[x00gler123]

I honestly think their process is pretty good (at least it was for me). There are no bullshit questions and they are honest in their feedback/telling you what you should and should not expect. A lot of people are pissed because they didn't get in and feel it's google's fault for having a 'broken interview process'. But reality is that google is one of the most profitable/successful companies in the world and gets voted "top employer" year after year. A lot of people want to work for them, so they can afford to be picky in the interview process; but that doesn't mean anything is broken. Quite the opposite actually.

As per the secrecy; I reckon they want to keep part of their process secret so that people don't try to game it. At the scale they are operating (millions of applications per year) you obviously need some standardized tech testing for the first interview rounds (where no engineer is present yet) and if everybody just posted their questions and answers online you could kind of forget about that.

[GVIrish]

"A lot of people are pissed because they didn't get in and feel it's google's fault for having a 'broken interview process'. "

I think that's a mischaracterization of what people dislike about Google's interview process. People dislike that it takes a very long time for Google to follow up and that the process wastes a lot of their time.

Yes Google is profitable and a lot of people want to work there but that doesn't mean their interview process is great too.

[nxb]

I'd just note -- both total-profit and profit-per-employee are REALLY terrible ways to predict employee happiness. Just look at some of those rankings. They're a really mixed bag of both exploitive and great employers. I mention this because I always hear inexperienced people from small towns using those metrics, in isolation, to rank their employer choices. It's such a bad approach.

I'd even argue that highly-negative profit, pre-profitability employers can be far better for tech employees.

- High profit = the company's major problems are already figured out, they already have a money-making machine, they don't desperately need you and in fact you may be a liability who could break their already profitable machine. They could fire nearly everyone and possibly continue making bags of cash for a very long time.

- Highly negative in profit = they desperately need someone to solve their problems fast, which in tech product companies is often by making some new tech. Without you, they're nearly guaranteed to end up very screwed.

[BinaryIdiot]

> A lot of people are pissed because they didn't get in and feel it's google's fault for having a 'broken interview process'. But reality is that google is one of the most profitable/successful companies in the world and gets voted "top employer" year after year. A lot of people want to work for them, so they can afford to be picky in the interview process; but that doesn't mean anything is broken. Quite the opposite actually.

I'm not convinced of this argument at all. Many of the people I know who either work there or have work there complained about the process as well. I didn't get in which was fine (I certainly wasn't pissed) but I was very frustrated with the interviewer being 15 minutes late and when he asked me to implement an object he cut me off when I was almost finished to optimize a method (I protested; premature optimization and all). Then, since we had a truncated time table, I just finished optimizing and he ended the interview. I explained to him what code was left in my object implementation and it didn't seem to be a big deal. Then when I didn't progress a few friends I have at Google looked into it and found out I didn't get to progress because I didn't get to finish my implementation.

I've been in a variety of different types of interviewing. The more academic, white boarding of issues that you never even see on the job just suck in my opinion. I like it a hell of a lot more when I get to work with whoever I'm interviewing with and we're working on a real problem. Or hell even homework. Anything to show how I solve problems and code versus simply remembering different algorithms or data structures from college.

[geebee]

I sympathize. My experience was much better than yours - nobody showed up 15 minutes late, and everyone was exceptionally polite. Given how stressful in person, whiteboard exams can be, I'd actually compliment google on how well their interviewers treated me as a candidate. But in the end, I was (as I stated below), surprised with just how much optimized, relatively clean and ready-to-run code I was expected to write at a whiteboard. Writing plenty of code that shows you are capable of doing so, along with outlining a strategy that shows you clearly understand the problem and are working toward a solution? As far as I can tell, that's a no-hire.

One disagreement, I don't think this is purely about remembering algorithms from college. That's a necessary but not sufficient condition. If you can't do the basic binary tree traversal (and so forth) stuff cold, I'd say there's no way you'll pass the exams, because you need to solve more complicated problems that are based on core data structures and algorithms.

Honestly, I have to say I'm pretty damn impressed with people who can do this at the whiteboard. I might be able to get there, with a few months of study, though this would probably require neglecting other aspects of my life and job for a while. The cynical part of me has this vague suspicion that this screening is actually designed to hire only those without these sorts of demands on their life (kids, family, outside interests).

[Bahamut]

As with any interview, YMMV. My own experience was pretty negative - I got asked a question where one pretty much has had to have studied & memorized the approach to solve the question, with no live feedback or direction but mostly silence. I enjoyed a far better process with Facebook, where the interviewers at least made an active effort to engage my mind in the right direction if I blanked or was missing a key piece of knowledge, although their methodology for evaluation leaves some to be desired.

Personally, I greatly dislike how software engineers are hired in the industry, coming from a prestigious graduate program of mathematics. There is an expectation at some of these companies that the candidate studied for their interview (or got just the right set of questions), even if they demonstrated highly valuable skills such as figuring out a complicated trick on the fly without any prior knowledge of anything similar. If one wants the smartest people, it is well-known in academia that that is not how one should evaluate a candidate, so I am not quite sure why such shallow & flawed metrics continue.

[geebee]

"A lot of people are pissed because they didn't get in and feel it's google's fault for having a 'broken interview process'."

I agree with much of what you said, though I don't agree with this particular sentence. I'm always pause a little when someone supplies an unflattering ulterior reason for a criticism ("oh, they're just pissed that they didn't get in"), since there may in fact be reasonable criticisms of google's interviewing process that shouldn't be dismissed so casually.

However, as someone who went through it recently (software engineer position, didn't get an offer), I agree with you that it is straightforward, no gotchas or gimmicks. I'd say that at least in my case, if you review data structures and algorithms, and are prepared to use them creatively to answer middle-to-difficult style questions from "cracking the coding interview" (not the exact questions, of course, but at that level of difficulty), at a whiteboard, in 45 minutes, without too much help, then yes, you're prepped. I'm not saying you'll pass the exams, I'm just saying you won't be surprised (again, just in my own experience, for a SE position).

The only thing I was "surprised" with is just how much progress you really are expected to make on these problems and the level of accuracy you are expected to show in your coding. I gave myself three weeks to prepare. I should have taken 3 months, maybe more. No, every parenthesis or semicolon doesn't need to be placed correctly, but (again, my impression here) you pretty much do need to be writing code (probably in Java or C++) that will compile and run with minor edits, and you need to largely solve the problem, at a whiteboard, in 45 minutes. If you get stumped a bit and need a prompt, I don't think that'll rule you out, but if you need too much help getting through, my guess (based on limited experience) is that this will be a no-hire.

Fair enough. I think there's room for reasonable disagreement about whether this sort of thing produces too high a "false negative" rate, but I suppose that's up to Google. Good software developers are hard to find, and while false negatives do cost you, false positives cost you too. I'd generally say this is google's business, they can afford to be picky, as you said.

With one difference… which is that Silicon valley companies almost uniformly insist that there is a severe shortage of software engineers, yet at the same time insist on being (this may just be my own opinion), extremely picky about who they hire. Should an industry that by its own admission tolerates an exceptionally high false negative rate be taken seriously when it says it has trouble attracting enough talent?