← AI System Design / 27
hardNewAppleGoogleMicrosoft

The cloud model is more accurate. Would you still put this assistant on the device?

Try it in the practice workspaceDraft your answer before comparing it with the rubric.

A model wins on connected tasks but leaves field staff without help during outages. Compare all scheduled work, then defend local data, expiring permissions and synchronization.

Updated Nov 2026 · Original practice questions with worked answers and teaching rubrics. Our editorial policy explains the sources, review process and limits.

Case briefRead the inputs, then make your call

Fictional routine-maintenance assistant. Two systems replay the same 1,000 visits. All initial replay devices support local inference. Acceptable notes contain correct identifiers, supported work descriptions and no invented completed work. Cloud results are needed at the visit, not after eventual reconnection.

Scenario inputs
Connected700 visits; 665 acceptable cloud notes; 574 acceptable local notes.
Disconnected300 visits; 0 acceptable cloud notes; 246 acceptable local notes.
Operational constraintsDocument freshness, offline permissions, device support and later work-order synchronization need a product policy.
Follow-upA new deployment has 1,000 devices but only 840 support local inference. Their visit distribution is unknown.

Your task. Compare cloud, local and a proposed hybrid route using consistent denominators. State which results are measured in the replay versus assumed, then define a rollout and offline-access policy.

A model wins on connected tasks but leaves field staff without help during outages. Compare all scheduled work, then defend local data, expiring permissions and synchronization.

This reference answer is in Premium. Free signup opens other answers.

9 answers per topic instead of 5, and your progress kept · no cardor unlock all 420 remaining answers
UP NEXT ON YOUR JOURNEY
NOTES & DISCUSSION · 0

No comments yet. How would you have answered?