertAIn reads each resume against the job description, the company profile, and any intake notes, then returns every candidate ordered by fit — with one paragraph of reasoning per candidate explaining where they sit and why.
What the output looks like
#3 of 47 — Strong fit with one gap. Candidate has seven years of B2B SaaS account-executive experience selling into mid-market, which matches the core of this role. She's closed deals in the $50K–$250K ACV band — below the $500K+ range you've set as target — so expect a ramp on deal size. Two things to probe in the interview: her experience managing a multi-threaded evaluation with procurement, and her familiarity with the vertical (healthcare) you're selling into now.
That is the output. Not "Score: 82." Not "Match: strong." A paragraph your recruiter can read to a hiring manager without translating.
How it differs
Keyword matchers count how many phrases from the JD appear in the resume. CertAIn reads the resume the way a senior recruiter does — weighing seniority, context, trajectory, and gaps. If the JD says "SaaS sales" and the resume says "selling software subscriptions to enterprises," the keyword tool misses it and CertAIn doesn't.
Regulatory fit
Every ranking output is a recommendation to a human — never an automated decision. That's the architectural line that keeps you on the right side of NYC AEDT, Illinois AIVIA, and EU AI Act Annex III. The reasoning paragraph itself is the audit artifact regulators want to see.
Targeted questions, grounded in what this person actually did.
What it does
Pick a candidate. CertAIn generates interview questions specific to that person's resume, this role's requirements, and — if you've added it — the intake notes from your hiring-manager conversation. Not generic "tell me about yourself." Questions about the specific project on page two of their resume that maps to the exact skill the role needs.
What the output looks like
On the 2024 pricing-migration project at your last company — you led the rollout of a usage-based model across 400 enterprise customers. This role owns a similar migration next quarter. Walk me through how you staged the cutover, how you handled customers who pushed back, and what you'd do differently if you had to start over. (Probes: change-management depth, customer-escalation instincts, lessons-learned frame.)
How it differs
ChatGPT can generate interview questions. It can't generate questions that cite a specific line on a specific resume tied to a specific JD requirement — because it doesn't have the workflow. CertAIn does.
Time saved
Senior recruiters spend around ten minutes prepping each interview and run twelve to twenty-five interviews a week. Interview Prep turns that into a one-minute review of pre-written, resume-grounded questions. Provenance: 2026-04-17 discovery call, Jordan, head recruiter — "Spends ~10 min per interview trying to prep and build questions. Doesn't have the time to prep properly."
A deeper, structured read — once a candidate is worth it.
What it does
Evaluation is the full work-up on a single candidate: a multi-section write-up covering fit against the JD, signal and concerns, strengths relative to the role, risk areas, and what the next conversation should cover. Heavier than Ranking; designed for the top of your pile, not the whole pile.
What the output looks like
A structured document, not a chat reply. Sections include: summary fit, experience map against JD requirements, notable strengths, concerns to probe, recommended next step. Exportable. Auditable. The kind of write-up a head of talent can forward to a hiring manager without cleaning up.
How it differs
An evaluation is the difference between "this candidate is 82% match" and "this candidate has done the exact work the role requires, twice, at smaller companies — the open question is whether they can operate at your scale." That's the work a senior recruiter does today in their head in thirty seconds. CertAIn does it in full, in writing, repeatable.
Regulatory fit
Evaluations are the clearest bias-audit artifact CertAIn produces — structured, exportable, demographic-free by design. The bias_audit export pulls Evaluation outputs by candidate × JD for any date range.
A structured read of what was said — and what was missing.
What it does
After an interview, paste in your notes or the transcript. CertAIn returns a structured analysis: what was covered, what was evidenced, what was asserted without evidence, what the candidate avoided, and the signal that decides next steps.
What the output looks like
Covered: prior pricing-migration work (matches Prep focus), team size managed, churn-handling approach.
Evidenced: specific customer example with named companies and outcomes; crisp explanation of the rollback plan.
Asserted without evidence: "I'm great at change management" — no second example when probed.
Avoided: direct question on conflict with the previous CRO went into a story about "philosophical differences" without specifics.
Signal: strong on execution, weak on conflict tolerance. Recommend a reference call focused on the CRO relationship before advancing.
Workflow gating
You can only run Interview Analysis on a candidate who already has an AI Evaluation on file. This isn't a pricing trick — it's the design. An analysis without a prior evaluation has no baseline to compare against, so the output would be weaker signal at higher cost. Evaluation first, analysis second.
How it differs
AI notetakers transcribe. Interview Analysis interprets. Transcription is a commodity (Metaview, Otter, Fireflies, Read). The thinking on top of the transcription is what decides the next step — that's what CertAIn does, and transcripts from your notetaker of choice paste right in.
Most AI hiring tools give you one of these. A few give you two stapled together. CertAIn gives you all four inside one workflow, so the output of ranking is the input to prep, and the evaluation you ran is the baseline for the analysis you run after the interview. One tenant, one credit ledger, one audit trail — four moments in a recruiter's day where AI judgment is actually useful.
That's what "AI hiring that explains itself" means in practice.