Quantum Learning Machines
RESEARCH

Research

We publish what we did and how. Cohort sizes, criterion correlations, confidence intervals, fairness monitoring, and errata โ€” including the results that don't favor us.

METHODOLOGY
How outcomes are calibrated
The measurement engine, role-demand profiles, and job analysis protocol. Published v1.0.
VALIDATION
181,800 evaluations
Three independent datasets. Republished every quarter.
FAIRNESS
14 dimensions, every session
Continuous monitoring. Items pulled within 24 hours of differential detection.
ERRATA
3 of 14 corrected
When we get it wrong, we republish with the diff and a note explaining the error.