What the assessment measures
The assessment contains 15 multiple-choice scenarios across critical reasoning, prompt literacy, AI evaluation, bias and safety detection, and model quality judgment. It samples how you interpret a situation and choose a response, not your performance on a real job.
Inputs and scoring
The scoring endpoint considers your selected option for each scenario. Each option has an editorial weight from 0 to 10. The overall score is earned points divided by 150, rounded to a percentage; category scores use available points in that category. No resume, demographic information, salary data, or employer outcome is used.
How role fit is suggested
Scores of 88 and above suggest AI Specialist, 78–87 Senior AI Evaluator, 65–77 AI Trainer, and below 65 AI Evaluator (Foundation). These are editorial score bands, not validated hiring criteria or a professional level you have earned. Compare the tasks with your actual work evidence and an employer's real requirements.
How the salary guess is produced
Each score band carries a fixed illustrative USD hourly and annual range from Thrive's existing assessment design. The estimate does not query job listings, salary surveys, your resume, or location. Hourly and annual ranges are separate guesses, not a mathematical conversion. Actual compensation varies by country, employer, experience, employment status, and contract terms; verify it in a real opportunity.
Limitations
A fixed multiple-choice set cannot measure every capability, and repeat attempts may reflect familiarity. Scenario wording and rubric choices affect the result. Role suggestions and salary guesses are not psychometrically validated, an employer endorsement, a credential, a compensation offer, or an employment guarantee.
What to do next
Practice one category using pairwise evaluation, source verification, or prompt auditing. Keep examples and checks as evidence of practice, then compare that work with actual role requirements.