Vouch ranks your own applicant pool on the only outcome that matters — who performs and stays — instead of who screens well. With reasons, coverage, and an audit trail behind every decision.
Keyword filters and "likely-to-be-hired" scores optimize for who looks good on paper. The candidate who screens best and leaves in three months is the one they hand you first.
Tools rank on who gets hired. Train on that and you relearn exactly who you already hire — the same bias, just faster and with a confident number attached.
A hire who performs and stays is worth more than one who interviewed well and left. Conventional screening never sees the difference — it has no line to the outcome.
When a ranking is questioned — by a hiring manager or by compliance — "the model said so" is not a defensible answer. Most scores can't show their work.
Vouch works on the applicants you already have. You keep the pool; we rank it on realized outcomes and hand back a shortlist you can actually defend.
Hand Vouch the applicants for one role. Nothing is scraped, nothing is sold, and nothing leaves the engagement except the ranking you asked for. We rank your pile for your job.
Each candidate is scored on reliability-weighted signals: verified license and tenure, schedule and location fit, history of staying. Verified facts count more than self-reported prose. Missing history is never held against anyone — it lowers how much we claim to know, never the score. New grads and returners aren't buried.
For each candidate: a score, a coverage figure for how much evidence it rests on, a plain-language reason trail, and a logged record for adverse-impact review. Auditable from the first ranking — not bolted on later.
The strongest signals are the ones a candidate can't generate: a verified license, real recorded tenure, an observed history of staying. Vouch starts where those are cleanest — and the same engine extends to the next licensed role, and eventually to agents.
The first working slice. Verified licensure and tenure, schedule and acuity fit, and history scored against a retention rubric for the specific role — say, a permanent med-surg day RN on a high-turnover unit. The steady local hire surfaces above the dazzling résumé that won't last the year.
Book a demo →As work shifts to AI agents, the same scoring core ranks and routes them on logged, realized outcomes instead of marketing claims. A labeled simulation runs in the demo today; live agent routing follows once the human side proves out.
See the simulation →Three things most screening tools can't do at once: rank on the outcome you actually want, weigh what can't be faked over what can, and show its work on every decision.
Vouch scores candidates on realized outcomes — performed and stayed — using signals a candidate can't manufacture, and explains every ranking in terms a hiring manager and a compliance reviewer can both read.
Your pool stays yours. We rank it and hand it back. Nothing scraped. Nothing sold.
We're working with a small number of design partners. Hand us the applicants for one role and see the ranking — with reasons and coverage — next to who you'd have shortlisted yourself.
Design-partner program · We never keep or sell your applicant pool
