Job tests used to come in one size. A warehouse worker, a salesperson and a nurse often answered the same battery of questions. It rarely mattered whether those questions predicted success in the actual job.
Miami-based PsyMetrics wants to end that. The company has launched an AI tool called Test Architect. It lets employers and HR consultants build their own hiring assessments from a library of more than 200 validated scales.
From Months to Minutes
Building a custom pre-employment test the old way took months. It meant hiring an industrial psychologist, running a job analysis and waiting on validation studies. PsyMetrics says its AI assistant, called Psy, compresses that timeline sharply.
A hiring manager uploads a job description or types in role requirements. Psy scans the text and identifies the technical, behavioral and cognitive traits the role actually demands. It then recommends scales from PsyMetrics’ existing library, rather than generating new, untested questions.
“Different roles require distinct combinations of cognitive capability, behavioral traits and operational skills…” said Jesse Llobet, founder of PsyMetrics. “Legacy assessment vendors continue to force candidates into rigid, outdated testing molds.”
Llobet founded the Miami company in 1996. He built its scale library over three decades of industrial-organizational psychology research before adding the AI layer this year. He earned his degrees at the University of Florida and the University of South Florida.
That timeline matters to how PsyMetrics pitches itself. Plenty of AI hiring startups launched in the past two years with no psychometric research behind their scoring. PsyMetrics is arguing the opposite case: decades of validation first, generative speed added on top.
Shorter Tests, Company Says
PsyMetrics says the tool shortens assessments from the traditional one-to-two-hour gauntlet to 20 to 30 minutes. Shorter, more targeted tests are meant to keep candidates from abandoning the process partway through.
PsyMetrics also cites turnover reductions of 25% to 87% across what it calls enterprise client deployments. It ties those gains to better alignment between test criteria and the actual job.
Those figures come from the company’s own client studies, not independent, peer-reviewed research. They are best read as a vendor claim rather than an audited result.
The system keeps a human reviewing every recommendation before it goes live, PsyMetrics said. Hiring leads can add, swap or remove scales before publishing a battery under their own brand.
Built for White-Label Use
PsyMetrics is pitching the tool heavily at recruitment process outsourcing firms and boutique HR advisors. Those firms can apply their own branding to the finished assessments.
That white-label option lets a small HR consultancy offer a “custom” test without building the underlying science itself. PsyMetrics supplies the validated scales; the consultancy supplies the client relationship.
It is a model built for scale rather than for any single hire. A staffing firm using the platform can spin up a new branded battery for each client. It never has to commission a new study when a client signs on.
The scoring benchmarks are designed to align with federal Equal Employment Opportunity guidelines, PsyMetrics said. Employers still carry responsibility for how they use any results in hiring decisions.
Every scale in the library has gone through validation testing, the company said. That sets it apart from AI tools that generate test questions from scratch with no track record.
Why It Matters Here
PsyMetrics has operated out of Miami for nearly three decades, well before AI hiring tools became common. Its wager is straightforward. A South Florida assessment company with a long validation record can move faster into AI than newer rivals. Those rivals often have flashier tech and no testing history behind them.
For Miami employers weighing whether to trust an algorithm with hiring decisions, that history is the pitch. Whether the turnover numbers hold up outside PsyMetrics’ own case studies is a separate question. Employers will have to answer that one for themselves.