Dynamic, not scripted
Characters respond to the candidate's choices. There is no fixed script to memorize and no single 'right answer' to look up.
How it works
A structured behavioral simulation, from role to evidence
Every candidate for a role gets the same live, standardized crisis — and you get a scorecard where each judgment is backed by their own words. Here is exactly what happens at each step.
The four steps
-
1
1 · Define the role
Search any job title. hire.center resolves it against a self-hosted copy of the U.S. Department of Labor's O*NET database (1,000+ occupations) and returns the standard skill profile for that occupation — ranked by how important each skill actually is to the job.
-
2
2 · Confirm the blueprint
The system pre-selects the skills that clear the importance threshold for the role. You keep 2–8 of the 14 assessable transferable skills. Optionally add role context — key tasks and a note — to ground the scenario in your world. The five-trait Composure Profile is always measured and never costs a credit.
-
3
3 · Candidate runs one live scenario
The candidate opens a link and enters a ~15-minute, multi-character workplace crisis compiled specifically for the role's pressures and job zone. The characters react to what the candidate says and push back — so a rehearsed, polished answer doesn't survive contact.
-
4
4 · You read quote-backed evidence
Within moments of finishing, a scorecard is ready: a 1–5 score for each blueprint skill and each composure trait, each with a verbatim quote from the candidate as evidence, plus a plain-language rationale and an overall recommendation you are free to override.
The scenario engine
Why a simulation beats an interview or a quiz
Static tests measure what a candidate knows about behaving well. A live simulation measures what they actually do when the situation fights back.
Role- and pressure-matched
The scenario's setting, stakes and escalation are compiled from the occupation's O*NET job zone and work styles — a nurse and a sales director face different crises.
Hard to game
Because the scenario probes behavior in context — with pushback, time pressure and competing demands — AI-generated or memorized answers tend to fall apart under follow-up.
What gets measured
Two axes on every assessment
The role decides which transferable skills matter; composure travels with every candidate. Together they answer both 'can they do the work?' and 'can they do it when it's hard?'
Axis 1 — role-relevant transferable skills
Social & Team
Social PerceptivenessCoordinationNegotiationPersuasionInstructingService Orientation
Problem Solving & Systems
Complex Problem SolvingJudgment & Decision MakingSystems AnalysisMonitoring
Strategic & Resources
Time ManagementPersonnel ResourcesMaterial ResourcesFinancial Resources
Axis 2 — the Composure Profile (always included, free)
Stress Tolerance
Staying effective and even-keeled when the situation escalates.
Self-Control
Keeping strong reactions in check; responding rather than reacting.
Adaptability
Adjusting approach as new information and constraints appear.
Cooperation
Working with — not around — the people in the room.
Empathy
Reading and acknowledging others' concerns while still moving forward.
The scorecard
Evidence-first by construction
The scoring model is instructed to cite, not to impress. Three design choices make the output defensible.
A quote for every score
Each 1–5 score is paired with a verbatim line from the candidate. The server re-checks each quote against the transcript and flags any it cannot verify.
'Insufficient evidence' is allowed
If the conversation never tested a skill, the model returns insufficient evidence for it instead of inventing a number. Honest gaps beat false precision.
Tone is not evidence
The judge is calibrated to separate confidence and charisma from demonstrated skill — so a smooth talker with weak decisions doesn't out-score a substantive one.
Run one yourself
The fastest way to understand it is to take the simulation and read your own scorecard. 12 free credits on signup.