Purpose
Defines the intended outcome, audience and standard of success.
Interview probe“What outcome were you trying to achieve and how would you know it was good enough?”
Home · Methodology · PRIME
The behavioural standard for human-AI collaboration.
Core-6 measures baseline knowledge through Sova. PRIME is a separate framework for discussing and developing applied behaviours; it is not part of the Core-6 assessment.
Open the rubricRole & context
Supplies domain constraints, relevant background and boundaries.
Interview probe“What context did the system need and what information did you deliberately withhold?”
Instructions & steps
Breaks complex work into clear actions, checks and deliverables.
Interview probe“How did you structure the work so that errors were easier to find?”
Models & examples
Uses strong examples to anchor format, depth and quality.
Interview probe“What example or comparison helped the system understand the standard you needed?”
Evaluate & refine
Verifies claims, tests quality and iterates with accountable judgement.
Interview probe“How did you check the output and what caused you to revise or reject it?”
Enterprise application
Use PRIME where behaviour matters.
A shared rubric helps hiring managers and L&D teams evaluate the process behind an answer, not just the polish of the final output.
Structured interviews
Probe for purpose, context, verification and accountable judgement with consistent follow-up questions.
Workflow reviews
Examine how a team delegates to AI, checks results and retains decision ownership.
Targeted development
Turn observed gaps into practical coaching rather than another generic prompt-writing course.
Separate companion resource