Identity and eligibility
Confirm the participant and authorization context without treating identity as competence.
Applied assessment laboratory · proposed pilotThe assessment begins
The assessment begins
where the résumé ends.
Realistic work. Changing constraints. Inspectable artifacts. Independent corroboration. Outcomes with attribution and uncertainty intact.
Verification architecture
A claim becomes useful only when its chain survives inspection.
Each layer answers a different question. None is allowed to substitute for the others.
Work under constraint
Observe decisions, artifacts, revisions, and recovery—not recall of benchmark language.
Independent corroboration
Ask references about specific work, scope, contribution, and outcomes they directly observed.
Outcome and attribution
Separate what changed from why it changed, who contributed, and what remained outside the practitioner’s control.
Minimum viable assessmentTwelve capabilities.
Twelve capabilities.
Twelve distinct pieces of work.
The first pilot should prove depth before breadth. These are task hypotheses, not a released test form.
DISC.STAKEHOLDER_DISCOVERY · proposed pilot task
Not administeredMap the contested intake redesign
Given interview excerpts from a sponsor, operations lead, frontline user, security reviewer, and an absent downstream team, propose a stakeholder and decision map for the next discovery step.
- Stakeholder and influence map
- Question plan with owners and decision dependencies
Midway through the task, disclose that the executive sponsor cannot approve workflow changes and that a contractor performs the highest-volume step.
- Revise decision ownership
- Add the contractor and actual approver without discarding earlier evidence
Do not infer
This exercise would not establish effectiveness in live customer interviews or politically sensitive environments. A polished map would not establish that the candidate earned trust from the represented stakeholders.
Activation sequence
Fielding is a gate, not a launch date.
Every stage produces an inspectable receipt. A missing receipt stops the next state.
Blueprint
Critical tasks, conditions, artifacts, and non-inference limits are frozen before participant work.
Response process
Cognitive interviews test whether the exercise elicits the intended capability without irrelevant barriers.
Rater calibration
Reviewers train on shared anchors, score overlapping units, and resolve material drift.
Controlled pilot
A consented, preregistered cohort completes tasks under declared conditions and accommodations.
Outcome study
Independent outcomes are linked only after leakage, attrition, access, and attribution controls pass.
Standard setting
A qualified, conflict-disclosed panel may propose criterion anchors after evidence exists. No cut score is authorized.
One number would hide the evidence we need most.
The pilot preserves capability, evidence, uncertainty, and readiness as different outputs. A future standard-setting study may propose criterion anchors, but v0.1 cannot emit an official threshold.
Inspect the scoring boundaryCapability
Criterion-level observations from work; no résumé or title multiplier.
Evidence
Coverage, independence, maturity, and provenance remain visible.
Uncertainty
Missingness, disagreement, task conditions, and limitations travel with every interpretation.
Readiness
A governed decision state, separate from proficiency and evidence maturity.
Current authorization
Inspect the field research program Design, synthetic rehearsal, and critique only.
No recruitment, real evidence intake, result publication, certification, ranking, or employment decision is authorized.