My Google Data Scientist interview - I got DESTROYED

so a couple years ago i applied for a data science job at google and it was a disaster keep watching and i'll tell you all the story i'm richard and this is richard undata [Music] alright so in my last video i gave some tips about how to get a data science job and i mentioned that from my own personal experience i've generally found it pretty easy to get jobs there have been a few exceptions along the way there and the most memorable of all of them was google i mentioned this in my video and i got a couple requests that came in asking me to tell my experience interviewing with google so i figured you know what this is a fun story some of you out there might find it useful some of you might find it entertaining who knows maybe some of you will find it both of those things so i'm gonna tell that story before i do that my usual asks hit the subscribe button to my channel if you haven't already for all kinds of content just like this smash the like button because that really does help my content reach a larger audience also i will have a link in the description of this video to my patreon account so if you guys are willing to support me that way that would be enormously appreciated so first of all just a little bit of background on me so when i was in school and in the couple years just getting out of school i never really had a lot of interest in working for the big tech firms or the fang companies that is facebook apple amazon netflix google and i mean these are great companies and everything and i really do understand the appeal but it just wasn't something that i was personally interested in and for a few reasons i mean number one i live in the detroit area that's where all my friends and family are for the most part anyway and those companies don't have a huge presence over here and it's just a thing for me personally if i have to weigh between some big career move and staying close to my relationships here my family my friends i'm always going to err on the side of my family and friends so that's one thing and the other part of it is that culture that you get ingrained in over there where it's very workaholic based i mean i know some people who work at google i know some people at apple and generally they tell me pretty similar things you're kind of living to work at least in the short term and honestly the the whole concept of just pouring that many hours of your life every week into one single company is not something that i've ever been interested in i highly prefer i work for 40 to 45 hours a week at whatever company i have around here and then in my spare time i'll do other things i'll enjoy uh spend time with my family and friends do my own side projects whatever it may be i just don't like working all that much for one single company you might but that's just me all that to say i was never super pumped about working for any of these companies in the first place but i had a buddy who's over at google and he would always tell me a lot about it like oh man it's you should totally apply over here it's so much fun over here you're in sunny california you'll learn so much you can take the experience you've gathered here if you leave a few years later you can take it anywhere else and i mean that's a totally valid point not that i was super interested but i kind of went along with it just to kind of get this guy out of my hair so i said all right yeah if you want to refer me go for it and so he did and it's my understanding that google gets tons and tons of applications from all around the globe every single day but if one of their employees refers you you sort of get moved to the top of their stack so a few days later one of the recruiters did get a hold of me and they said i would be a good fit for a data scientist position and they gave me this questionnaire to fill out first kind of as a screening measure and the questionnaire was pretty straightforward it asked me just previous applied statistics methodologies that i've implemented in the past how i've gotten data in the past and the data quality issues that you kind of run into from a practical standpoint pretty straightforward stuff also my top technical skills and how i rank my capabilities in all of them so whatever i did all that turned it in nothing to it well then a week later a recruiter from over there got a hold of me to set up a google hangouts interview now one thing i really do have to hand it to google on is when they do their interviews they are very organized and very transparent about it so they basically told me up front this is going to be a seriously technical interview um the things we're going to focus on are going to be applied stats and programming capabilities and they're also very transparent about the criteria that they're using to evaluate you so they're basically looking at four different things and that's stats knowledge your intuition about data related problems and data in general programming capabilities and then how googly you are which basically means how good of a culture fit you are and i had a week's time up front but i barely prepared for this interview at all like if i remember correctly i took this interview at like 5 p.m on a day in which i was taking half the day off of work and then in the second half of that day like after lunch i started going back through some old course materials i did in school for stats but that was really the extent of all the preparation that i did for this interview like basically nothing so i got to the interview and then the very first question was some kind of coding related question um and i gotta be honest i don't remember these questions at all there are some major themes here and there that i remember from these from these questions and how i felt about the questions after the fact but unfortunately if you came to this video looking for specific google interview questions i don't have any of those for you i do remember the first question was just a basic coding question and when i say basic it wasn't super elementary but it wasn't super crazy or advanced or hard either and the guy asked me right off the bat like what language i feel the most comfortable coding in and i told him r and he said okay well for any of these questions you can go ahead and use r and the whole time we had the shared google doc that was open so he could see my thought process and the things that i was writing down so there was that the next question was this applied stats question and i don't remember all the details about it but i do remember it was something that could be modeled using a binomial distribution and he went on to ask some questions about like what do you do for a confidence interval both in the large sample and the small sample cases as well as the behavior the estimate and the confidence intervals if you have a p-hat that is the estimate of the of the parameter you're trying to look at here if that's close to zero or to one so there's those questions but then there was a third question and he told me there were multiple parts to it but there were only like three or four minutes left in this interview so it was really i gave a few thoughts on like part a of a three-part question and then the interview just just had to end because we were out of time after this interview i gotta be honest i had no idea whatsoever how i did like i knew i was maybe a little slow on the first couple questions but that generally i had all the right answers but because i was slow i barely even got to the third question at all so i knew that part of it kind of sucked uh but you know a week went by and then i was sitting at my job and then i received an email saying hey can we give you some feedback about your interview at 1pm i said sure so then the recruiter calls me on the phone and they tell me okay richard you did a great job so we'd like to fly you out to mountain view to the google headquarters and have you do a full-blown day of on-site interview over there for those of you who are overseas mountain view is in northern california it's in the san jose area of california so at this point my mind is kind of blown right like all this started because my buddy was telling me i should apply there and now all of a sudden i'm flying to the other side of the country pretty much uh in order to do this interview so all of a sudden this is getting pretty serious and so the way they structure their interviews there is at least this is how they did it for me there are four interviews each of them lasts an hour and actually the way it worked with me is you had the first interview and then there's lunch and then you have the three interviews after lunch so they gave me two weeks of prep for this and then the interview was going to be on a friday and i probably prepped a little bit more for this one but not all that much so i would do an hour and evening here and there and the flight from detroit to san jose is right around four and a half to maybe five hours so i was prepping on things like doing some r code here and there reading my old applied stats stuff pretty much consistently the whole time but nothing too seriously intense so the morning of the interview i wake up in the hotel i'm feeling good i'm ready to fight dragons i mean really what i was doing was fighting dragons with a dime-sized toothpick because i didn't do much prep work but regardless i felt ready for this thing now i do have to mention of course google's campus is absolutely gorgeous like from a landscape perspective it's in an absolutely beautiful just rolling hills type of area the campus is full of souvenir shops and museums it really is huge too i mean it can be kind of mind-blowing if you're used to working in like a little office building where you have your parking lot and your building and that's it it's nothing like that with google or facebook or any of these other tech firms and obviously i can't do a video about google without talking about the food the food was awesome so a lot of you probably know this but google has these cafeterias there where they're basically serving food i want to say 24 hours but basically around the clock and obviously they're doing this uh just because it makes it that much more difficult for people working there to go home it's a lot easier to you know you're working you eat you work some more so they're doing that for a particular reason but regardless there's a lot of food and it was really good then i go in for the interviews and you come in and there's this lobby and it was packed full of people who were waiting to go into their interviews for various positions so they grab me and like i said there's four interviews to this process and when i look back on it the first interview was the worst of all of them at least i think you got in there and the first question was some kind of conditional probability question and this wasn't any stats 101 level stuff no this was some serious math stats type of stuff that you know maybe had i been in school at the time with the math stat stuff a little fresher in my head that i could have answered a little bit more clearly but it wasn't good then there was some question about the cdf of a distribution that came up a little bit later and it was just awful like have you ever been in school and then you see this problem like usually in the back of an exam or something and you're literally looking at it and you have no idea how to even start it i remember that feeling a couple of times here and there when i was in school and it was exactly like that the second interview as i remember was pretty heavy on applied statistics and this one was pretty similar overall to the one that i did over uh the google hangouts interview like i remember there was some topics that came up about counts and rates going back and forth between the two and just the poisson and exponential distributions uh putting these things into a multivariate context like again i'm very hazy on the specifics here just because of how long ago this was i remember this interview wasn't too bad but then i got into the third interview and this is when my ass started getting kicked all over again like the first question on that one was very coding heavy um like once again we're not talking your straight comp sci 101 stuff like this was some actual like heavy stuff again it was using r i think there was a sequel question that was in there too but it was some real uh creating algorithms using r and i remember i was getting pretty stumped by this up until like 40 minutes in or so when eventually they moved into some more applied stats heavy questions now how i did on these was kind of hit or miss like there was a question about regularization like what's the difference between ridge regression and lasso when do you use each mathematically what's going on there's a question about feature selection and what happens when you have covariates that are pretty uh heavily correlated like what happens to the covariates that you keep uh so stuff like that then i got to the fourth interview and this one started out pretty good uh they asked me how you explain concepts like p-values um conducting multiple hypothesis tests and bonferroni corrections and things like that and i taught stats for a little bit while i was in doing my master's degree so explaining these sorts of concepts wasn't something i had any kind of problem with but then they threw this finance related question at me and i don't remember any of the details about the problem because they kept throwing just one piece of information at it after another after another there was like six or seven different variables i remember to this problem and i mean this was a real serious applied sort of problem and once again it's one of those problems where like i was looking at it and i i legitimately had no idea where to even start and the problem with a google interview is it's not like exactly like how it is in school where sometimes maybe you'll get a problem like that but you'll see it and then you'll skip to another problem that you know how to do you can't do that in a google interview because you can't so if you can't answer one question they're just not going to move on to the next one right then they are going to want to see what your thought process is and that's one of the things they're evaluating you on like even if you can't answer the problem all the way they want to know what's on your mind honestly with this problem i mean my mind was so jumbled from everything that was going on it was challenging even to articulate all the hazy stuff that was going on in my mind so that that interview was just a complete and utter catastrophe like i i came out of this interview feeling like my brain got hit by a bus this was on a friday then the weekend passed and i got to like monday or tuesday and the recruiter called me on the phone and told me no i did not pass my final interview with google and so here we are the recruiter told me you can apply again after a year and i remember just as i went through the process i was starting to get kind of hyped up about it like before i wasn't interested in it and i mean to this day i'm not but while i was going through the process i'm like okay i could i could really see this right like the money's really good the point about it's great experience that you can get now you might as well get the experience now and you can go wherever you want later all those are really valid points and this story is really fun to tell it's fun to tell people uh like over drinks and dinner or whatever um but it's never been something that i've been particularly interested in going back to just because i'm happy with where i am right now uh just not really interested in trying again with google and again google is a wonderful company for those of you who are interested in it it's just again not something i was really ever super into myself it really does require prep like when you're in school there's a lot of things that you will learn but over time you're going to forget it if you're not applying it and using it from the day to day and honestly you learn so much while you're in school there's absolutely no way in any job in the world that you're going to use everything that you've learned all the time so if you're really interested in a job at google you have to really want it and you really have to do the work up front to prepare yourself but if you come from a good background you really know your stuff and study it up again if you want it bad enough i think you can get it now there are several reasons why i decided to share this story on here with you guys number one was it was requested so hopefully somebody out there finds it useful or funny or whatever but the other reason is i do want to make it clear with my channel that i don't profess to be a data science guru and honestly anybody out there who calls themself a guru in data science is lying to you for their own ego or because they're trying to sell you crap there's tons of stuff out there in this field that i don't personally know and the same could be true of virtually anybody in the community and i've had instances where i've applied for jobs and i've not gotten them you just heard a story about it now maybe this one would have been different had i done more prep work up front to put myself in a better position um who knows not really something i think about but maybe it would maybe wouldn't but have been in the data science world for close to six years now and i have learned a lot as i've gone along the way that i wish i could have told myself six or seven or eight years ago and that's why i use this channel to share those things i know and the things i've learned with you guys so you're better educated and better informed so i'll leave it all there thanks everybody for uh for listening to the story again hope you enjoyed it if you did hit the like button share this video leave me a comment down below then i'll see you all in the not so distant future until then richard on data

Channel: RichardOnData

Views: 12,563

Keywords: google data scientist interview, google interview story, google data scientist, faang data scientist interview, google interview, data science interview google, data scientist interview google, google interview tips, google data scientist interview tips, google interview process, google interview experience, google data scientist interview process, data science, joma tech, data science jay

Id: IvWZcibbdQ4

Length: 18min 55sec (1135 seconds)

Published: Mon Aug 10 2020