EASE. Evidence behind readiness.
Engine for AI-based Skill Evaluation. EASE grades work produced during defined Nuvepro Competency Assessments in an agreed practice environment. The available work record supports grading against the assessment criteria.
Choose evidence for the responsibility
Knowledge questions and practical work answer different assessment questions.
Multiple-choice questions can assess knowledge and recognition. For work such as configuring infrastructure, debugging a service or handling a failed workflow step, practical evidence shows how a person applies that knowledge under the defined conditions. Choose the mix of questions, work products and observation needed for the responsibility.
A course completion record alone does not establish whether people can do the changed work. Readiness needs evidence from defined scenarios, including judgment, exceptions, and the handoffs people will own.
What the assessment can examine
Assessment criteria can cover implementation, quality, recovery and outcome. Agree the scenario and which evidence the environment will capture.
State which criteria EASE grades from the captured work record and which a human reviewer assesses.
Implementation
Did the candidate deploy or configure the required components in the assessment environment? Check the submitted state against the scenario requirements.
Quality
Did the work meet the agreed security and quality criteria? Review configuration, permissions, exposed secrets and resource choices where those are part of the scenario.
Recovery
When the exercise includes a failure, did the candidate inspect the available evidence and choose an appropriate next step? Review the recovery actions captured for that assessment.
Outcome
Does the resulting work product satisfy the scenario's acceptance checks? Review the result and any unresolved exceptions.
Reading a score
Review the grade alongside the submitted work, available action record and candidate explanations required by the assessment.
EASE produces a normalised score and grade for the configured assessment. Interpret them against the rubric and evidence for that run.
A score reflects performance against that assessment's criteria. Review scenario coverage, allowed assistance, available evidence and any grading uncertainty before deciding what further practice or assessment is needed. A high score does not establish performance on unassessed work.
Where it sits
EASE provides assessment evidence for the human responsibilities being evaluated. A readiness decision must state the role, scenarios, criteria, and work covered by that assessment.
When scoping a Competency Assessment, simulation or Project Readiness engagement, agree the responsibilities, rubric, environment and human review process. Confirm which evidence EASE will grade and how questions or disputed results will be reviewed.
People, agents, and the shared workflow
Human skill assessments examine what a person can do. Agent evaluation tests agent outputs against workflow-specific scenarios and criteria. A shared workflow also needs its handoffs, exceptions, and outcomes reviewed. One score does not establish readiness for all three.
Your operating owner decides what further evidence, controls and acceptance are required before production use. As agents, tools, or responsibilities change, agree which scenarios and assessments need to be revisited.
See Project Readiness for Your People