4 ms·

So if you're a tester or a UI interface person or a sysadmin whose work shouldn't really mean worrying about the big O order of algorithms - should these people

by redcap 15y ago

So if you're a tester or a UI interface person or a sysadmin whose work shouldn't really mean worrying about the big O order of algorithms - should these people also be asked questions about algorithms? The comments at the end of the blog post talked about a tester with years of experience being asked coding questions after she told the interviewer that she wasn't a coder. Another comment mentions a sysadmin applicant being asked similar things. The impression I get is that Google hires a lot of smart people to make technically hard things work very well (e.g. search), but fail in other areas that require a softer approach (e.g. Google Wave). So perhaps instead of focussing on algorithms, would it be wise to talk about problems that an interviewer (say a tester or sysadmin) might be facing in their daily work?

cletus 15y ago
> So if you're a tester or a UI interface person or a sysadmin whose work shouldn't really mean worrying about the big O order of algorithms - should these people also be asked questions about algorithms? Let's address each individually: - UI design person: we have UI/UX people, which is essentially a non-engineering discipline so won't have the same requirements. There are FE (front end) engineers who will be expected to have the same theoretical foundation as any other engineer; - the guy says he interviewed as a tester. Google's definition of a tester is different to that of most company. We have SETs (System Engineers in Test), who are expected to have a solid theoretical foundation. This makes more sense once you understand that most of our testing is automated rather than, say, writing and executing manual test scripts. - Sysadmins (SREs; Site Reliability Engineers) fall into two different categories: those with a more programming bent and those with a more sysadmin bent. The first lot will be asked algorithm questions. The second are more likely to be asked questions about networking, Linux administration and so on (the first will get these too but probably less). As for Google Wave, my personal opinion is that it was a solution in search of a problem so I wouldn't look at it as a failing of "softer" disciplines. Google+, as an example, seems to have been received very well, including on the UI/UX front, which would seem to fall in the same "softer" category. The early G+ successes and positive reaction IMHO stem from a more solid design that delivers value to users, something I don't think Wave ever did. One last thing I'll add is that Google's career ladders don't necessarily match up exactly to what you'd expect and those ladders are constantly re-evaluated. New ones come into being. Some disappear entirely.
- Lewisham 15y ago
  I've personally found the Google hiring process disappointingly vague, and I felt a lot like the OP. Do you want me to build a solution or do you want me to tell you what the Big-O notation is for an algorithm? These two are not necessarily the same thing. The one-size-fits-all interview style doesn't work, particularly if you get a crappy draw on the interviewers. The odd thing to me is that the Google hiring process optimizes for the theoretical, but everything I've seen seems to indicate that the practical is what moves you up the ladder at Google. And by practical, I really do mean it. Google (rightly, IMHO) values Getting Stuff Done and then optimize later for a number of job roles. And the people who I know who are very good at the theoretical can be quite poor at the practical, and it's detrimental to Google to hire those people into those roles. The hiring process simply doesn't seem to link to the reality. (I'm a Google intern, so I agree there's likely to be some skewed viewpoint)
  - johnx123-up 15y ago
    Reminds me "You cannot find a person who knows everything." How to interview a candidate? http://rajeshanbiah.blogspot.com/2009/01/how-to-interview-candidate.html http://rajeshanbiah.blogspot.com/2009/01/how-to-interview-ca...
- wpietri 15y ago
  As for Google Wave, my personal opinion is that it was a solution in search of a problem so I wouldn't look at it as a failing of "softer" disciplines. It kills me that so few people at Google even understand what good product people do. We engineers love solutions. We have entire books of them. We are hypnotized by tool catalogs and hardware stores because they are full of lovely, lovely solutions. Good product people, though, focus on the actual problems that people have. When solutions are proposed, they test them intensively on real people to see if they actually deliver benefit. If not, they don't ship them. Google Wave is absolutely a failure to appreciate "softer" disciplines.
- SergeyHack 15y ago
  It's too early to say "G+ success".

by redcap 15y ago

cletus 15y ago

> So if you're a tester or a UI interface person or a sysadmin whose work shouldn't really mean worrying about the big O order of algorithms - should these people also be asked questions about algorithms? Let's address each individually: - UI design person: we have UI/UX people, which is essentially a non-engineering discipline so won't have the same requirements. There are FE (front end) engineers who will be expected to have the same theoretical foundation as any other engineer; - the guy says he interviewed as a tester. Google's definition of a tester is different to that of most company. We have SETs (System Engineers in Test), who are expected to have a solid theoretical foundation. This makes more sense once you understand that most of our testing is automated rather than, say, writing and executing manual test scripts. - Sysadmins (SREs; Site Reliability Engineers) fall into two different categories: those with a more programming bent and those with a more sysadmin bent. The first lot will be asked algorithm questions. The second are more likely to be asked questions about networking, Linux administration and so on (the first will get these too but probably less). As for Google Wave, my personal opinion is that it was a solution in search of a problem so I wouldn't look at it as a failing of "softer" disciplines. Google+, as an example, seems to have been received very well, including on the UI/UX front, which would seem to fall in the same "softer" category. The early G+ successes and positive reaction IMHO stem from a more solid design that delivers value to users, something I don't think Wave ever did. One last thing I'll add is that Google's career ladders don't necessarily match up exactly to what you'd expect and those ladders are constantly re-evaluated. New ones come into being. Some disappear entirely.

Lewisham 15y ago
I've personally found the Google hiring process disappointingly vague, and I felt a lot like the OP. Do you want me to build a solution or do you want me to tell you what the Big-O notation is for an algorithm? These two are not necessarily the same thing. The one-size-fits-all interview style doesn't work, particularly if you get a crappy draw on the interviewers. The odd thing to me is that the Google hiring process optimizes for the theoretical, but everything I've seen seems to indicate that the practical is what moves you up the ladder at Google. And by practical, I really do mean it. Google (rightly, IMHO) values Getting Stuff Done and then optimize later for a number of job roles. And the people who I know who are very good at the theoretical can be quite poor at the practical, and it's detrimental to Google to hire those people into those roles. The hiring process simply doesn't seem to link to the reality. (I'm a Google intern, so I agree there's likely to be some skewed viewpoint)
- johnx123-up 15y ago
  Reminds me "You cannot find a person who knows everything." How to interview a candidate? http://rajeshanbiah.blogspot.com/2009/01/how-to-interview-candidate.html http://rajeshanbiah.blogspot.com/2009/01/how-to-interview-ca...
wpietri 15y ago
As for Google Wave, my personal opinion is that it was a solution in search of a problem so I wouldn't look at it as a failing of "softer" disciplines. It kills me that so few people at Google even understand what good product people do. We engineers love solutions. We have entire books of them. We are hypnotized by tool catalogs and hardware stores because they are full of lovely, lovely solutions. Good product people, though, focus on the actual problems that people have. When solutions are proposed, they test them intensively on real people to see if they actually deliver benefit. If not, they don't ship them. Google Wave is absolutely a failure to appreciate "softer" disciplines.
SergeyHack 15y ago
It's too early to say "G+ success".