Rubric
Rubric-anchored grading.
Every submission is assessed against published criteria across the six competency dimensions. The same rubric applies to all examinees, so results stay comparable and each grading decision can be traced back to the criteria it was measured against.
Dimension I
Abstract Manipulation
Operating on abstract structure across domains.
Dimension II
Swarm Intelligence
Orchestrating multiple AI agents.
Dimension III
Generative Art
High-quality multimodal artifacts.
Dimension IV
Practical Application
Real business problems, end to end.
Dimension V
Generative Coding
Working code, verified.
Dimension VI
Social Implementation
Ethics, law, and societal effects.
Scoring
Calibrated, multi-evaluator scoring.
Each submission is scored by a swarm of AI evaluators and calibrated to reduce single-model bias. Because scoring is anchored to the rubric rather than to a single reviewer, the approach is built to stay consistent as the assessment scales.