Estimation question

We want to improve Toronto Streetview. Google cars will drive in Toronto and take pictures. How long it will take?

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

Operational estimation combining geography, vehicle throughput, and logistics, not just a raw distance calculation.

Clarify scope: total road distance in the Greater Toronto Area to be photographed, and assume a reasonable coverage target, most public roads.
Estimate total road length: a metro area the size of Toronto, roughly 6 million people, likely has around 10,000 to 15,000 kilometers of drivable road.
Estimate Street View car throughput: assume each car covers roughly 200 kilometers per driving day accounting for city traffic and stops.
Estimate fleet size: assume a realistic deployment of 5 to 10 cars working the region simultaneously.
Compute total car-days needed: 12,000 kilometers divided by 200 kilometers per car per day equals 60 car-days, divided across 8 cars, roughly 8 working days of driving.
Add buffer for weather, equipment issues, and data processing time, landing on a total project estimate of roughly 3 to 4 weeks including data upload and stitching.

Builds the estimate from road length through car throughput to fleet size, a complete operational chain rather than a single guess.
States assumptions about fleet size and daily coverage explicitly, so the interviewer can challenge specific inputs.
Adds a realistic buffer for weather and data processing, showing awareness that raw driving time isn't the whole project timeline.

Estimating only raw driving distance without factoring in fleet size or the throughput of a single car.
Ignoring non-driving time like data processing and stitching, which can meaningfully extend the total project timeline.

How would you prioritize which roads to photograph first if time were constrained?
How would seasonal weather affect your rollout schedule?