Repeatability
Equivalent cases are run more than once.
Interactive demonstration · No account
See how we evaluate an AI agent—in two minutes.
Replay a synthetic run or take the six cases to your own agent and record only what happened. Construct calculates the signals in this browser.
Local interaction · Synthetic cases · Your observations are not transmitted. Normal hosting logs still apply to the page load.
01 · Choose a use case
02 · Run or replay the cases
Run or replay the cases
Psychometrics, used carefully
Not a personality test. A measurement system.
We test whether behaviour is repeatable, comparable across controlled contexts and supported by the evidence needed for your decision.
Context sensitivity
Relevant context may change behaviour; irrelevant labels should not.
Coverage
Every score states which intended behaviours were actually sampled.
Uncertainty
Small samples produce wide ranges. We show that instead of a trust score.
Synthetic profiles are controlled test conditions. They expose possible failure modes; they do not represent or predict real people, and they do not replace user research.
Ready to test your actual system?
We first agree the claim, cases, allowed data and one controlled connection route. API access, expert analysis, timing and commercial terms are then defined privately in writing.