Lab 12 — Defensible QA Verdict

Time: 50 minutes of execution plus presentation
Team: three or four, with individual role rotation

Mac / Linux / Windows

Submission: prepare the deliverables below using the public-safe packet and review workflow. One PR per lab; private review is equally available. Never publish grades or personal information.

Choose a capstone executable on your supported workstation using the platform guide. Include OS/browser/shell and tested versus untested platforms in the verdict. Save Markdown/PDF and evidence in portable formats; present from a browser if desktop slide software is unavailable. No score depends on owning a particular OS.

See the platform guide and included local lab track.

Scenario cards

Choose one of the four runnable capstone cards with your instructor:

CRM role/permission change;
Community Launch responsive preview and intentional theme change;
inquiry-email integration change;
small repository regression and proposed fix.

Mission

Deliver a five-minute quality verdict to a mixed audience of product, engineering, and operations reviewers.

Required artifacts

Test charter
AI task brief showing approved scope and verification plan
One or more evidence packets
One test/automation proposal or a reason manual exploration is better
Ship/fix/investigate recommendation with uncertainty explicitly stated

Presentation structure

User and business outcome
Top risks and test approach
What happened and the evidence
What remains unknown
Recommendation, owner, and next action

Review questions

Reviewers must ask at least two of these:

What would disprove your conclusion?
Which behavior did AI propose, and how did you validate it?
Does the evidence show a technical symptom, a customer impact, or both?
Why is this a release blocker—or why is it not?
What would you automate next?

Individual scoring

Each learner is assessed on their rotated role, their evidence reasoning, their safe AI use, and their ability to name uncertainty—not the number of defects their team found.

The cards are public and open-book. Cite assistance and produce your own evidence. After the initial verdict, the instructor selects a small changed condition from the card; each learner predicts or explains its effect before verification. Fresh reasoning and an individual defense—not hidden answer files—establish understanding.