These are HLD questions of the kind Uber actually asks — the patterns reported from Uber's AI-engineering rounds, where design trade-offs, scale and failure modes matter as much as definitions. Treat this page as a mock interview: say every answer out loud before revealing it. If one surprises you, the lesson behind it is linked at the bottom.
Uber HLD concept questions
Uber interviewers commonly ask how to efficiently find the nearest available drivers to a rider. Why is a plain SQL query scanning all drivers' locations not a viable approach at scale?
Asked in
During a brief network partition between Uber's matching service and its driver-location store, should the system keep matching riders to drivers using possibly-stale location data, or pause matching until it's sure the data is current?
Asked in
Uber HLD applied & hands-on questions
Uber-style: each of 500,000 active drivers sends a location update every 4 seconds. Design the ingestion path so this doesn't overwhelm the database, keeping in mind that only the LATEST location per driver actually matters.
Asked in
How to use this page: Uber rarely asks something you've never seen — they ask a standard HLD concept and then push one level deeper ("why?", "what would you do if..."). Master the concept in the HLD course lessons, and the follow-up stops being scary.
Keep practising: Caching Explained, Databases at Scale, Design a URL Shortener and Interview Framework cover what most Uber HLD rounds test.