Diagnostic Work
Diagnostic work turns vague discomfort into a named problem, reviewable dimensions, responsibility and appeal paths, and a decision-ready output. Public samples reveal the diagnostic surface while keeping private engines and thresholds bounded.
Public samples
AI Deployment Risk Clarity Review
Reviews verification, escalation, rollback, accountability, and user or frontline burden before AI deployment expands.
Fast Answers, Slow Responsibility
Tests whether AI customer support removed the bottleneck or moved it into legal responsibility, human escalation, and hidden labour.
Pharmacy Safety UX Review
Reframes a retail-pharmacy controversy as a safety-visibility and intervention-timing problem.
When Part-Time Work Stops Functioning as a Job
Examines job-package viability, short peak-time shifts, schedule instability, and hidden transition burdens.
Audit architecture and protocols
CIEA Automation Quality Audit Extension
Tests automation claims through output warranty, verification burden, rollback cost, and accountability collateral.
Agentic AI Authority Cutoff Review
Determines when an execution-capable system should remain read-only, stay gated, or be frozen before authority expansion.
Stop–Go Threshold Protocol
A rule-based gate for deciding whether an AI system should receive one more step of delegated authority.
Deployment Review Form
A structured record for delegation stage, control, containment, and stop/go decisions.
Typical public output
A public diagnostic page normally shows:
- the situation or claim being reviewed;
- the named structural problem;
- the review dimensions;
- the responsibility, escalation, appeal, or rollback path;
- the decision-facing conclusion;
- the boundary between the public sample and the private analytical engine.