You get 45 minutes, sometimes more if the interviewers can see that youre close. Some were disinterested, some were engaging. Again, every word of that is important. No meta whatsoever. The first team asked me how I would implement some weighted search function and I had absolutely no idea what to do. Two or three noting it would. On the other hand, I haven't declined any that are given as homework, except one where the guy said it would take 8-10 hours. I had 10 years experience at that time and that was never discussed and also would not answer my questions about working at Google. the typing kind of defeats the whole "understand what it feels like to work with you on a daily basis" goal. I believe this is part of why the technical interview gauntlet has slowly become the norm and persists, is it's usefulness for that, although I don't think it's consciously happening, just a side benefit that those that hire are picking up on subconsciously. (Which is what I am assuming they are doing anyway -- as it sounds likes the interviewers are rating them across different dimensions). This information is for outside and inside the company. You're saying it is not. How many of those who didn't solve (or didn't even get on the track of solving) their problems actually ever got hired? Individual interviewers have their own style, approach, goals. That is, you're unlikely to have someone who's capable of writing down a nontrivial piece of code who can't actually write any code. Why shouldn't they be hired over you? It's like the Five-Year Plan in the Soviet Union. Hmm I thought they only send you to HC if they feel like you have a decent shot. Months of preparation + LeetCode + Multiple interviews are critical to increase your chances to get an offer. Maybe the guy was just having an off day but having spent a month preparing I felt cheated. But a large company that puts out talking points about hiring great leaders and team players probably can't even elucidate what that would mean (especially when the team member isn't interviewing, so you go for the generic common denominator and average blandness, with no real idea of what it means in particular), and if they can then I'm skeptical they are finding it-- there's a limited number of great leaders around and they typically aren't interested in doing the grunt work of maintaining an obscure cog in an obscure system in perpetuity, reporting to a faceless committee so they can make more ad revenue. Keeping things boring. But I have a hard time believing this author's claim that "it's not about solving the problem.". I just interviewed for an embedded position a few weeks ago. Incredible. I stopped interviewing with them, forever. I conducted dozens of interviews at my last place. Some people are bad under pressure with hard problems. I can come up with a million examples like this. Homework is much more close to the real world experience: get a task and would you have to figure it out. I'm not sure, but I'd wager that it might actually come closer than any other interview process. It's just absurd. Youre purposefully misrepresenting what I said to satisfy your preconceptions. You decided to latch on to one possible inaccuracy in what I said and disregard the overall point. Doesn't bother me. Even if it's not standardized, a rubrik across X attributes with a scale and a number. This means even if I end up getting the job at a company with an interview process like Google's, I'm unlikely to be surrounded by people who I enjoy working with. - that quote that is contentiously attributed to that person That's a reasonable range. So, while I realize the content of the questions themselves could be problematic, that's not generally what I push back on. a lot of high performers only get in 2nd or 3rd time and barely. Interviewers are nice and everything, however this whole process is flawed, and the post nowhere depicts the reality. Sorry for the pedantry, but I think maybe you mean "uninterested" rather than "disinterested". Unfortunately many (IMHO) get this part wrong. I had a google recruiter reach out a couple years back, and I asked for just some ballpark idea of salary range. That is why candidates are encouraged to try again, if unsuccessful. This is very well documented on sites/apps like Blind and accurate from my personal experience. Then again, another leading tech company loved me and stuck me on their core design team. I am not sure if this is in the interest of any company but why not provide feedback, as it can only help the interviewees in the future and the companies attract better talent. > Why does everyone on HN seem to care so much about what Google does in interviews? Three of my friends (all similar backgrounds) interviewing at Google this year have had completely different interviewing experiences. But has anyone done any proper large scale research on what is the best way to hire software engineers? Really? Hiring is hard, and I do think asking hard questions (on any subject really) does show a lot of the candidate Also the Dunning-Kruger effect is strong among interviewers. You'll almost always be quoted for a single, perhaps 2 levels. Writing down tricky algorithms on a whiteboard without notes is not a skill that's necessary to any occupation as far as I can tell. My own theory was that recruiters largely controlled this and they had quotas to meet of phone screens, onsite interviews and hires and this just ended up wasting a lot of interviewers' time. But why would they change it otherwise? Some asked relevant questions, some clearly had no idea what my background was or what position I was interviewing for. Sounds like a way of hitting your talking point of "we hire for great leadership!" To any hiring managers that might be reading this, please watch this alternative method of interviewing: Thanks for posting this! Id gladly interview with Google again. The only companies that can do that at scale are FAANG. Google, Go to company page <= 4 is a firm rule with me, unless they want to provide a stipend. Without such tight time pressure and scrutiny as an interview implies, I would expect any working software engineer to be able to come up with working code in a fairly short period of time. Individual interview scores are definitely not predictive of job performance. I've never interviewed with them, much less worked there, so everything I know or think I know about their interviewing process and work culture is second- or third-hand, but it sounds pretty miserable from the outside looking in. Personally I don't mind coding on a whiteboard but only if you understand why you're asking a candidate to do it and what you hope to gain. I also think that since this process helps select for fresh graduates who don't have any idea of their worth, have few responsibilities, no extracurricular hobbies, and are willing to put in an insane amount of time and effort in order to prove themselves is another big reason why this has become the norm. You usually find great leaders somewhere more exciting and independent, it's the nature of the beast. Just curious! I haven't done that, and I suppose I should expect to get rejections until I do. I don't see the confusion. Thats sounds like a very interesting, expensive, and borderline unethical experiment. There are alternatives, the current interviewing system is not a good solution. I have never interviewed for Google so wouldnt know if thats actually how it goes. Maintaining your level of aggression and resentment cant be healthy. Can't give an O(N) solution but come up with a correct O(N^2) solution? thanks for explaining how coding interviews work to me thou, really had no idea based on the 200 i've administered in recent years or dozens taken in my career. For example, the coding questions that I ask are not that complex. Handle upper and lower case (they should figure out this is an issue). In theory yes, probably that's the intention. Next for the technical part, a homework assignment. Yeah bro that's the plan . If not you may figure it out from first principles but if you don't it doesn't mean you're a bad programmer or you shouldn't be hired. And even if it does get past HC, will that mean a lower level? > The problem is that there is almost no correlation between how candidates do in whiteboard interviews, and the ability to perform the job on a day to day basis. "Googliness" is just a qualitative lever in hiring that allows Google interviewers to enforce whatever biases they have. There are probably millions of High IQ people applying and vying to work at Google. > I bet it will never become a standard for the whole industry. Sure, give someone a problem slightly harder than fizz buzz just to make sure they aren't a complete fraud. Both of these are really simple programming problems that the experienced coder should be able to knock off in 5 minutes. taking notes or rubric scoring, then it's just a matter of sending a summarized version of that and hopefully it's feedback from multiple people and not just one person's opinion. I've been interviewed by Google dozens of times over the past 7-8 or so years, have received two written offers and one verbal one. Perhaps if the question was totally off the wall rather than merely a gotcha. As you see, neither is solvable in 5 minutes. Before an onsite? If I do not, all is not well. How do you solve a problem? Good interviewing techniques aren't formally taught there, and only learned via shadowing and sitting in on other interviews. Simply because people of such personality will confuse friendliness for weakness and start to complain and argue about questions or even mild non-positive feedback. I ended up saying something along the lines of lets just pretend the method is called isPresent() because I forget to every interviewer. I feel like the most interested and serious candidates will be shut out by this. The employer's goals are to minimize time spent per candidate (since this is expensive) while hiring a sufficient number of qualified candidates with a minimum of false positives. Interviewers are human. Talking about technical topics? Im in a similar boat right now. Don't get me wrong, what they say they are trying to hire sounds great, and I think they think they are trying to find those things, but I'm highly skeptical of both the process and effectiveness. I'm reminded of an amusing rant by Adam Carolla: "I had no idea how f*ckin horrible most people are at their jobs.". So maybe? Much cheaper than the alternative which is to waste 2-4 hours per candidate. Practically, interviewers more often don't care anything other than solving the problem. I really dont hope that writing blog posts on twitter with one tweet per sentence and then threading them together becomes the norm rather than the exception, but my expectations are low. If they reject candidates who don't/can't consider algorithmic complexity of a particular solution, then they'll likely reject some candidates who would have performed well at a job there. It's important to remember to that the goals of the employer and the candidate are different. Those interested in learning more would do well to read Kahneman's "thinking, fast and slow." If anything I think I messed up the softball question. There are probably laws against it (at least in California) but the variation in interview experiences in this thread indicates that Google could possibly benefit from recording (video or at least audio) of interviews for quality control. As a category I now decline any questions at a site like coderpad.io where you have to code with a gun to your head. If its something that is never used in that particular job, why ask it in the interview in the first place? Great. Regardless, any of the points i can think of don't really seem that relevant to your original point and my response. It's not all about solving the problem, but to be clear if you don't solve the problem optimally you're not going to get an offer. (If u ask me, prolly the fish will become an amphibian lol). I like how the article tried to portrait these interviews as super professional and as if they have like a plan or something. The reason why these were my favorite phone screen questions was if the candidate couldn't hack a question like that, I could very confidently write up my interview report and tell the recruiter --- don't bother with the expense of bringing the candidate on site and asking 4-6 software engineers to spend 2-3 hours interviewing the candidate and then writing up a comprehensive set of interview notes/report. Personality imbalances in groups usually lead to dysfunctional situations over time: too many sloppy people who just cowboy code until dawn, too many excessively cautious people that won't touch a line of code any more with fear of breaking something, etc. I counter with "who cares?". That the interview process isn't more streamlined is a failing on the company's part however. Im an average looking dude and my interview experience at Google and comparable companies has been pretty similar to the one described in the article. I can get good work, with people I know, making good money, without going through any of that. But that sounds like how I run interviews myself. I dont think its fair to play with peoples lives like that, and firing those who dont work out sounds like a nightmare, but Id love to see the data that would come out of such a study. All the comments here seems like opinions of people based on themselves and no one provides a proper alternative. The easiest way to get there is by already having seen the problem. Pick any category of answers you like in fact. The candidate's goals are to try to get a positive signal to the employer. You can make them simple, and anyone remotely competent who isn't completely crippled by nerves should be able to zip through them. *. If you have a pool of 100 people to fill 10 roles and 20 are will work out then, as the employer, you don't often care which of those 20 you get, as long as you get 10 of them. We get it, no one likes weird off-topic brainteasers. You could criticize it, but it's so internalized in how the place works that there's no way to get rid of it without blowing up everything. But you also sometimes get a feeling based on the interviewers reactions etc. If you are in California, the employer is now required by law to disclose salary range. I think there is more clear incentive for interviewers to be more interested. For a majority of candidates, there is a correlation (both aspects). 2 of 5 bad interviewers. This seems like a highly optimistic, happy path description of a nigh-impossible process. Yeah, I struggle with that, too. Personal opinions. the typing thing is a killer- they practically have to transcribe the entire thing so there is 0% chance of making any kind of human connection with your potential coworker. > The problem is that there is almost no correlation between how candidates do in whiteboard interviews, and the ability to perform the job on a day to day basis. They kept giving me hints but it was clear the problem was above my skill level at the time. By all means write a blog post telling what you yourself is like as an interviewer at google. An inexperienced one will knock off in 5 minutes. 10 years ago I was at EA and they would throw me into interviews with a couple hours notice and no resume, so. No excuse for typing loudly + not stopping after you addressed it however. And I honestly think FizzBuzz is the right way to think about live coding tests because it's simple. Under interview conditions, where one has to not only come up with working code, but chatter constantly while doing so, it becomes a lot harder, simply because one has to multitask a little. the interview process is testing do you know what i am going to ask and that's all that matters. I was just reading a thread on Linkedin by a recruiter who lined up 30 candidates based on the client's requirements, only to have the client reject all 30 of them. L5 is a reasonable terminal level there, as in you can have a successful career and never go beyond L5, and no one will bat an eye. That's why it's an assignment in some beginning CS classes. They just let you type it in a google doc. I had same impression of my 5 interviews (3 well, 1 ok, and 1 bad), and got rejected at hiring committee. But it's impossible to really know. Even if we agree that millions is not correct, 100s of thousands or thousands of people? So annoying. How is the average newcomer to recruiting supposed to be savvy enough to suss out the 12 different ways someone can be a "Docker Expert?" These days, Google's interview system is more like getting better/cheaper replaceable labor/commodity. I conducted many interviews while I worked at google and I wouldn't trust anyone who tells you the interview situation as one particular, well-defined, thing. alternatively, to show that they're not wasting my time, they could just answer the question. I bet it will never become a standard for the whole industry. My interviewer showed up late and basically spaced out for the entire 40 minutes. They mostly involve translating a process that a non-engineer might have to deal with in everyday life into code. Ive only heard about the law that says employers are no longer allowed to ask candidates about salary history in California. It was recursion based, and my mind SHUT DOWN on recursion once it started. Your interviewers and and the type of questions asked depend on your potential level. A single bad interview does not kill your chances. All that said, being on the receiving end of these rejections is a real disappointment. If was just me with that feeling, I would agree that I was biased for not receiving an offer, but I see this happening a lot with good engineers and throwing all the responsibility on the interviewees shoulders with "You are bitter for not receiving an offer" or "gitgud" arguments is just avoiding the discussion. (LC and TC or GTFO, in other words.). When I've worked with interview plans it was around having the right mix of different technical interview styles and not really technical interviews (like often someone should run through the resume and validate things claimed). I would like prospective candidates for programming jobs to know that there is a thing in their language of choice that lets them make fast lookup tables. Go to company page > It is my experience that interviewers from countries with rote-learning based educational systems ask canned questions and expect regurgitated answers. I meant to post this earlier but forgot to, sorry. Go to company page I'm not sure I follow. The problem is that there is almost no correlation between how candidates do in whiteboard interviews, and the ability to perform the job on a day to day basis. I've switched to 1-2 hour homework exercises on the advice of an experience colleague though. In the third, the first phone screen's interviewer was constantly typing on his keyboard while I talked, and it was so loud. This definitely isn't advertised, I interviewed there on Friday for frontend and was presented with an entire wall to write on. maintenance, bug fixing) so maybe they just want effective robots? Clearly some people can do that, since Google hires thousands of people every year. Or they know and then force the candidate to use it and then mark them down for not knowing it (when the candidate never claimed to know it). Cool people on the other hand will be able to open up and show their skills more easily. So as a candidate, after you've been interviewed enough times you realize that X,Y,Z is really the name of the game and posts like these just seem totally out of touch. The singular most useful skill anybody can master in IT or any other profession is the ability to write off a loss and not let it worry you. And I want them to have heard of a hash table before. If you are a small business owner, you interview and have direct skin in the game. How do you solve someone else's problem? > I'm fully able to admit that I just wasn't smart enough for their organization. I enjoyed my interview at Google. I don't know what our offer rate is, but I think it is probably well below 50% even once you make it onsite, so being above average wouldn't be enough. The Silicon Valley interview system is a major blind point for these companies: a structural weakness which internal dogma paints as an unquestionable strength. Based on my conversations with Google's engineers during the on-site, I did not feel like I was dealing with the best and brightest, whereas at my new company I felt awed by the depth, experience and accomplishments of the team. We definitely should have hired them and we didn't. regex, then your results are biased in favor of people who have worked with regex. Certainly not a conversation despite everything I had seen and read about their interviews. What does it even mean for things like ideographic characters? True bro. Our hiring process is also tuned for low false positive rate, at the cost of high false negative rate. Several other companies claim the same. FizzBuzz is a simple problem aimed at providing an early negative signal on a candidate. Id qualify 4 as going "well" (two very well - gave them exactly what they wanted, two went ok - had solid real good answers at end) and 1 was terrible. Curious if they downlevelled you coz of the one bad interview. - Given an ordered sequence 1..10 and the operators {+,-,/,*} where you can put any operator between two numbers but maintain the number order and operator precedence, find the number of operator combinations that yield 5 digit positive integers. I don't know if other interviewers at Google have put in that kind of thought to improving the interview experience. However, as of 2014 average interview score was very well correlated with job performance (I can't give you further exact details because they are still marked confidential and i can't find a paper or anything that has disclosed it). This is ridiculous, you will be given a no hire recommendation by the interviewer at a FAANG company if you do not provide an optimal solution. But, the positive signal, the "can this person do the job" signal, is very very low. What else would you use as metric? I got it to 2, interviewer said "no, you can do it in 1" then refused to move forward. In some cases I saw the phone screen feedback (as part of the whole packet) and I honestly don't know why it didn't end there. When searching for jobs, I actually find the interview process of a company to be an extremely useful filter/signal. So it's not an effective mechanism for single interviewer to enforce biases anyway. The reason is I've given hundreds of interviews with that question, so I'm prepared for almost all the ways people can solve it, and especially all of the ways people get stuck or approach it in ways that would probably be useful if you had more than 45 minutes but aren't going to work in an interview setting. the interviewer i had couldn't even manage to come up with anything he liked about google, when i asked. The examples in a neighbor comment are very good, BTW. The negativity may likely exist from no response/lack of feedback -- which can be frustrating for ppl that spent time/effort), but at least as the previous poster mentioned some people do appreciate feedback. As I get older I've started pushing back on these bad interview decisions, but can't help feel it's a losing proposition without support from fellow developers. The quality of engineers and managers varies a lot. I know that interviews at big companies can be screwed, but among the top ones, I doubt any company is as terrible as Google is. No real conversation happening. No dice - they wouldn't discuss even a ballpark range until I'd flown and and given up a couple of days or my time for travel and interviews.