Google Interview Question for Software Engineer in Tests

Country: United States
Interview Type: Phone Interview

Comment hidden because of low score. Click to expand.
of 0 vote

This reminds me of how Amazon works i.e. filters and sorts

1. Keep a pair of test data. The first would be ranked lower that the second on only a SINGLE parameter e.g. rating, price or else.

From the resulting list find whether the comes after the second. This should tell us whether the sort works well.

- deep.kulshreshtha July 04, 2017 | Flag Reply
Comment hidden because of low score. Click to expand.
of 0 vote

Very open question, I'd cover these aspects:
1. What is purpose of this test? Verify the search function, verify correct rendering, ... what is already tested? A single test should fit in some kind of test concept which gives an idea on what to test. A reasonable assumption could be, that search has been tested already on different known samples and information is around how the search has performed compared to this fixed set of data.
(notice, there is no perfect search result, but there will be indications whether it improved or not compared to other "generaions")
2. Define what must be tested, based on 1) e.g. verify that each items price is <= then the next, maybe for the first 20 pages, supposing it is very expensive or impossible to fetch all results. How ever, we still could miss a very low priced element that would only come after position 1000, or maybe even last due to an error. Now, several questions:
- sometimes search is designed to be fast vs. "perfectly accurate", that means, for the case of Google for example, that servers that participate in a query and do not respond within a certain time, will be ignored. Usually this is the case for not so important pages that have less redundancy etc. etc. So a certain error might even be tolerated.
3. Maybe we have statistical data, which gives a price distribution for the first 1000 items of a search and information how this distribution changed over time. Maybe we only want to accept a certain change in this distribution
4. Maybe the test should be around performance, e.g. how fast is the result served or how fast is the result rendered. Here we had more information, like a SLO (service level objective) where we'd say 99.9% of time the first 50 result must be served within 100 mS.
5. Or other tests could be that the system accepts "bad searches" which are searches that will tear down a server, etc. etc.

- Chris July 04, 2017 | Flag Reply

Add a Comment

Writing Code? Surround your code with {{{ and }}} to preserve whitespace.


is a comprehensive book on getting a job at a top tech company, while focuses on dev interviews and does this for PMs.

Learn More


CareerCup's interview videos give you a real-life look at technical interviews. In these unscripted videos, watch how other candidates handle tough questions and how the interviewer thinks about their performance.

Learn More

Resume Review

Most engineers make critical mistakes on their resumes -- we can fix your resume with our custom resume review service. And, we use fellow engineers as our resume reviewers, so you can be sure that we "get" what you're saying.

Learn More

Mock Interviews

Our Mock Interviews will be conducted "in character" just like a real interview, and can focus on whatever topics you want. All our interviewers have worked for Microsoft, Google or Amazon, you know you'll get a true-to-life experience.

Learn More