USA | Google | SWE III 1st Round Technical
Summary
I had a first‑round technical phone interview with Google for a SWE III role in the USA; the interview focused on a stream deduplication coding problem and follow‑up design questions.
Full Experience
Had 1st round technical phone today with Google and it was overall not too bad.
No tricky DP or weird leetcode trivia. The coding is done in a google docs like environment without ability to run code. Rundown was:
- Intros
- Question
- Test case walkthrough
- Followups that tested some pieces of system design
- Q/A
The question that was asked was doing a dedupe of data that comes in as a stream. We may have a lot of physical devices that emit events like "Battery at 10%" at certain timestamps, but we need to remove utterances of the same event that are within 10 seconds of each other to reduce noise.
Almost like a mock logging service but no need to worry about File I/O
Followups:
- Time/Space complexity
- If you were to deploy your code to real production environment, what bottlenecks do you anticipate? How can you handle it?
Interview Questions (1)
Deduplicate Event Stream Within Time Window
Given a stream of events, each with a device identifier, an event type (e.g., "Battery at 10%"), and a timestamp, remove duplicate events that occur within 10 seconds of each other. The goal is to output a stream where consecutive identical events from the same device that are less than or equal to 10 seconds apart are omitted, keeping only the first occurrence.