Frame your answer around reversibility and learning: a quick version to validate demand is different from a quick version that creates permanent debt.
Best answers explore context: what is the urgency? Is this a one-way door or reversible? What technical debt would accumulate? Can we ship the quick version to validate demand, then invest properly if it works? Look for nuanced thinking rather than always choosing speed or always choosing quality.
Reveals how the candidate partners with engineering. Do they respect technical concerns? Can they think in terms of sequencing (validate, then invest)? Red flag: always defaulting to speed without considering consequences.