(Almost) Everything You Need to Know About the Data Science Interview (Nick Singh) - KNN Ep. 80
Updated: Jul 31, 2022
Today, I had the pleasure of interviewing Nick Singh. Nick is a career coach and the best-selling author of "Ace the Data Science Interview". Previously, he held a variety of Software & Data roles at Facebook, Google, and VC-backed startups. He aspires to be the Drake of Data Science....whatever that means!" Today we learn about Nick's fascinating career journey through tech, the keys to landing the data science job, and his unique experience writing his book. I hope you enjoy the episode, I know I enjoyed our chat!
[00:00:00] Ken: If you were an employer, how do you think is the most effective way to do this since you've been...
[00:00:03] Nick: I would definitely. Yeah. I would definitely ask them about SQL and it's okay, pick whatever flavor SQL you want and I'd make up some hypothetical table and I'd just make sure they understand concepts like joins. Maybe you can write a window function, just basic queries.
[00:00:25] Ken: This episode of Ken's nearest neighbors is powered by Z by HP. HP's high compute, workstation-grade line of products and solutions. Today, I had the pleasure of interviewing Nick Singh. Nick is a career coach and the best selling author of "Ace the Data Science Interview". Previously, he held a variety of software and data roles at Facebook, Google, and VC-backed startups.
He aspires to be the Drake of data science, whatever that means. Today we learn about Nick's fascinating career journey through tech, the keys to landing the data science job and his unique experience, writing his book. I hope you enjoyed the episode. I know I enjoyed our chat.
Nick, thank you so much for coming on the Ken's Nearest Neighbors Podcast today. Obviously we've met another incredible guest source from Harpreet's data science happy hour. You're a book author. You're a former, you know, you have a great, incredible pass that I won't lead too much into yet. But again, I really appreciate you coming on the show and having a talk about data science jobs and your experience and whatever that might be.
[00:01:31] Nick: Yeah, man, I've been loving your content. I've been watching your YouTube for a while, so I'm really happy to be here and shout out her ... for helping get this together too. So, yeah.
[00:01:40] Ken: Awesome. First, I have to comment on how much better your background is than mine. You know, I'm like sling it with my bed and like, I barely have my papaya and frame.
[00:01:50] Nick: Yeah, man. No, I'm gearing up for that YouTube life. You know, I'm not on YouTube right now, but maybe sooner than later I will. So I just gotta get the background check for that too.
[00:02:00] Ken: Awesome. Well, if by the time this comes out, you have a channel, I'll make sure to link it in the description below.
So in order to get people a bit more acquainted with you, I think a really unique way to do that is to talk about how you first got interested in data. So was that a single pivotal moment or was that a slow progression over time?
[00:02:20] Nick: Definitely a slow progression, and, you know, I've always been interested in data. And my career is a little bit weird cuz I've done data in a whole bunch of different ways that maybe isn't as traditional as holding the role of data scientists for long periods of time. So in school, my first, so in school I studied systems engineering, which is kind of like industrial engineering or operations research shows this good mix of math and coding and modeling and a little bit of business.
And what kind of always guided me throughout my career has been building, making better decisions through data, right? So I think of myself less as a data scientist, data engineer, software engineer, pretty much the thing that's been guiding me is helping orgs, P teams, people, products basically make better decisions through data.
So that's kind of got what got me into systems engineering. I minored in computer science, I intern for a defense contractor that did work for the Naval Intelligence Research Lab. So for the intelligence community, then I worked at Microsoft. Then I interned at Google's nest labs as a data engineer where I got to see a different side of data at scale looking at all these logs that nest thermostats and nest cameras are spewing across, you know, the whole system.
And then I started my career as a software engineer at Facebook on the Growth team. So technically my title is software engineer, but even that team, the Growth team was all about using data to build better products and to experiment our way to A/B test our way to higher engagement, higher retention. So I joined the new person experience team, which was basically all about figuring out how can we build better products that increase early retention of users so that they'd sign up on facebook.com or the app.
And make sure that they didn't quit the app within two weeks of joining, you know, so that was a lot of SQL queries, a lot of A/B tests and a lot of different like data exploration to figure out like, Hey, where should our product roadmap go? Where are the most opportunities to increase retention and engagement? After that, I switched things up, man.
I'm a big fan of Drake. He says, YOLO, you only live once. I knew, man, Hey, I've been so interested in the business side of things. Let me go, let me go give that a spin cuz you know, I did the whole Facebook software engineer thing. If I needed to go back to Silicon Valley and code, I could do that. So I took a big risk and I joined this alternative data company called Safecraft that does mobile location analytics.
They had this big panel of mobile phones from which they collect GPS location data, anonymize it, and then put out aggregated insights such as, Hey here's how many people are shopping at Walmart and Target, you know, never was localized down to a person. But at that aggregate high level view, it helped people understand what's happening with retail, what's happening with commercial activity.
So there I was actually the head of marketing, which is pretty random, right? Cuz it's like, bro, I'm like technical as hell. How am I doing marketing? Well, it was just a small company. The CEO took a chance on me and the whole job was trying to explain why this dataset was good to other data analysts, machine learning teams and data scientists who work at hedge funds or retail analytics companies.
So it was almost like I'm trying to position this product and sell to people who I used to be on the other side, using data to build better products, right. So I did that for about two years and you know, the COVID hit, people, lost their jobs, offers got rescinded and I've been writing on LinkedIn for a long time.
And I was like, all right, maybe this is my time to be a career coach. And that's where I've done my most recent thing, which is write a book on data science interviews. So it's kind of a weird career in, and around data in different ways through engineering, data science, software on the Growth team, or even running marketing, and now the book, but yeah, I just love data and helping people. So that's kind of my career.
[00:06:11] Ken: Awesome. And so, I mean, it seems like you have really unique experience from these huge tech behemoths to kind of more startup type businesses. I'm interested what that transition was like for you. I mean kind of jumping into the meat of it here, like first, how did you get signed on with like in that, in that marketing role, how did you meet these people? What was that process like? And just tell me a little bit more about the experience between like a huge tech company and something that's a bit smaller that's growing quickly.
[00:06:39] Nick: Absolutely. So actually, it's funny enough. I sent the CEO of Safecraft, a cold email and within 24 hours, he is like, Okay, cool.
Let's interview this kid. And then a few weeks later I had the job offer. And that's actually why I tell so many people about the power of writing good cold emails. It's literally chapter three in the book, a data science interview. It's all about how do you send effective cold emails?
Cause I actually emailed my way to interviews at Airbnb and Snapchat. So it works at big companies and small companies. So I just sent the guy a cold email saying, Hey, like here's my background? You know, things at Facebook are, are right, but I'm looking for a change. And he like, kind of liked my energy.
He liked my portfolio projects. He liked my past and he, you know, I had a convincing story because. Besides just the work I did day to day, I had all these other side projects and hustles and my DJ business that I did in high school that show like, Hey, you know what, even though I don't have traditional marketing experience in different ways, I know how to use data.
And I know how to maybe advertise and market in any ways. This is a very technical product. We were selling data, you know, there's no UI. We were literally selling a CSV once a month, we would drop a CSV to a machine learning team. So, you know, it kind of worked out that they needed someone technical like me to really explain the value of these alternative datasets to other teams. So that's kind of the story of how I landed at safe graft through a cold email happens and sense, and it worked.
[00:08:05] Ken: That's incredible. And you know, something that we talked about a little bit offline is your experience with startups or sort of that entrepreneurial nature. And I think that at least from my perspective, like data science is a very entrepreneurial, like vertical within business, right?
Like you're scoping problems. What is entrepreneurship is your defining problems and you're building solutions to them. Like in theory, a lot of that is what we do with data science. Can you talk to me, I know you had I think it was like a startup or a really in intense project you worked on around rap music. And I'd love to hear this story of that and how that integrated into. Broader picture of your career.
[00:08:47] Nick: Life is wild. So this random thing I did in my sophomore year of college, it was a startup called rap stock.io. It's dead R P but back in the day it was kicking and it was alive. What it was was I love people would get really obsessed with fantasy football, where their draft teams. And I know you're the whole sports analytics guy, you know, all about the fantasy life and how intense people get. Yes.
[00:09:08] Ken: Although I'm absolutely getting dominated this year, my fantasy football league.
[00:09:12] Nick: Right. And guess what, Ken, I hate to break it to you. I don't really like sports. You know what I like, I like music. I like thinking about, Yo,is Drake overrated or is there more mileage in his career or is Lil Nas X one hit phenomenon or is he here to stay? And in hip hop, everyone's always yelling like, Oh, this guy's just a one wonder or, Oh, this guy's the goat, the greatest all time, right? These are debates that are happening.
So I wanted to make people put their money where their mouth was. Basically, I kind of built this fantasy football, but for hip hop music where you kind of draft artist your fantasy label. So this is a little bit ahead of its time because now they do this stuff with like cryptocurrencies and betting markets and all this stuff.
But this was just me in my college dorm room, trying to figure things out and try to quantify your ability to like say who was doing good in music. So what I'd use was Spotify data to kind of price each of these artists in real time so that people could go long or short on different artists and kind of get credit and show off like, Yo, I actually have a good portfolio because look at how well my artists are doing.
I actually know who's an up and comer and who's just, you know, fake in the industry. So I did that. I grew it to 2000 monthly active users and doing this project. And you know, it's funny, I even did this project cuz I used to DJ in high school, right. So I've always been about like music and taste making.
And this was a project around quantifying basically musical taste. That was kind of what I wanted. I wanted to prove to my friends that I knew how to pick 'em better than they did. You know what I mean? And I mean, that's kind of the ego that drives people maybe in sports too. Like, Hey, I know how to draft better than you do.
Cause I actually understand the game of football or basketball better than you do. So I had that same kind of ego. I wanna build a platform, but you know what I fell in love with. I love the music part. I love building, but what I really fell in love with this growth engineering, this idea of like, Wow, I have users coming and with Google analytics, I see their retention data and I can split test things and I can send an email and they'll come back to me and I can build email campaigns.
And I got really interested in. Not just marketing, but this intersection of like, how can we build products better that engage users and do it in a data driven way. And I think these days, the VC's coined the term product led growth, which is, Hey, changing your product to lead growth and not thinking of marketing or design is a separate function from like the actual product, you know, they're all very tightly integrated.
So that's what got me interested that in that little field, my sophomore year of college, and that's kind of what led me to Facebook's growth team way later, where, Hey, on Facebook's scale, I'm running A/B tests with millions of users and trying to figure out like, what, what should the product do next?
So that's kinda the story of how my little side project turned into my full-time job and I mean again, in the book, chapter three is I think two or three is all about this of building kick ass portfolio projects. Cause it's what I tell all the people in data as well, like Kaggle is free. All these datasets are out there for you to explore and build stuff.
And if you can cold email that and sound intelligent about the work you've done and show how it's relevant to the future company. Like I made rap stock relevant to all the growth teams I interviewed with you are in a much better position on the job hunt than other folks who just have the same boring projects that everyone else has done.
[00:12:40] Ken: I mean, to me, that's such a, like an incredible case study in and of itself. Is your experience there? Something, I think a lot of people struggle with is motivation or just like what project should I do and being introspective, understanding what you like and what challenges you face or, you know, even if it's just, Hey, I wanna...
One of my friends, I mean, another guy interviewed from Facebook a couple weeks ago. He started tracking all of his ping pong games, right. And they wanted to see who would, who would perform the best. And they created an ELO rating system and whatever it is or one of my other friends, or maybe I've read an article mine, but they created a analytics for all the settlers of Catan games.
They were playing, right. And you know, like these things, they seem like kind of dinky and it's like, Oh, this is just for us to get us to understand our own stuff better or like competing at your friends or have a benchmark, but they go a tremendously long way in the job market. And it's not like you don't need super sophisticated projects.
You have to show technical skills, right. But if it's something that is like, Oh, this person clearly did this in their free time because they enjoy it. They get immense joy out of like working with data that really comes across you. Like, if you're a hiring man, would you, would you like someone who is excited about doing data work every day and like might need a little work or do you want someone who's just like hates being there, but is like a really good coder.
[00:14:11] Nick: Exactly, man. People hire for like potential and there's nothing more that shows your interest, passion potential than these portfolio projects. And another big thing is most people quit when they're working on their projects, cuz they get bored of it in a week or two, but I didn't quit cuz I love hip hop music and I had users and I was like, Wow, this is so exciting.
And my friends are using it. Same way that guy making ping pong. He's not quitting. Cuz his friends play ping pong. Versus if he did something more academic, maybe it wouldn't have worked as well. So I think that's another super thing that I talk about in the book. Like a hack, which is go use your passion to drive these projects.
Even if your passion is something that's fun, like ping pong or hip hop music that's might not be very serious or businessy. It still will help you make sure you actually put these projects to completion, right? Cause if I did something way more boring and relevant, would I have had any users? Probably not.
I wouldn't have even known about growth engineering or my passion for marketing. And you know, that, that led to a whole bunch of other things happen in my career, but it all started with, Wow, I'll love music. Let's go make something so huge fan of that as well. And got to see that you've seen it work in other people's careers. Well, and it's definitely worked in mind.
[00:15:17] Ken: Yeah. Well, something. I think is I wouldn't say unique about your situation, but something that's really highlighted there that I don't think I've talked about enough is like creating positive feedback loops. So you were able to like, through your interest and your passion create like a spark to get something going, right.
And then you had users, or you had like some system that kept you wanting to like go back to it or hold you accountable. Like if you have people using your platform, you're like, shit. Like I have to yeah. Have to keep doing it, right. And the more you do it, the like more viewers you get or users you get or whatever it is.
I think, for example, I saw that with YouTube, right? Is that, Hey, if I do these specific things, these are specific outcomes that I, that I can get from this. And the more I do these things, it keeps building and building and building on itself.
[00:16:05] Nick: It's addictive, man.
[00:16:07] Ken: Yeah, well, but I mean, there's so many applications of that within the, within the data science career or within the job market itself is that if you get better at, for example, like deep learning, right.
That enables you to do more deep learning projects, right. That enables you to find cooler solutions and outcomes. That again just makes you want to do more and more because your capabilities are expanding, right? Yeah. And you know, I just wanted to call that out. Cause I think that that's a lot of people cut the feedback loop too short. They either don't share. And you know, like I think being social with projects is one of the best ways to continue a feedback loop to get buy in.
[00:16:44] Nick: A hundred percent.
[00:16:45] Ken: But, it definitely is not, it's not the only way, right.
[00:16:48] Nick: That little feeling for me was when I coded up the project, it was a school semester and I saw someone in a lecture hall, two rows in front of me use my website. But they didn't, I didn't know that person. They didn't know me. So I don't know. I mean, this is a friend of a friend. It's not like a total total stranger, but it was somehow, maybe on Facebook, but I was just like, that's my website. And he's placing trades on my website. Shit. Like, this is real now, you know?
And I don't know that guy's name. I, you know, I've seen him around, I don't know him once that clicks. It's like, it gets really exciting and you start growing way faster. And I know a lot of people are like that about data science. So, you know, it gets addictive once you start getting good and it doesn't get so hard and daunting.
So I know for people who find these things daunting, you start at an easier place, lower that activation energy, you know, make it something you like. And it's easy to do. It's way better in the long.
[00:17:40] Ken: I mean, that is so meaningful right there. Switching gears just a little bit, I've been interested in this and wanting to ask questions related to this. So let's take your experience interviewing at Google interviewing at Facebook for data engineering roles and software engineering roles. One, what was that like? And then how are software engineering interviews different from a more pure data science interview?
[00:18:08] Nick: Totally. Yeah, man, there's been so much out there on software engineering interviews, so I don't have any alpha, you know, I don't have too much more to say besides grind, beat code and read books like cracking the coding interview or elements of programming interviews.
I mean, the whole reason I wrote this. Face the data science interview is because I was such a fan of Gale's book cracking the coding interview and how it helped me in my data in my software engineering career. And I was like, why does this book not exist for data science and data science careers?
So honestly the path for interviewing for these data and software engineering roles is similar in the sense that, yeah, you're gonna be asked these technical questions that require some SQL for the data engineering ones. A few of my software engineering roles also ask me some SQL. And then yeah, you gotta do grind lead code and answer questions.
And I would code in Python. So it's not that different. I mean, of course in data science though, they're gonna ask you more conceptual stuff too. And some of the audience might be thinking about like, Oh, Nick, wait, some software engineering, data engineering. Why are you writing a book on data science while I have my co-author as well?
Who's the face ex-Facebook data scientist turned quant on Wall Street to help me out with some of these things, cuz I've been around the data science block, but not nearly as much as he has for the interviewing side. But what we see from coaching candidates and working on the book and just talking to hiring managers, conceptual stuff matters a lot more.
And then actually an emphasis on projects is a lot more there. Cause in coding, you can get signal, Hey, solve this, reverse this link list. Okay, you get some signal there and data science, you can ask people that. That really doesn't map to the actual job, because like yo, which data scientist is even using a link list.
I mean, truth be told what software engineer is using a link list in production. I mean, no one really not money. I know, but even that's still somewhat more relevant than compared to a data scientist where it's like, Yo, ask me about Pandas, ask me about ... ask me about NumPy, you know, ask me something there, ask me about modeling.
Like, what are you doing? Ask me about breath first and depth, first search through a graph, you know, so I think there's an more emphasis on conceptual stuff around stat pro ML. And there's more of an emphasis on projects because Hey, at some level you just gotta ask people, what's your past experience, spin and.
Why did you make the decisions you did? And you can do that in software engineering interviews too, but it's definitely, I've seen that way more for data science because the field's so broad. So it's hard to like, you know, I can't ask you about SVMs maybe a good data scientist, but not know too much about SVMs and maybe that's okay.
You know, like maybe it's been a while since you've read about that technique or reviewed it. So who am I to say? So it's better. I ask you about your logistic regression project that you have written on your resume versus in software. It's sort of like, Hey, even if it's been a while, since you've worked with a load balancer, you should still know conceptually what it is in the last you about it.
[00:21:05] Ken: This episode of Kens neighbors is brought to you by Z by HP. HP's high compute, workstation-grade line and products and solutions. Z is specifically made for high performance data science solutions. And I personally use the ZBook Studio and the Z 4 Workstation. I really love that the Z line can come standard with Linux and they also can be configured with the data science software stack. With the software stack, you can get right into the work of doing data science on day 1 without the overhead of having to completely reconfigure your new machine.
Now back to our show. That's awesome. I think, I think something that I've noticed is very different is for the most part software engineers. I think that there's more of a consistent structure. You know what you're gonna expect. You know, I've a long time ago, you. I was, I took a couple inter job interview, like data science, job interviews, just to see what it was like, I wasn't panic on changing or anything like that. Yeah, yeah, sure. I just wanted to like, get familiar with the, what the market was like and the, the breadth of the different types of interviews that I got were fascinating.
Like one, I had to do like a PowerPoint presentation with like a past project that I did and explain it. And then a couple runs of like pandas related stuff. And like, honestly, that wasn't too bad at, I thought that was like a reasonable process. I had some prep, but I didn't have to like, do a whole nother dataset.
So I had like a take home quiz that was like all statistics problems, right. And then another one, they sent me like a dataset and set out, like find something out of this. And I think that can be really overwhelming for a lot of people. Like it is. Yep. Not knowing what you're gonna get. And, you know, hopefully we can start mapping those and saying, Hey, like, these are like, you know, if it's like, Hey, these are 15 things you can expect. That's better than not knowing the number of finite different things out there.
[00:22:52] Nick: Exactly. Exactly. And that's where our book, I mean, that was our aim. That's one of the reasons we wrote the book was because, Hey, people have, you know, we were like, Why does this book not exist? And one of the biggest reasons we saw was like, Well, how do you speak intelligently about like 11 different things?
Cause there's coding, SQL, product sense, case studies, ..., stat, ML. Like that's a lot of stuff. There might be a visualization part, take home challenges. There's so much. And this is before we even talk about behavioral interview questions or anything like that. So it's definitely hard to map. We tried our best to do it, and we think we came really close.
So it really does, you know, we aligned it with what we've seen at some of these big companies and what we've seen people write about on Glassdoor. So it's hopefully aligned to the market, the book, but definitely as a candidate side, it can be overwhelming because someone's gonna ask you a probability brain teaser.
Someone else is gonna give you a take home project with, to blow and SQL. And then a third, person's gonna ask you conceptual stuff in ML. And then a fourth, person's gonna ask you to reverse a link list. And it's like, Wow, how does one person do all that? Honestly, it's practice. And honestly, practice is one thing.
And the other thing is, Hey, maybe you don't have to get the perfect answer. Maybe being able to explain your thought process, even if you can't write the perfect SQL query is okay. Or maybe you don't know all the different assumptions for linear aggression, but at least, you know, some of them, you know, so that's sort of like that.
[00:24:12] Ken: Yeah. I mean, I would hope that most interviewers do not write people off for a single mistake. Yeah. I mean, I've gone through some interviews where like one of the rounds, I was like, man, I bombed that. And you know, you get to the next round. You're like, how did that, how's that possible? What mistake did they make?
And to be fair, they could have made a mistake. But yeah. You know, like it's not, it's not like this like very clear pipeline. It's not like people in this circumstance are like data, right. We have emotions that are, and like, Hey, if I are really like this person, I think they'd be incredible to work with.
[00:24:45] Nick: Yep. Exactly. So there is variability and there is a wide range of things, but I think we distilled it to basically the most common ones. So hopefully the book achieves that part.
[00:24:59] Ken: Interesting. So if you were If you were interviewing a data science candidate, what do you think would be the most representative way to do that? Like this is maybe outside of scope. Yeah. But it's like flipping the table. Like if you were an employer, how do you think is the most effective way to do this since
[00:25:18] Nick: you've been kind of in this? I would definitely, yeah. I would definitely ask them about SQL and it's okay. Pick whatever flavor SQL you want and I'd make up some hypothetical table and I'd just make sure they understand concepts like joins.
Maybe you can write a window function, just basic queries, make sure that they can do 'em. I would do that. Cuz and you know, you don't even need an ID for that, you know, just put in a notepad or just tell 'em conceptually how you do it. I would definitely include some SQL there and then I would definitely grow them on past projects.
Cause I feel like it's easy and there's a lot of like resume inflation where someone writes like, Oh I did all this and then you. Ask him about it. And it's like, Ah, I was just one person in a four person team, and I really didn't do that much. And I can't really speak to regression, you know? And it's like, Hmm.
You know, why'd you list that? So I think in an era of a lot of resume and LinkedIn affiliation and everyone just trying to stuff in random skills, I would definitely ask a lot more about projects. And then I think a take home for more junior candidates is good. I know that senior talent often refuses to do take homes because they have a lot of other job offers on a table.
So depending on I would administer take home projects, like depending on the situation and the candidate, or if I'm on the fence with somebody and I just make sure that they're able to summarize data, maybe narrative. But of course it depends on the what kind of, kind of role it is and then do some basic modeling.
But yeah, really. Just make sure that they actually know how to work with their different tools and have some base level of communication, which a take home project can test easily. So that's kind of what I've seen as really indicative. And often another good thing I think is that asking people about projects is a good way to start asking technical questions.
So in a vacuum it might be stupid or trivia to be like, Hey, tell me about the math behind PCA. But if someone says they used PCA and did a lot of work with it, or, you know, they did a lot of stuff in regression, I would ask them more and more questions and use that to just see like, Hey, is this someone who's just touched the surface of things or they actually know what they're doing.
And it speaks to their like, you know, that they take their craft seriously as like, Hey, I didn't just import from psychic learn. Like I actually know what I'm talking about. So that's kind of what I would structure in my interview process. Yeah.
[00:27:31] Ken: You know, that's a really interesting like, concept of like probing and seeing how far you can go until they run out of answers. Like I think a lot of, like, I would do that when I would interview, we'd talk about a random forest and like versus XG Boost and I'd be like, Okay, like, what is the difference between these algorithms? What's the difference between bagging and boosting? Like how does that? And it's like, Okay, you know, they get to the one level down and then you ask another more specific question, like, Okay, like what does that mean?
Like related to residuals, whatever that might be, right. Yeah. And it's not to say like, I'm testing to see if you're getting the answer, right. Yep. But it's testing the depth of knowledge, which I think is a very effective interview technique. Yep. But on the flip side, as an inner ..., you're like, Oh my goodness, I got that wrong.
And it's like, no, they're just seeing how far they can push you a little bit, right. Good candidates. You wanna...
[00:28:26] Nick: Yep, exactly. You wanna push here's something else. That's funny if this is a little bit of a. This is a little bit of like a controversial thing, but I've seen in my little like, experience to be true that if a company doesn't ask me and push me hard, I think, Oh, this company has a low technical bar.
And I know definitely for these hyper elite hypercompetitive jobs, it'd always be worse, worse to like, be like, Ah, yeah, cool. You're cool. You know, come on through, you know, then you're like, Ah, do I really trust these team or do these people. So, I mean, if you probe people until they can't go any further one, you learn about their limits.
And two, it leaves candidates feeling like, Ah, maybe I didn't get it perfectly, which is. Which is, you know, it's, it's sucky, but I'm telling you, like, there's a real psychological thing. Like, Oh. And they called me to the next round. Okay. That's really cool. Like, Okay, I'm gonna give this company more of a chance.
You know what I mean? Like even you wrote, I mean, even you said you were excited, like, Oh yeah, I bombed something and they're like, Oh, did they make a mistake? You know? Yeah. You kind of want you candidate feeling just a little bit like that. Now. Of course don't P someone I'm not saying that, you know, you know, these things have limits, but it's an interesting psychological thing I've heard from hiring managers and everything that it's beneficial to push people to their limits.
[00:29:40] Ken: Yeah. You know, on the flip side of that, we have just a lot of interviews that I guess, like, don't really go to plan. Yeah. How do, how do you reconcile something like that when. You feel like it's going off the rails or something along those lines. I've definitely been there. And I, you know, it's like, I don't wanna give up, but like.
[00:30:05] Nick: Yeah. Yeah. So you just keep asking them other questions and you fill the time and then you end the interview at the appropriate time. When, for the time scheduled, I know it's tempting to be like, Oh, let's call it early or whatever. But it leaves a bad taste in your mouth. And honestly, the person who's on the other end, if they feel like the interviewer's not into it might write you a bad cluster review or tell their friend, Hey, this company's not good or whatever.
So you just wanna give some people the respect they deserve. So I've just, you know, if things don't work good, then I just keep it behavioral and you know, it's like, Okay, great. You don't know anything, you know, I just try to get to know the person and just take an opportunity to talk to someone honestly.
And there's no, I mean, and it sucks, right? Cuz you feel like you're wasting your time, but it embarra, you know, avoids another embarrassing thing of like, Oh my God, like. I pro them twice and this ended a disaster. Should I ask some even more hard things? Probably not, you know, just keep it, keep it there, yeah. It happens.
[00:31:01] Ken: So, you know, we're talking about qualifications. I lost my train of thought a little bit back there. Yeah. That wasn't my best question. But you know, we're talking about qualifications, right. And like how you're qualifying the candidates. I think as you become more senior in these roles, You're not just looking for any job, you're looking for a job that is really meaningful to you.
And I think it's important to qualify the employer as well, with questions and things along those lines. You know, how do you do that effectively? And what are some, you know, you'd mentioned like a red flag is like, maybe they're not grilling me enough. This is a cake walk. Yeah. That could mean that they're like the work is not gonna be meaningful because they can't ask like intelligent questions about right.
How it would be a fit for it. You know, what are some ways that me going in as a, like maybe a senior data scientist, I can say, Okay, I think this would be a good fit for me. They have certain criteria. Like, what should I ask? How should I approach that conversation?
[00:31:58] Nick: Yeah. So I often find that the best opportunities at companies that are growing fast and now of course it depends on if you're prioritizing work life balance, but you know, I'm a Silicon valley mindset, you know, and I'm interested in startups and working at these top tech companies.
So usually growth. If a company's going fast, that means there's a lot of opportunities. And I feel like work environment wise, it stops being zero sum. If there's more work to be done, people are less territorial. There's less politics or just. Just do more work and people don't fight. I feel like a lot of the fighting happens when there's not enough work to go around and then people are fighting for credit and there's not, you know, people are worried about layoffs and they're trying to compete with their coworkers versus a startup is just like, do whatever you want.
Like we have too much work, so whoever's good. Do more work and you'll be promoted and recognized for it faster. Cuz anyways, we're hiring so much. So I think for senior data scientists where they can align themselves with a culture of high growth, it's always good. And actually honestly, that's for basically any candidate, I would always tell people to optimize for growth rate and like try to assess that.
So it's a little bit biased, right? Cuz I mean, I know I took the, an question in my own way of like how do you look for companies that are growing fast because you asked like how do you just make sure it's a good fit? I think that's too general cuz you know, maybe someone's looking for a work life balance.
Maybe it's good that they didn't grow me. You know, maybe it's good. It was really chill and really casual and they're like, Oh when can you start? You know, come on through. It's really fun. Maybe if that's what you're looking for, maybe it isn't a red flag, but for me looking for like an intense role generally that where I can learn a lot, what I find to be good is to talk to the people in the interviews and get a sense of what are they working on actually.
So usually it's like, Hey, tell me about like what challenge you're facing right now. Usually in the, ask me a question part, and then I also try to understand the leadership and their mindset. Well, so where I can, I try to say like, Hey, you like me? Can I talk to not you, but your boss or your boss's boss.
And usually if you're a good candidate, you can like go up a level too and have them try to close you and try to understand from their perspective, what does the work you're doing or the team you're gonna be joining fit into the larger organization or branch or the whole company's. Objectives, you know, and that, that can give you a good sense of like, Yo, is this work gonna be meaningful or is this some random work that's gonna get canned?
And this is some random like R&D project that's probably not going anywhere. You know, you can get a sense of that when you jump up a level or two, and usually you just have to ask for it. You're not gonna get like a VP, maybe talk to you if you're just joining as an entry level role. But if you ask for it, they will make it happen, especially if you're a good candidate.
So just talking to folks, who've been a really good thing. And then just seeing like their, like seeing the rate of employees joining on LinkedIn and seeing like, you know, if there's only 50 people see, when were they hired? Is this company hiring fast? Are they growing or have most people been there for a while?
And then same way see their ex employees, did people just stay for a year in dip, have a lot of employees dip recently, or as generally people are with the firm and they're hiring fast, you know, that tells me there's a good, the company's growing well. And it also tells me like, Hey, probably my career will grow well too here.
[00:35:04] Ken: That's awesome. You know, I think. Using data sources and like collecting your own information is really valuable. I get asked by a lot of people, but Hey, should I do a master's in data science? Right? Like what program should I do? And I'm like, I don't know. I like haven't gone through every single program out there to tell you if this one is like a high quality program.
Like in order for me to feel like I've given you, like, I could tell you if you should take it or not, I'd have to take it myself to see, Oh, what's a quality, but right. But the, but the point is that I always tell them, find a graduate of the program and like shoot them an email, or like shoot them a LinkedIn message and be like, did you find this as a valuable experience, right?
And you can do the same thing with a job. You can say, Hey, this person used to work at this company. Let me, they're probably gonna give you a realistic appraisal of their experience there if they were a data scientist, right. And you know, if they left and they still say really good things about it, probably a good sign, right.
[00:36:04] Nick: And see the tenure. That's the other thing like, Hey, have people been around for a little while, is it's a more mature company or are they just churning through talent? And then often people write about their own promotion or you they'll hold multiple roles and see like, Oh, this company, this person just had the same title for four years, or they just stayed for a year and left.
But recently I was looking at a company where I was like, Oh look, multiple. These people have been promoted once every year, year and a half. And they've been with the company for 2, 3, 4 years. I saw another company the other day. I was helping someone with their job hunt where someone interned at the company work somewhere else for nine months and then came full time back to that company for four years and then left to join a very well known fortune 500 company.
I was like, Okay, great. Like clearly this company's doing good. Sure. They're more regional smaller company, but clearly they're able to retain talent for multiple years. Send that talent to other bigger companies later. And clearly this guy interned there. Did some other job and then came back and like had a really solid career.
So clearly they're able to retain talent and like this guy didn't just come back randomly after his internship for some random reason. So it's just like little signals, right? So it's a mix of qualitative talk to people, quantitative, look at these people's tenure and it's all on LinkedIn, man. You just had to mine it a little bit.
And I know we're all data scientists and data folks who want big datasets, but sometimes you just gotta look at 30 people's LinkedIn profiles at a company, put it in your own Excel sheet and just see what it is, you know? No, no Spark, no Hadoop, you know, just gotta put it in Excel and calculate yourself and see what's happening.
[00:37:38] Ken: Yeah. Well, I think that that's a really important point is that, well, people put so much effort into their projects, right? They put so much effort into interview prep. Yeah. And they forget. They eventually have to do a job afterwards. Yep. And it's like all about getting the job, but they don't, I don't think people do enough homework. And if they'd really like the job, right. You're like, Okay, go ahead.
[00:38:01] Nick: No, I've seen the same thing. And I think few things to people for people realize like, guys, I wish I was cool doing all these fun things, but I spend most of my waking days make waking hours working just by the nature of I sleep a lot.
So when I'm awake, I have to, you know, clock ins for some time, right. So it's like, if you're gonna spend so much of your waking hours, basically your life, that's not asleep working, let's do some better due diligence, right. And people are like, Nick, you're gonna look at 30 LinkedIn profiles. It's like, Yo, like you guys spend so much time researching a company.
How are you not gonna look at every person's profile? See their tenure, write it down. Especially if you have two offers in hand. That's one thing that I tell people. And then the second interesting thing I tell people is like sorry, I'm losing my train of thought here.
[00:38:47] Ken: No, no. It's okay. Well, like the flip side of that, I think we touched on it a little bit is like during the interview, I really think it's important to like ask qualifying things. Like, you know, they're interviewing you, but from a psychological perspective, if you're also interviewing them that, you know, it's like dating, right. If you have standards, there's a higher likelihood that someone else is gonna be attracted to you. You're not just getting anything that you could get.
Sure. And like, figuring out how to ask those questions, to see if this would be a good fit for you. Like, Okay. Like you know, how much do you have unlimited PTO? That's something that I care about. That's something that's meaningful to me. You know, I'll have to think a lot more if that's not something that you offer or if you don't have like a certain type of plan or like health benefits or even related to salary or the type of work it's like, look, I don't want to be able, I don't wanna be touching XYZ system because I think it's antiquated.
Yeah. If you're gonna tell me I'm gonna be, be using you know, again, X, Y, Z, like every day on the job, like this isn't a good fit for me. Yep. I think that people don't realize that you can be, I mean, they're paying you a lot of money, like yeah. Like they want you, they need you just as bad as you need them. Probably even worse. Yeah. Like you should be qualifying them as well.
[00:40:08] Nick: Put one nuance here, cuz I know this is a little controversial. I hear what you're saying. It's a two-way street. I would just let people know. In my own perspective, I try to like back away from that mindset just a little bit until I have the offer, then it's totally a two-way street.
and here's the reason why, cuz it's so easy to play, to lose where it's like, Ah, I didn't care about the job anywhere. It's like, you'd start tanking. And it's like, Ah, but that guy was a little mean, you know, I don't want this job anyway, you know? So I always try to say that you are interviewing them.
It's a two-way street, but like in the beginning stages focusing, focus on getting the offer. Because if you start letting yourself be like, Ah, but I didn't really want the job anyway. It starts to kinda that that affects you in your interviewing, you know, and I'm always like, you know, if you have the free time and you're actively interviewing, try to focus on getting the offers and then later worry about like, once you have the job offer, let's do our due diligence.
Let's call up people. Let's not do that in the middle. Let's first focused on getting the offer. Now this is for more junior talent. I can understand how someone in your position, someone more senior who, you know, has a bunch of stuff to do and is not applying willynilly. Maybe it is more of a two way street, but definitely for junior talent, like suck it up interview as perfectly as you can then do the research at the tail end where be a little bit more critical about the role, because you know what, you know how early we said, it's all about that passion, enthusiasm.
right. It's like if you start getting in your own brain a little bit, like in one interview, like, Ah, I don't know if I really liked it. You know, one person was a little weird and you know, it's hard to be enthusiastic in the rest of your interview process and maybe you don't have to work with that one weird person or maybe that one weird thing was just a very minor detail in the grand scheme of things.
And they're offering a lot of growth and a lot of salary, maybe you should have been enthusiastic throughout the whole thing. So that's kind of why I'm always like, Yo, I'm always trying to be enthusiastic about the opportunity throughout the interview process and then only getting that critical two way mindset once I have the offer in hand.
[00:42:05] Ken: I like that a lot that makes it tremendous amount of sense. And again, the I'm speaking probably more for sort of like that senior position which is very different from the junior position, right. And to that point, how do you feel about like breadth versus depth related to the job search on that front? You know, I actually, for all people, I generally. Recommend a little bit more depth because I think, you know, I, well, I'll let you touch on that and I'll get, then I'll share my,
[00:42:35] Nick: I'm a little biased, cuz my breath has been crazy, right? I've done some marketing, some data science due engineering, software engineering, written a book, just all these random things.
And in my own life, I think breaths work really well. I think, I don't wanna say that there are two types of people, but I mean, there are definitely some people who are oriented towards breath who are just insanely curious, imaginative who are always tinkering with a bunch of different side projects for those kind of people.
I just wanna let them know breath can work. People tell you to specialize. And I think there's this famous, you know, quote like specializations for insects or something like that. I mean specialization does pay dividends, but I think in this kind of crazy world we live in, that's evolving really fast.
There are games to be held. For generalists and people who have a breath of experience, right? Like I always tell people with the book, I'm definitely not the world's best data scientist. I'm not the world's best writer. I'm definitely not the world's best career coach in college. I didn't think I'd be a career coach.
Maybe I thought I'd be a data scientist. And I definitely didn't think I'd be a writer, but I'm the weird person who's sort of into all three and was able to make the book. And now the book's been doing well. So I'm always telling people like, Yo, especially earlier in your career, get that breath, I think later in your career, if you know what you like, and you're just like, Yo, I love modeling.
And I don't like exploratory data analysis or I love, you know, technical, you know, being an evangelist maybe day to day coding isn't for me. And I want a more talk about developer tools and data science tools, you know, that might be fine later. So I think it's age slash career dependent. But I think that like, what's been interesting is breath has been a good strategy.
And I think society pushes you towards depth, cuz that's much more of an explainable path cuz no one can explain to me early in my career, like why would one person wanna learn about software and data and writing and like public speaking, like what does that get you? Like? What kind of job is that?
But you know, with the internet being so big and random niches occurring every day. It might not make sense immediately, but like breath can pay dividends, especially as long as the breaths and things that are all valuable, you know? So as long as you're getting, picking up valuable skills. So I think depth is a very easy case to argue and I think it's what society pushes you into.
It's really easy to specialize in one skill or one tool set. So I don't have anything against depth and I think people can earn very good livings through depth and have really good careers. But breath, man, don't sleep on breath. Cuz crazy thing doesn't happen that you can't, you can't plan for, you know, it's hard.
How can I advocate for something I can't even plan, but I'm telling you if I'd seek depth, I would still be a senior software engineer at Facebook doing growth. That's all I could talk about and sure I'd know a lot about it, but I would have no book. I wouldn't be able to be a career coach. I wouldn't have any of that. And that's what gives me so much value, you know, today.
[00:45:22] Ken: I love that. I think that for general life advice, you're spot on, like, how would you know what you like if you don't try a lot of things. Yeah, right. Like that's how I experience food. That's how I mean, obviously career wise, I've done something very similar on that front as well.
Like I, would've never known I enjoyed teaching or making videos if I didn't try it and go, go outta my way. I think I'm thinking more in terms of the, the specific job search, you know, rather, rather than sending out resumes to every company under the sun. I find that if I say, Okay, I would likely be interested in working in these four sectors.
I'm gonna, at least for this role narrow my search to these four. Then I can do one project for each four sector that like those project. Are very localized to the roles, which probably give me a significantly higher percentage chance of like getting interviews, getting jobs, whatever that might be. I generally think, I don't think you should like narrow and be like, I only wanna work in this position and I'm only gonna apply to Facebook because like probabilistically, you cannot
Yeah, exactly. If you don't get it, you're screwed, but no, but yeah, there there's like that trade off that I'm interested in between saying, Hey, like let's narrow it a little bit. And then like, you know, I'm not like sending out 700 resumes, whatever that might be.
[00:46:45] Nick: Yeah, no, definitely a big fan of what you said. And you've mentioned four projects. Let me show you how one project I did with rap stock can be spun into a bunch of different things, right. So I can show you. This way of like, I had one project that I talked about. I mean, I had won a hackathon. I'd done other projects, but primarily in interviews, they really only asked about this one project.
And of course my internship experiences, but this is the one project rap stock, the hip hop stock marketplace. Let me show you how that's spun with consumer oriented companies like Uber, Airbnb, and Facebook. I would interview with their growth team and I'd write cold emails to their head of growth, through head of growth engineering, showing them, Hey, look how neatly my project aligns to them.
Then when I talked to financial companies, I was like, Yo, check out how I made a stock market for rappers. I'm interested in FinTech and finance and look at this ledger I made blah, blah, blah. And I'd send that to Stripe and Venmo and a few other companies like that plat. And then I'd take that same project rap stock and be like, look how I'm pricing things.
Using alternative data from Spotify. I wanna work at your alternative data company or your consumer credit company or your like insurance tech company. That's pricing insurance premiums through alternative data, cuz that's the same kind of thing I really am interested in. So I think even just one project spun different ways can work really well.
But notice, you know, these are still niche things, right? I'm still not spraying and praying to random companies. Like I don't have anything random to tell CVS health based on my rap stock project, right. But I do have something to say all consumer companies, you know, Snapchat, Airbnb, Uber, they all have growth teams.
They all could have used someone like me to help them out. So I think there's this magic of yeah, don't spray and pray, but also, Hey, don't knock that one project could be spun so many different ways. And in each email, the person's like, whoa, this person's born for this role. Like they know some data, they know some engineering, they built this really cool project.
Like, Okay, we're gonna interview him. So I think there's magic to be done even just in one interview, as long as it's very compelling.
[00:48:48] Ken: Well, that that's something I really wanna highlight in. What you just said is, I mean, you kind of beat around it, but the story that you're telling is so important, right. When I look at someone's resume, what you're seeing is a story of their career history.
And the next step on their resume, that is not there should be at your company, right? Yep. If they've built their resume correctly, or they've told their project story correctly, or they've, you know, given their two minutes about themselves story, if you're not thinking that, Hey, like the next logical step is for them to be working at my company, like that person is doing something wrong.
[00:49:26] Nick: Exactly. And you, it's so funny to see people botch. Tell me about yourself. And actually we have a whole chapter in the book, a data science interview on behavioral interviews. And I give people this three step formula for how to ace that. Tell me about yourself question. But the first part of the chapter is, Hey, these interviews, behavioral interviews are not fluffy bullshit.
Like there's a real art to doing well. And you'd be surprised at how many people tell that kind of story poorly because every interview asks, tell me about yourself and that story. The three part formula we give in the book is, tell me about your past. Tell me about your current and then talk about the future and selectively pick things from your past and current, that map really neatly to a future where that future is exactly the job and the type of company you're being offered today.
You know, so you wanna make it look inevitable when you tell that story such that wow, it's lined up neatly for what we have today. So I just wanna give that little plug because so many people mess it up. And what you said about like, Hey, you want to talk about how this is perfect for the job, super clutch advice for answering these behavioral interview questions. And almost everyone that gets asked, tell me about yourself, so.
[00:50:41] Ken: Yeah. Well I think what a lot of people don't realize about the behavioral interview also is like, They're assessing skills, right? It's not just like, Oh, do I like this person? It's like, well, how much has this person like analyzed the company?
How much do they know about our business? How much homework have they done? How much research have they done? I mean, that is the first thing that I always ask is like, Hey, what's our company do like, can you explain it to me? I mean, like, it's kind of a jerk move, but it's also like, if you're working in a company, you should know what they do, right. How do you explain, like, what are lines of business? And you know, that is something I guess it's behavioral, right? Yeah. I mean, that's not a technical question. I guess in theory it could be, but...
[00:51:22] Nick: No, and it's kind, it's like, Well, one technical skill, it might be testing as communication. You're a data scientist. How are you gonna communicate your findings? How are you gonna advocate for this business recommendation? If you can't even tell me about yourself neatly, if you can't even tell me the story about you, how do I expect you to tell me the, you know, tell the CEO the story behind the data and like, put that into a neat presentation and.
Support this narrative that drives, you know, create a narrative out of the data that supports this decision that should be done. If you can't tell me about yourself or answer something like, Hey, what does this company do? How are you gonna do all these things? Right? So people are really quick to discount.
Like, Ah, I'll just wing it. It's just about myself or, Hey, what's there to improve in my communication, but Hey, it, it, it really is some, there's some real meat behind these questions and real, real value of becoming a better communicator and practicing these things. So you can show interviewers like, Yo, I've got the communication skills of charisma, the people skills of passion to do this job well, because honestly that's what makes a lot of this. Success as well. It's not just your modeling ability or your ability to do Pandas. You know, there's a lot more there.
[00:52:28] Ken: You know, that's something I think would be funny is if a behavioral interview, maybe half of it was over slack or over teams or something like that because people don't realize most of our communication in the workforce is written.
Right. And how effectively you can communicate via just like messages or emails, I think is really, really important and overlooked. Definitely. Obviously person to person communication is great, but like the written medium, I put a lot of stock in if people write a blog. Like I look at that. I mean, that is 100%.
[00:53:03] Nick: Yeah. Storytelling is everything. And that's how we communicate. That's how we make big decisions. And I think what's interesting is what you're alluding to this written interview. If you are in interviewing for an investment. Associate at a VC firm. One of the quick things they say is, Hey, write a one page memo for why we should invest in any company, you know, and you get to pick your company, but you have to convey like why.
And it, it comes in a one page memo and actually one something interesting at Safecraft was I was giving a written interview and why we do that is to make it even because, you know, maybe some people just aren't great communicators. And let's talk about, you know, maybe a software engineer doesn't have to communicate so well compared to a data scientist.
Who's such a cross functional role. So yeah, we would do that to even the playing field and just be like, Yo, can this person write. Neatly enough, you know, nothing crazy, but just like, you know, three, four sentence answers to like five questions, you know, to just level the playing field of like, Hey, maybe you're just more Wellar, you know, you can articulate better in writing, but it's such a crucial move.
And yo, my teachers would always tell me like, Oh, when you're in leadership or management, you'll be writing a ton. Like maybe this math stuff, won't be so important. I'm like, nah, I have you guys like I'm, I'm gonna be writing code till I die. I had no intention of writing a book and here I am nowadays day to day writing, way more in English for the book and blog posts and content marketing rather than writing code, so.
[00:54:26] Ken: I know it's funny how that works. I just a reminder, everyone that will be giving away five of "Ace the Data Science Interview" books. So just be sure to be tuned in to how we're gonna deliver that. That'll be in the description and in the pin comment around how those will be administered.
Before we, before we kind of move on, I want to. Touch on the book just a little bit more. And maybe what's not in the book, right? Like what have you learned since writing the book that you think could make an addendum or a second book that, that maybe surprised you.
[00:55:01] Nick: Okay. So I'm gonna let the audience in a secret. So we self-publish the book, which means we can update the book continuously and nobody knows this. So basically we've updated the book like 17 times since it's come out in three months and we will continue to keep updating it. So, yes. The page count literally keeps increasing. So literally, and we tried to do this, like the Silicon valley mindset of quick iteration, like yes, the book, like anytime I learned something, we're putting it in the book.
So, we're always catching mistakes and adding new value. So, a little bit of a cheat to say, Hey, we haven't learned anything. It's all in the book. But I think one interesting thing just through talking to people and I mean, it's gonna be hard to distill this into the book. Like it's not gonna make it in is mindset, how important mindset is.
And I don't wanna talk about this woo stuff. Like, Oh. Nick, what are you talking about mindset, you know? But like, what's been really interesting in coaching different people, helping them out with interviews. It's just like how some people just. Hey, I can't do it. So they don't even try. And how other people are like, Yo, I'm gonna make it work.
And I'm gonna listen to you, Nick and Nick, you say to do 201 problems in your book, I'm gonna do them. Okay. Maybe, I don't know all of 'em, but I'm gonna do a hundred. You're still far beyond the person who was like, Ah, yeah, I got the book and I didn't really look at it cause I ran late and this and that.
And you know what, I just kind of gave up and you know, it was a little hard and whatever. So I have, I've been really surprised by how important mindset is and like how, you know, they talk about like, think about like a graph, like a line there's like a why intercept and there's slope. And people say like, judge people on slope, not on why intercept don't judge people on where they start judge them on how fast they're growing.
And day after day, I'm more impressed by like, Wow, there's some people who just don't have great skills, but they grow really fast. because they're into it. And it's a mindset thing. It's not cuz they're like 50 IQ points higher. It's just like, they actually give a shit and just giving a damn is like actually something that I've just been very surprised to see like, Wow.
It correlates really well with who actually does well or not. It's not about your baseline knowledge. You're coming into this process with it's it's been mindset like when does one person give up and when does another one? And I don't know how to teach that. I don't know how to say in the book like guys, trust me.
If you can work through these problems, not give up, actually do the cold emails, not just read about them. Be like, Ah, this is really creative. I'll send two, but like actually send 30, send 50 cold emails. I want to put people in that mindset of like, Yo,it's possible, cuz yeah. We've seen it happen. So that's ones that one thing that's not in the book that I just don't know how to communicate, which is why these kind of videos work so well.
Cuz like maybe me telling it to you and maybe can you chime in? Have you seen that too? I mean us telling it to you right now maybe does better than be like, Oh yeah, just read the book and it'll work out.
[00:57:51] Ken: Well, you know, I think that that's definitely something you can figure into the book. I mean, something yeah, that is related to that is having is like how you quantify quantify this process. From some data from, I think it was Sharpest Minds, they're saying that around like 2% to 3% of applications get even an interview, right. If you're going through a job board, right. If I'm going through a job board and I'm getting 6.5% of like a callback rate, right.
I might think that that's terrible. But if I'm basing it off of like a line in the sand, that baseline that's actually really good. It's like double than the average is. And people get really discouraged when they don't actually have a reason to be. And I think that that's like a psychological trick is that we're basing a lot of our judgements off of like flaw to metrics or the lack of metrics and using data.
We can actually create a better mindset or a more realistic view of the world. I mean, in some sense, right? Like this process is a numbers game. Like if you do the right things and you apply to enough jobs and you don't stop, like I see it as being almost impossible for you not to land a job, right.
Like the only way you don't land a data science job, unless you just like. Or doing literally the worst work ever, constantly. and like, can't figure out how to improve, right. Is to stop interviewing or stop applying, right.. Yeah. And I don't know, to me, I think that's like a hopeful thing it might take. So you longer than other people or whatever.
[00:59:36] Nick: I wanna put a third thing, interviewing, applying, and then also reflecting on what you messed up, you know? Cause people are like, Yo, you just don't know these skills or it's like your SQL is trash. Let's go, let's go improve that. But as long as you keep doing it, Hey, six months a year, like people are able to break into this because it's not rocket science.
It's just SQL and SQL is SQL can be hard. Like I don't wanna knock it. And I know a lot of people learning SQL, but I'm just trying to say like, Hey, SQL was designed to be kinda accessible to business stakeholders and like a wide variety of people. I don't know if you know that, like I think they took truck drivers and said, Hey, can we teach the average truck driver SQL in a week?
They wanted a standard where it's like, Oh, so that's why they called from and select select cuz that's exactly what it means, you know? So I dunno. I feel like this feel. So that's why there's not anything too crazy and SQL naming wise, you know, where means where like or an an like, there's nothing crazier than that.
Right. And that's like the foundations of a lot of what data science and data analysts do. So I dunno. Yeah. I just definitely think that people shouldn't give up and be demoralized too early. And that's where then mindset comes in is like people who believe in this felt and themself and like don't give up early, keep sending hold emails, keep building projects and iterating. Like they're able to achieve crazy things. And that's where that slope comes in of like, they just keep improving versus other people just give up.
[01:01:02] Ken: So that last point you made about feedback. That's something, honestly, I've always struggled with, how do you ask for feedback after an interview and actually get it? So a lot of the time I've asked feedback, I've sent an email that just crickets. Like...
[01:01:15] Nick: Yeah. I mean, companies don't give feedback cuz of like lawsuit reasons, which honestly I think is the stupidest thing cuz you know how coffee has to say like caution hot cuz someone spilled coffee at McDonald's and then they got sued for.
So, but, you know, let's be honest, there's enough cafes. Now that serve coffee that doesn't say caution hot, right. So same way. And I'm sure someone somewhere sued some company cuz they got feedback. They didn't like, but I think most companies should be giving feedback, especially for those take home challenges.
It's brutal to like have someone spend six hours on a project and then just like ghost them. That's terrible. So I don't want people to see that like companies to do that. So any hiring managers, please do not do that. But speaking back to that point of how do you get feedback? I don't know. I mean, unless companies change, I think one thing is, did you know yourself often where you messed up.
I feel like people kind of know. So I think that's one thing that people kind of kind of know where they mess up too is maybe you didn't do anything wrong and they're just stronger candidates. Like that's something you can't account for. And it's a mindset change of like, Hey, sometimes I will go head to head with someone who just has a few years more experience and I did fine, but they did better.
Or they they're just as fine, but they're more conservative pick cuz they had more years experience. Honestly. There's not much you can do. So feedback wise, I feel like feedback has not been a bottleneck for most people. Usually, you know, you work through these 201 problems. They're hard enough where people know like they're no one's solving all the problems then bombing interviews.
Usually this book, you can only work through 40 or 50 of 'em and you're like, Yo, I can, you know, I can only get through so many of 'em before they start getting really hard, you know? So I think people, I think it might be too much of a cop out to be like, Yo, I didn't know where I went wrong. I think it's easy to figure out your own balance.
And. I think you just gotta build some taste, right. So how do you build taste? Like how do you get more critical on yourself is to see what is good for take home challenges. Go look at Kaggle and look at what other people's notebooks did and see how they explore data and what they took into account or in SQL.
See how other people solved hard or SQL questions. And you can do currently for ML, see how people explain, like I'm five on Reddit. You know, people always explaining concepts, like see how other people explain PCA or SVM intuitively, and try to write it down yourself. Like, look, you know, SVM, you know, what a PCA is, but could you write about it neatly in like three or four sentences and explain it to someone who.
Might not know what it is like that's, that's hard. Even I would struggle to write about it neatly in like four or five sentences. Talk about dimensional or reduction, for example, you know? So I think like, just build good taste, see what else is out there? Cuz there's so many talented people putting up their work for free on GitHub and Kaggle and where else.
So you can build that own taste and that's really how you level up. I don't think I don't want, I don't like that mindset of like, I would be perfect. If only someone could grade me and like, give me real feedback, you know, I think it's like a lot more inside that you can do like a lot more initiative. You can show yourself to do it. Yeah.
[01:04:15] Ken: Awesome. I really like that. So those are all my questions. What do you have going on? You know, feel free to plug your stuff. Talk about anything that is coming up in the pipe.
[01:04:26] Nick: Yeah, guys, I'm excited. I mean, just mostly promoting the book and iterating on it and improving things, you know, as I mentioned, we're just constantly iterating. So if you guys have read the book and have feedback, come reach out to me on LinkedIn and yeah. Just excited. Get a YouTube channel going at some point soon then later you might be, see me hitting some dance moves on TikTok. Do a little bit of a, you know, shout out some data science concepts and then hit the Dougie.
You know, I don't know how these mil you know, Gen Zs do, but I'm gonna figure it out. So that's, what's exciting. But yeah, guys connect with me on LinkedIn: Nick Singh. The book: "Ace the Data Science Interview", check it out on Amazon. If you've been reading it, let me know how it's going. Write a review, but yeah, those are all the. But, yeah. Just been really excited to do this with you, Ken. Thanks for having me.
[01:05:11] Ken: Yeah, it was a lot of fun. Thank you for coming on. Just remember that Nick should be doing his own research and figuring out how he could improve. So give him feedback, but just remind him of no, I'm just kidding.
[01:05:22] Nick: Yeah, no, I mean, honestly, y'all tell me what you guys were vibing with on the comments down below on YouTube, or tell me what kind of didn't vibe, you know, cause I'm always trying to hungry to improve, so.