Estimation question

There is a colony of 1000 people on the moon. Estimate the bandwidth needed. How would you provide internet to them?

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

Estimation combined with practical infrastructure design reasoning for an extreme environment.

Clarify scope: estimate total bandwidth needed to support 1000 people's typical communication, research, and entertainment needs on the Moon.
Estimate per-person bandwidth usage based on typical modern internet usage, covering video calls, data transfer for research, and streaming.
Multiply per-person usage by 1000 to get an aggregate bandwidth requirement, stating the per-person assumption explicitly.
Address the real constraint: Earth-Moon latency (about 1.3 seconds one-way) affects real-time applications but is far more workable than Mars-level delay.
Propose an infrastructure approach: a dedicated relay satellite network or laser communication link given the Moon's proximity makes higher bandwidth links more feasible than for Mars.
State a final bandwidth estimate as a range, noting the dominant constraint is relay infrastructure capacity, not per-person demand.

Uses a grounded per-person bandwidth assumption, e.g. 10 to 50 Mbps average across a colony's mixed usage, marked as an assumption.
Correctly identifies Moon latency (about 1.3 seconds) as manageable, unlike the much larger Mars delay, showing distinct reasoning per scenario.
Proposes a specific technology, laser or high-frequency relay links, appropriate given the Moon's closer proximity to Earth.
Delivers the result as a range and flags relay capacity, not raw demand, as the real bottleneck.

Treating Moon latency the same as Mars latency, missing a key difference in feasible real-time communication.
Estimating bandwidth without proposing any concrete infrastructure to deliver it.

Chapters of the AI PM course that teach what this question tests.