AI built for the way agencies actually bill.
Credits, not seats. Reasoning paragraphs, not keyword scores. Deliverables you can ship to an end client as the work product — because that's what agencies sell.
Credits, not seats. Reasoning, not scores.
Agencies run many roles for many clients. Volume is the job. Cost-per-hire is a line item, not a tools budget. CertAIn's credit-based pricing matches that: you pay for the AI work you actually do on a real requisition, not for seats sitting idle between searches. A Pro plan at $499/month buys 2,500 credits — enough to rank ~2,500 candidates or run a mix of ranking, prep, evaluation, and analysis across dozens of active roles.
Provenance: pricing from SAAS_ARCHITECTURE §Subscription plans. Agency-first ICP per April 2026 investment analysis.
Four jobs the AI actually does on a search.
- Job 01
Triage a flood of applicants
You post a role on behalf of a client and the flood starts. 300–1,000 resumes in a week is normal. You can't read them all at the depth the client expects. CertAIn's Ranking action reads every one, orders them by fit against the JD, and returns a paragraph per candidate you can defend — so you can pull the top 20 confidently and move.
- Job 02
Ship a deliverable to the end client
Every reasoning paragraph is formatted for export. The three-finalist write-up you'd normally spend a Friday afternoon assembling is already there — ranking reasoning plus an AI Evaluation on each, styled as a PDF or copied into an email. The work product itself is the deliverable.
- Job 03
Prep the interview without losing the billable hour
Senior recruiters spend roughly ten minutes prepping each first interview. On a 30-candidate week that's five hours you can't bill. Interview Prep cuts that to a one-minute review of resume-grounded questions. The hour gets put back on a new requisition — which is what pays.
Provenance: ~10-minute prep-time figure from 2026-04-17 discovery call, Jordan, head recruiter at enterprise. Needs validation with agency operators specifically — flagged for Jon.
- Job 04
Run multiple clients without the context switch
Each client has a different company profile, different JD nuance, different definition of "senior." CertAIn stores the company profile once, layers a JD-specific override on top, and runs every AI action against that stacked context — so you don't carry 12 clients' idiosyncrasies in your head. The AI does.
How credits map to a real search.
| Action | Credits | What you get |
|---|---|---|
| Ranking | 1 per candidate | Fit paragraph + ranked position, full pipeline |
| Interview Prep | 1 per candidate | Resume-grounded questions, per interview |
| AI Evaluation | 2 per candidate | Full structured write-up for finalists |
| Interview Analysis | 3 per candidate | Post-interview signal + next-step rec |
Worked example
A search with 400 applicants, 20 advanced to phone screens, 8 to panels, 3 to offer:
- Ranking all 400 → 400 credits
- Interview Prep for the 20 screened → 20 credits
- AI Evaluation on the 8 panel candidates → 16 credits
- Interview Analysis on 8 interviews → 24 credits
- Total: 460 credits for a full search.
A Growth plan ($249/mo, 1,000 credits) runs two of those searches concurrently with room. A Pro plan ($499/mo, 2,500 credits) runs five. Credits reset monthly; you never get charged for a seat that didn't run a search.
"I went through 300 resumes and only passed 3 along."
— Jordan, head recruiter, 2026-04-17
"500 to 1,000 resumes at any moment for any role."
— same
"Tyler and I received 15,000 resumes between the two of us in 30 days."
— same
The volume isn't the interesting part — every recruiter knows the volume. The interesting part is that the pass-through rate is 1% and almost all of the work is sitting at the top of that funnel, unbillable. CertAIn moves the work into the part of the day an agency is actually paid for.
Provenance: 2026-04-17 Jordan discovery call. Jon flagged the need for direct validation with agency owners (Jordan is in-house enterprise). Tyler meeting scheduled 2026-04-22.
Your client’s GC will ask. You’ll have the answer.
If you place candidates in NYC, Illinois, or the EU, your clients are asking (or will ask) about AEDT, AIVIA, and EU AI Act exposure. CertAIn's architectural human oversight — every AI output is a recommendation to a human, never an automated decision — gives your team the foundation regulators expect across all three. Bias-audit data export is built in. Candidate AI disclosure copy is tenant-editable so your clients' legal teams can adapt the language. Details on /security.
What CertAIn is not.
- Not an ATS. Your candidates still live in Greenhouse / Bullhorn / Loxo / whatever you use. CertAIn adds judgment on top.
- Not a sourcer. We don't find candidates. We read them once you have them.
- Not a scheduler. Paradox does that for hourly. We're the evaluation layer above it.
- Not a chatbot. Every output is structured, exportable, and auditable — not a conversation window.