Worked sample · printable PDF
TRD-04 · QC that catches model mistakes
Which verification steps does a credentialed technician never skip when the suggestion came from a model?
Specialty centers · AI-Assisted Trades & Credentials
Two printable shop-floor briefs you can assign this term
Free briefs that treat the model as a tool on the bench, not a subject in a book. Students measure how often an AI assistant sends a technician down a wrong path, then build the QC checklist that catches it — and prove the checklist works against suggestions they already know are wrong. TRD-04 and TRD-01 have single-brief PDFs; every brief is in the bank PDF; the other ten are full on the page. No account. No customer data leaves your shop.
The shop rule
Model output is a suggestion to be verified, and the verification is the assignment. No brief authorizes acting on it on live customer property, a vehicle returning to service, energized equipment, or food for consumption. Customer, vehicle, and patient data never enters a public model. The capstone stacks onto a Board-approved credential — it is not evidence of one.
How to run one
- Students choose the brief. You name the trade and task family in writing first.
- Two scheduled checkpoints. Score with the six questions below.
- Mentor is advisory — school-supervised, never one-to-one unsupervised.
Featured packets
Featured · printable PDF
TRD-01 · Diagnostic assistant limits
How often does it send a technician down a wrong path — a test, a teardown, or a part that costs hours and money before the error surfaces?
Catalog (10 briefs on the page)
- TRD-02 · Service documentation
Can it write one that survives a warranty audit, a service manager's review, and a lawyer's reading?
- TRD-03 · Estimating under uncertainty
Where does the model run over, where under, and which of those errors is the one that costs you a customer?
- TRD-05 · Code awareness
How often is it right, how often is it fluently wrong, and what lookup workflow replaces trusting it?
- TRD-06 · The non-automotive variant
Pick a trade outside the shop bay — culinary, cosmetology, health support, early childhood, horticulture — and define where AI assist is safe, where it is merely useless, and where it is prohibited. Then prove the boundary with a test rather than an opinion.
- TRD-07 · Parts and supply chain
Photograph a part, or hand over a spec sheet, and ask a model to identify it. Measure the mis-identification rate — then price what a wrong part actually costs in return shipping, delay, comeback, and a customer waiting on a vehicle or a building.
- TRD-08 · Customer explanation
Does it stay accurate while becoming clear — or does it become clear by quietly promising something the shop cannot deliver?
- TRD-09 · Credential crosswalk
Where does the capstone genuinely document practice toward a competency, where does it merely touch one — and where does it prove nothing at all?
- TRD-10 · Shop data privacy
Customer names, addresses, plate and VIN numbers, building layouts, security details, household schedules, patient information. Write the rule for what never enters a public model — and make it something a busy technician will actually follow at 4:45 on a Friday.
- TRD-11 · Apprenticeship narrative
Turn your capstone into the artifact an employer, a hiring foreman, or a joint apprenticeship committee would actually respect — and find out by putting it in front of one.
- TRD-12 · Put the tool down
What was the signal that told you to stop, and how would a first-year apprentice recognize it?
Six-dimension rubric
- R1 — Is one trade, one task family, and one point in the workflow named — where does the suggestion enter the job?
- R2 — Are the trials logged so another student could re-run them with the same tool and scoring?
- R3 — Is there an honest comparator — the shop's existing procedure, the manual, or a journeyman's method?
- R4 — Are wrong suggestions counted and costed in shop terms: comeback, rework hours, warranty, safety?
- R5 — Could a working technician pick up the artifact and use it Monday without a phone call?
- R6 — Who is harmed when the model is wrong and a tech does not catch it — and what did the student refuse to let it decide?
Feedback
Scope unclear? Shop language off? Tell us here — you do not have to leave an email.