Product design question

Design a shared carpool experience for Lyft.

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Marketplace product design balancing rider cost savings against driver earnings and route efficiency complexity.

Users: cost-sensitive riders on common commute routes, and drivers willing to accept slightly lower per-trip fares for higher trip volume.
Pain points: solo rides are relatively expensive for riders, and matching multiple riders efficiently without excessive detour is technically hard.
Prioritize: focus on high-density, predictable routes first, like commute corridors, where matching riders with similar origins and destinations is easiest.
Solution: a carpool matching algorithm that caps added detour time and price riders based on shared cost savings versus a solo ride.
Guardrail: driver earnings per hour should not drop below a threshold compared to solo rides, or drivers will stop accepting carpool matches.
Success metric: match rate (percent of requests successfully pooled), and rider cost savings versus solo ride price.

Scopes the launch to high-density predictable routes first, a realistic way to solve the matching difficulty problem incrementally.
Sets an explicit detour cap as a design constraint, addressing the core technical challenge of carpool matching directly.
Includes a driver earnings guardrail, recognizing that driver supply depends on carpool remaining worthwhile for them too.

Ignoring the detour and matching complexity, assuming any two riders can simply be paired.
No consideration of driver earnings impact from lower per-trip fares.
Launching broadly everywhere instead of starting with high-density predictable routes.

How would you cap detour time without making too many riders wait too long for a match?
What would you do if drivers started avoiding carpool requests?