If you want to land a job in data science, then this podcast interview is for you. Because today we have Nick Singh on the show. He's the best-selling author of Ace the data science interview, and also the founder of DataLemur. And he helps him breakfast SQL questions and data science interviews in general. He has held various roles at Google and Facebook and came over 130,000 followers on LinkedIn. And that's for a reason because in this in this interview in his podcast, he shares a lot of insights on how to get the interview in the first place. And then how to ace the interview, how to make sure you tell about your experience in an understandable way, in a concise, concise way that appeals to the interviewer also he's going to share about how to communicate your insights, how to communicate your data in a way that is understandable, especially when you talk to non technical audiences. So basically, he in the interview, he shared how to explain technical concepts, like p values, and confidence intervals, like you're talking to a five year old. So a lot bags in his interview with practical insights and funny stories in his interview with mixing, enjoy this episode.

Hi Nick, welcome to the MindSpeaking podcast.

Nick Singh:

Gilbert, thank you so much for having me.

How did he come up with the name DataLemur

Gilbert Eijkelenboom:

So where I would like to start because the name of your company so you found that data linear and you're it's very triggering for me the image of linear, a little monkey and I was wondering, how did you come up with the name?

Nick Singh:

Yes. I love the movie Madagascar. It's like a Dreamworks movie. They have like several sequels and they feature a lot of numbers, because they're native to Madagascar. So first of all, I already love to merge and there's even a spin off show called around those numeric characters and blanking on the name but it sounds like Netflix. And I used to watch it as a kid. So I love lemurs. And then I you might mean had to have the word data in because and that's the key thing is helping data scientists and data analysts prepare for their technical interviews. So I saw other companies with random names like data, bricks, data dog data camp. So I'm like, alright, I'll be data something. And then I looked at a list of animals and domain names that are open and remember was open, reverse work queue. Lemurs are easy to brand. And there's no other like kind of connotation for lemur, you know what I mean? They're kind of rare animals. So I was like, alright, let's run with it.

Founding DataLemur

Gilbert Eijkelenboom:

Yeah. Awesome. So that's where he's coming from. And it's, it's very catchy to me, it's memorable. It sticks in my head. So I think that's one of the benefits. And of course I'm curious about just about the name, but also how are you how you got there. How do you come to founding Datalemur?

Nick Singh:

Yeah, it was a pretty natural evolution to start dating lunar. It's through my own book that I co wrote with my buddy Kevin, who was an ex Facebook data scientist turn on Wall Street. We wrote a book called ACE the data science interview, and it sold 16,000 copies in its first year. It's all about technical interview prep. And the biggest feedback our readers gave us was look, your SQL chapter about SQL interview questions. Asteron data views. It's great. It's fun. To do these on paper, but like, real interviews asking these questions with like a coding editor or through a technical assessment platform. So I want to practice the same content online. So I'm like, Alright, okay, whatever. So, you know, we keep hearing the feedback, but after a year, I'm just like, Alright, I've heard this so many times. That's probably the best next thing I can do with a boat because, you know, I'm just trying to mission to help data people land more jobs. And it just seemed like, hey, that time would be better spent on making an interactive and that's kind of how you do anymore was born because the big sell for data and brands look for hundreds of different SQL interview questions. You can practice them online, write the query itself, get it graded, and then see how other people solve the query as well, in the discussion boards.

Gilbert Eijkelenboom:

Yeah, and I can't imagine that interactive art and, and making a spectacle makes it very useful for people because just reading it from a book, of course, you can learn a theory and you need to practice to in order to get better and I'm, and I'm sure that helps so much in order for people to let those interviews and later also, we'll talk about you know how to how to land those interviews. How to reach out to people because you're very good at very good strategies in cold emailing and also how to communicate an interview. So I would love to dive into that a bit later. Absolutely. Yeah, sounds great. Great to see the success with your book because this is really helps people with a concrete problem, right? How do you how do you deal with those interviews? How did you get the show up? And how do you solve those SQL problems?

Nick Singh 5:50

Yeah, exactly.

Presenting at Fox News

Gilbert Eijkelenboom:

What else I saw is that you're you've been on Fox News, right? And and Dr. To talk about COVID data, so can you tell us about that? I was that?

Nick Singh:

Yeah, before I took the leap to work on the book and data lemur and just kind of take this full time career around helping data scientists look at jobs. I worked at a geospatial analytics firm called Safe graph and safe graph what they had was to see you need access to this dataset of GPS based location data around from two coming from like a panel of smartphones. And they also had data around what are all the stores in the US and where are they located and they located? I mean, like the exact polygon footprint, the exact like bounding box for where does Walmart start and where does it end? So if you know where Walmart exactly is, and you know where the smartphone devices exactly are, you're starting to be able to understand how many people miss it. What stores, right, and that's what Seacraft specialized in. So we were able to kind of repurpose this data set to understand how people stay at home or not. It's really funny, like, while maintaining privacy we can understand if you're at home or not based on home is defined as where your cell phones sits in the nighttime hours, most days of the week, right? Because that's our concept of the home. So if we have 20 days of location data, and we notice like oh for like 20 of those days, you're in the same spot every night from like 1am to 5am You know, that's a pretty good guess. And that's where your home is going to be make sense. So we're able to do is meet this data set called like staying at home index, which helps us track like, what counties are staying at home. So of course we wouldn't ever reveal like person level data, but at a county level at a district level at a state level, we're able to understand how are people complying with stay at home orders, which we might forget about now. But back in 2020, that was the talk of the town like oh, people have to stay at home and like wow, what's going to happen to retail what's gonna happen to malls, what's gonna happen movie theaters if everyone's staying home? That was just having the towel. There's no more business aspect. Of course, there's other ramifications of COVID. But in the retail community, and there's customers we had that was their concern in regards to COVID and their business impact. So yeah, we were able to make this dataset was really went viral. People kept linking to it all the news media picked it up, because it led to all these very interesting stories about like, Oh, look at Texas. They're not staying home but California is staying at home, which is cool because Californians are being more healthy except, oh wait movie theater traffic today in California has dropped 90% and in Texas only dropped 50% And yet these x's you know, the California movie theaters are getting COVID stimulus, and it started just driving all these kinds of stories. Plus, it drove a lot of unique research from the academic community. Economists, public health people, computer scientists, all kinds of people, sociologists, everyone who had something to do with COVID and staying at home, which basically was an entire academic community in 2020. Started citing our data set analyzing it and reading recruiting original research papers on that data. So that kind of popularity led us to a lot of news stories and of course, big firms like Fox News, CNN, NPR, Washington Post New York Times, reached out to us for comment, and see I was wearing a whole bunch of random rolls. That safe graph including a bit of product targeting and spokesperson because that's kind of like a public face of this data set. So I would help out with analyses a little bit, but more so in communicating the value of this analysis to PR and spokespeople sorry, I suppose the PR people and then reporters was kind of my job. Of like translating it, because I have this skill and I kind of naturally have combined with like years of practice of about being both technical and then also being able to communicate with non technical audiences. So that kind of made me the person to do it. So all of a sudden, I go from discipline to random, you know, I was 24 at the time, and this 30 person startup that 99% of the country never heard, but within like a month or two, you know, I was being on like the nightly news and podcasts and we were mentioned in like New York Times, like eight or nine times Washington Post. So it was really, really fun. It was great for my ego. Cuz my parents when I left my job at Facebook, they were like, What are you doing joining this random startup that we've never heard of? Like, are you sure you want to do that? Like, Facebook is the talk of the talent like That's awesome. But once they saw me on TV, they're like, oh, okay, so this is like a real company like okay, I kind of get what you do now. You know, they didn't it didn't click in their head that this was a good career move until almost one and a half years later when I'm on TV, talking about this company. And all our family friends are watching the same news and they're like, oh, okay, is kind of bored.

Common pitfalls and tips for communicating insights to people who are less analytical and less data-oriented

Gilbert Eijkelenboom:

And then they realize, hey, maybe it's not that bad movie Oscar all because if you get his publicity and, and, and these are kind of opportunities that I can imagine. It makes him problems. And I'm wondering because you're, you've been, you know, talking to a lot about you know, how to communicate those insights also to people who are less analytical, less data oriented. Once Can you share some of your strategies or how you do that how, what are common pitfalls and tips?

Nick Singh:

Yes, I pretend them and the trick is, I actually am kind of dumb, but sometimes people think I'm smart. So the first thing is usually in a lot of the rooms I'm in, I'm not the smartest person. And I always crave that, right? So throughout my career, even when I'm very technical person or working at Facebook or Google, I'm still very technical, but I want to surround myself with people who can teach me something, which means I usually don't, which means I'm usually asking questions or asking for things to be simplified. And then once I have this mental model of how something works, I try to understand like, oh, how can I simplify even more, right? So this thing that I just do in my head of being dumb, asking dumb questions and being confident enough to ask other people to simplify their thoughts, and then we working with them, and then it's sort of my own thing, if that makes sense. At that point, if they've simplified it down to me, and then someone asked me about it. Secondly, I'm able to regurgitate that but it's almost like attributed back to me, right? Because, unfortunately, this is how it works. A lot of science and a lot of just real world stuff. The person who gets famous is not the person who maybe invented or discovered a thing. It's the person who kind of popularized it. So there's so many quotes, wrongly. attributed to Einstein and there are so many random discoveries attributed to Ben Franklin that he didn't actually do, which is that we know about it because of him. Basically, that's just what kept happening. I was able to simplify things I said, it was just me more being dealt with the other technical person and then figuring out what simplified version look like. So it's just really about dumbing it down. The second thing is attention span. I don't have a long attention span, and a lot of technical people like to make fun. So as we get older, go over how old are you? Ben? You're 33 I am 27. And I'm just starting to feel older. Older meaning like, oh, there's a different generation and they do things differently than me. Like this whole tick tock thing was like beyond my time, right. So I mean, I worked at Facebook. I grew up with social media. I'm familiar with Instagram, YouTube, no problem. I grew up with all that. But the Tick Tock thing is the first day where I'm like, solidly out of college solidly new mouth things, all the busy my career, and there's a new social platform and you wouldn't be right. For a lot of people. As we get older. It's very easy to dismiss this kind of social media trend, right? And then tactical especially, it's very easy for us to be like, Oh, I only do long form technical guides. I make my own projects. I'm not going to watch a six second dancing video about Python. Like I don't have time for that. That's all garbage like I'm a man of serious intellectual curiosity. I read Wikipedia, you know, but I don't. I kind of recognize that that's not where the world is headed. We're not heading towards more intellectual discussions on Wikipedia. We're headed towards short form six second videos, and I kind of embrace that I've been able to embrace that just because I've been limited in social media for so long. I've been writing on LinkedIn for about four years. Definitely encourage your readers to give me a fall at like 130,000 followers on there. Before that, I would post funny things on Facebook just for my friends and like it was a game for me to kind of get likes, you know, and I, I love comedy. So I like being funny, and I'm posting my own jokes and memes. And now looking back, it's kind of cringy but in high school, that was my first time like, writing stuff out in the public. And basically the issues embracing just me embracing less attention spans, which means you have to explain things simply. And you have to explain things in a not long winded way, has given me the gift of being able to communicate with non technical audiences. Because the fundamental truth is, as soon as you talk technical jargon, people's eyes glaze over. They don't want to hear from you that no one cares about you. No one cares about you. No one cares about what you have to say. They care about what you can do for them. Which means as soon as I talk 100 details that I care about. But you're not technical. They don't mean anything to you don't care about them. I quickly realized no one's listening to me no one cares. So as soon as I started talking in the way that people want to be talked to, which is short form, because our attention span that's been ruined by tick tock, right. That's a simple attention span we take to our daily encounters. Right. So first is starting with long videos that are shorter and like short Instagram stories, and reels. And then we have tech talks, right? Just attention spans are coming down. And to realize that and know that other people don't care about me. They don't give a shit about what I have to say. They just want to know how I can help them and if I can just deliver that quickly. That's the ultimate way to communicate with non technical audiences. And I recognize the irony of what a long winded answer that was.

Gilbert Eijkelenboom:

I appreciate that answer because it it taught me a few things because you're talking about pretending to be to be done or dumbing it down. Right? And to me, that's such a powerful mental model because we're all curves with the with knowledge, right? The curse of knowledge. You might you might know this concept where it's very hard to empathize with people who don't have certain knowledge or experiences, like stakeholders, business stakeholders don't have the same data and technical experience that data professional professionals have. And to overcome that. dumbing it down is powerful. And and there's something interesting also in the answer, for your second point about the attention span, because I think even the older generation, so if you talk to an executive, right, that person is older and didn't grow up with tick tock, but still the attention span got shorter, and in general that they spend on those two more senior stakeholders is fairly short as well, where they end up people like to present many details and, and all the technical stuff. So also even for those people a bit more addictive mindset might actually kind of fit in those presentations.

Nick Singh:

Right? Like the way technology is permeated through all age groups. If you live with older people, they will complain oh, these kids don't read books anymore. They're always just reading these you know, their minds addicted to these things. But if you ever see an older person how they like juggle their email, and they're getting a WhatsApp message or getting a Slack message and thinking about their kids or thinking about their wives and thinking about their business isn't about the gym. You know, in this day, in a day and age, it's like distraction. People don't want to admit it, but everyone's attention span has gone down. Because there's just so many more cool things to do on your phone, your laptop. And there's so many people to talk to, and there's so many messages to write back. So it just works through all age groups and all attention spans.

Writing a book

Gilbert Eijkelenboom:

And if we're talking about attention span, you wrote a book a very successful one, but I know it takes a lot of time to write a book. So how did you manage that? Because your attention span might be even shorter bedroom people 20 years ago, and it takes a long time and attention span to read a book right? So how take us through that journey? How did you have the the mental clarity and patience to do so?

Nick Singh:

Right so I'm a big fan of atomic habits which is all about rising too. You don't rise to like your aspirations, you have to kind of fall to the level of your systems, right, which is this idea that like look, I want to be a good writer and I want to be dedicated and all that but I'm not actually that good with willpower. I eat tons of candy and junk food like that's just how I am so how can I write a book? It's because when I quit my job, and I knew I had to do it. Okay, it's this find out. So let me take it back. Let me tell you a story about losing for is a famous rapper for the audience who might not know distribusi little was Eva gives us a really good story about how he was working like a random retail job and wanted to be a rapper. So one day he got a face tattoo because he knew that once he got a face tattoo, hidden we can't work a normal retail job couldn't be a cashier like he had to make it in Iraq because there was no other choice. Yes, this is a little bit of a self destructive thing to do. For doing really hard things. Honestly, I don't have a better way like there are other people man they eat right to go to the gym to sleep eight hours a day. I'm not that kind of person that you want to interest me. I stay up late playing video games and writing goals or doing whatever. Like I'm not that discipline. So when I quit my job during the COVID time, I knew like hey, everyone always has an idea and filter. You also have a great book too. So kudos to you and I don't think you quit your job to write the book, right?

Gilbert Eijkelenboom:

No, yeah. So that was partly on the sabbatical. And I could imagine it's so much harder if you are in a job and it's writing a book, but I was the last part I did while still having a part time job. Yeah,

Nick Singh:

gotcha. Okay, good. So you had some flexibility. to I need to meet people who have ideas for startups, ideas for books, ideas to be a Social Media Creator and they never get anywhere. I'm also a little bit that way because if I'm tied to a job, it makes sense. But when I quit my job, that face tattoo element was there like look, I don't have a job to go back to I have to make this work. So I can't just be s around writing the book. I have to get it done. The second reason why I do I had to do it was because just genuinely in 2020 when COVID hit, I saw so many my friends peers, the community lose their jobs, because all these companies are doing layoffs because they're not sure pay for the travel hospitality space. If you're expedient in your travel last year, a pen break. You know, is there going to be you know, a future in your business right and just the whole world slowed down as things shut down. There are layoffs and there's this really famous book called Cracking the Coding Interview that helped me get jobs at Facebook and Google and Microsoft and I was like, Okay, why does it look like this? Is this good for data? So around that same time, I'm like, Look, I know this, hopefully it sticks to my LinkedIn posts. Were talking about this stuff are already doing really well. So I have some sense that this idea is popular and it's a little bit de risk. Three. I've just quit my job. I have no other thing to do, like I have to make this work. And then four isn't the right time to do it because he does a really big thing and 2020 and I mean even now and secondly, people are really experiencing crazy turmoil in their careers. And then college people, a lot of people, a lot of people transition from a retail job to something more technical. It was just what was needed at the moment. So all those things kind of, they need to have this sense of inevitability. I need to get this done. It doesn't matter if I'm disciplined or not. And you just sit there and granted. It was really bad. It sucked. It sucked but it got done.

The advantage of social media

Gilbert Eijkelenboom:

Yeah. Oh fantastic. And I think the the piece that was really interesting and fascinating made a metaphor you know to go all in and there's no way back and burn your bridge basically. And also what you pointed out about your LinkedIn posts, I think it's such a underestimated value of social media and LinkedIn in specific that not of course it's a general to to promote your work and get ideas out but also to get the feedback. You know what resonates with people and it's so, so interesting. And there's also a lesson for data professionals data scientists to to see what people value right to you and always need to do kind of market research, even if you're just working as a data scientist to see what problems are hot. And what problems you can solve with data, not just when you're writing a book. Yeah, so it's fascinating to hear that story.

Nick Singh:

Exactly. A lot of people think about social media as a way to help themselves. So if I post I'll get my ex or I'll get more consulting engagements are all good fits for that. Which is fine. And it's totally a real thing. The way I like to think about it is first like think about like, how can it help you learn about a customer's pain point better, right? And we talked about this all the time in the Lean Startup approach of like developing an MVP, listening to customers iterating fast. We talked about this for product development or technology all the time, but we don't apply it to our own like skills or writing. Right. So let's say let me give you a concrete example. Let's say Gilbert, you're trying to really up your data visualization. You could read a lot of books about if you could make some projects you could do some projects at work about it for over two months. Or you can make really quick visuals. From a data set. Take 10 different data sets on Kaggle spend two hours on each dataset. Make of this make these 10 Interesting infographics and post them on social media and see which gets the most likes see which gets the most like oh wow this is cool or oh wow, I love how you did this right? What you just build your portfolio, but to this process of seeing what hits with people what doesn't hit with people helps you understand how to better communicate data driven insights in a visual manner. And suddenly the length of time when you're doing it for your work or doing it for a client. Then you know you're going to do amazing because you got 10 repetitions from a big audience of people who will not lie to you, right because if you don't like something you don't have like, right, like people just keep scrolling. So no one will buy a PSU when you post it on social media, so I just love it for getting rapid feedback. And that's kind of how the book started as well. I posted a whole bunch of stuff. I saw what things resonated. What didn't and the long form parts that did resonate the most basically became the course of the chapters and the rest is history.

What he learned and tips he can give about written communication

Gilbert Eijkelenboom:

Fantastic. And and a book is obviously written communication. So part of our communication and you know working when you work in data or software engineering is also written what have you learned about written communication or what tips can you give for the audience?

Nick Singh:

For written communication, again, people i i like to think like psychology dress so much of this like, Look, I'm not a psychologist or have any background in that, but just from knowing people, people do all the things to help themselves. And that works in the short term isn't helping the long term. So what I mean by that, when people write books they often write to make themselves look good. Or hey, I write in a complex way people will think I'm smart, or if I you know, explained this in a long winded way, and it's really long, people will think I'm really technical, and they'll poke me and they'll think I'm a really great news, right? But I always flip it universally and say okay, what does the reader want? The reader wants to learn this and the easiest, fastest quickest way and if some technical details are missing, if they're wrong, well, I wouldn't say it's wrong. But if there have been like, simplified, so maybe they're not correct in all edge cases, but they're a simplified medical model, and usually it's simpler enology that might not be 100%. Correct. It's a 99% correct way of explaining something in a much simpler way. I'll always take no shortcuts, and trim things down, shorten them and put them out there. That's my biggest thing for communication, which is almost the same insight I had for oral communication too, which is attention chants are short, right? Simply talk normally don't talk to make yourself look cool or look smart. Because in the short, in the short, long term, you'll be like, Wow, that guy is so smart. I have no idea what he's saying. Breaking the short term. That's what we think, Oh, wow. He's so smart. Look at all these complex that is sustainable. In the long term. That person never gained an audience or never gains authority because no one has a clue of what you actually mean.

Gilbert Eijkelenboom:

It makes so much sense and it all comes down to helping other people right to empathizing with your audience what they want, and as you mentioned, not to write for yourself to make yourself look better, because in fact, you don't make yourself look better by writing this much right just by writing three sentences, which are concise and to the point and the narrative. There are a lot of parallels because you know, how I see your work is that you help people get jobs, right, very smart people who have done great projects, but some often the hiring manager to the recruiters don't know about them right or they don't know how to ace the interview. And also my work. The reason I do my work is so a bit later in the stage when you do have a job. When you communicate the insights. They often have amazing insights to share but some but often the audience the business doesn't appreciate them or understand them because it's all wrapped up in such a mess of detail that technical people find interesting but the business stakeholders find not so I see a lot of similarities and parallels where we need to think more about other people rather than ourselves.

Nick Singh:

Exactly. That's what we always have to do is this is like an active service. data scientist is all about helping your stakeholders. Social media is all about helping your followers cope meter book writing is all about helping your readers so it's always like an act of service towards those people and you never lose sight of who your end, customer client and reader follow up whatever they are. You can't lose sight of that.

Thoughts on cold emailing

Gilbert Eijkelenboom:

Yes. So let's talk about a bit about helping people landing that dream job right, which you have achieved many times in the past whether people are data scientists or people working somewhere in here. Let's talk about two things. First of all the application right how do you get the interview in the first place? You talked a lot about cold emailing. So share, share a bit about your thoughts on cold emailing?

Nick Singh:

Yes, I'm a big fan of reaching out to people who don't know, right? So instead of getting a warm introduction, or someone introducing you to someone that's called warm, I'm saying going cold, right to somebody you do not know. I'm a big fan of this because if you write the email in a good convincing way and keep it short, you say why you're responding like you, you know, you don't play coy you don't play you don't be bashful. You really come out and say hey, look, here's what I'm gonna do for you in your business. Here's why I think you should talk to me. Are you free next week to talk is my calendar four or five, six sentences can start opening doors if you're writing that email in a good enough way, and you're writing to the right people. And most importantly, you are someone who's respond worthy. So I want to emphasize that last place, so I don't sound like a snake oil salesman. Obviously there's some base level of like skill and portfolio projects and resume work experience. You need to have to make the strategy work, right. So if you've got an eight years of data analyst experience, and you're applying to a machine learning job and you cannot speak to anything about machine learning, it's not going to randomly get you a job right or if you don't have any science experience and suddenly read a cold email. It's not gonna look just like magic and get moved to data science. But if you don't have data science experience, you've built some portfolio projects. You've taken some classes online, you have a GitHub project or two. You don't have a formal degree and you used to work as a nurse. And you're writing email to a health care company, or health tech company. And you say, Look, I know your recruiter or your resume screener. I mean, you don't want to say this, but we both know a resume screener recruiter is not going to want to interview you because they're just looking at Oh, bachelors in nursing Okay, moving on next, right there they you know that this is how it is already say this is the CO requisite remind right. But where this works is if you're right the cold email and write to the hiring manager directly look, Gilbert I see you're a hiring manager in healthcare. I love what your company does and how they use data science to help nursing professionals make less mistakes. I mean, nurse, turn data scientist. Here's my portfolio project grant analyze Kaggle data to do X and Y and also go to Tableau dashboard which you can find here. I'm also analyze the data and explore it with SQL queries, and you can find the GitHub repo here. I'd love to talk to you about this, John. And do you have time to get it done? I was about six six sentences, eight sentences, probably like 120 words. You can read this in a minute or two. And you write to the right person and just appeal to them. Like, hey, look, I'm a nurse transitioning into tech. Suddenly, you might just be ahead of someone with a master's degree, right? Who didn't do this and just easily applied on LinkedIn, amongst 100 other people and didn't really showcase why they're good for a healthcare tech company, or what they can really build or do you know, it just claim Oh, yeah, this certification and this degree, you know, give me a look.

Gilbert Eijkelenboom:

All right. I liked it because it contains three things it contains. It shows you know how to send a message that is concise, to also that you understand the position of the other person, you've kind of done your research and you you understand what they are in what industry they're in, and you're connected through your backgrounds, right being a nurse and how you can solve their problems. So I really liked that, that method.

Nick Singh:

And it was fourth thing that I like to emphasize in this is links, right? One is, oh, I studied this, I did that. Emphasize, I emphasize here like hey, here's a link to my Tableau dashboard, and you just hyperlink. And by the way, I have my SQL query. In my GitHub repo hyperlink. If someone doesn't even click the hyperlinks still, they're like, wow, this person didn't just BS their way here. It's not that they've had some random surf or pores online. And they claim they know SQL. It's like, oh, yeah, we've actually done something and even if I don't ever click on it, I'm like, wow, they really saved they could know how many nurses don't get home. How many nurses put a dashboard on Tableau, sorry, on Tableau Public, right. Very few. So I think that's a big emphasis. And of course, this isn't, like downside case of no one even clicks the link it's still helpful and of course, most people will click that dashboard and say, Oh, it's an interactive fun dashboard that looks at patient mortality, you know.

Gilbert Eijkelenboom:

Exactly. So index you also need to kind of sell right why it's worth watching and were spending at least a minute on and and I think many people working in technology, they kind of hate the word sales, because it's it sounds sleazy, it sounds like it's not academic. But I think data professionals and engineers can go to help from salespeople actually, by by understanding their audience and how to communicate in a persuasive

Nick Singh:

way. Exactly all these preconceived notions on like, Oh, why are you dumbing it down? Oh, sales is bad. All all these social media stupid, right? These are all things that like, we techies kind of permeate our brain. And I know this because look, I've been in engineering school. I've been around really smart people who love math, love coding, who think oh, all these fads are stupid are men of science men and women of science, you know, like, what are we doing on this stuff? But like, the reality of it always is is like, hey, sales and marketing people know something about human nature to social media is successful because it taps into something we all know about human nature. Short Form, content is successful because it taps into something about human nature. So it's like always leaning into the human nature to get better results. Instead of going back to the grain and being like, oh, like why should I sell sell myself? Why should I reach out to someone you know, like, that feels weird. That feels too aggressive. Like, you know, they don't know me, doesn't have to reach out to them because you know what recruiters reach out to you and they don't know you. Companies reach out to you. They don't know you. So why can you reach out to the hiring manager recruiter yourself?

Why are people reluctant to reach out?

Gilbert Eijkelenboom:

Exactly, exactly. Why do you think people are reluctant to do so or why do you think people don't think about this?

Nick Singh:

A lot of it just comes from mindset and security. If I sent an email, you don't respond, I'm going to feel bad. I haven't anything, I'm bad. But if I apply on LinkedIn, and you don't respond, I don't feel bad because well 200 Other people responded. Most people on LinkedIn just don't respond when you apply to a job. So whatever, right? If I write to you and I spent five minutes on it, and you don't respond, I'll feel sad. So why even take the risks? Right? That seems scary. That seems bad. That seems not fun. Who wants to be rejected by tend to lean on nine out of 10 Holding short one in 10 cold emails works and gets you a interview with a company that you normally wouldn't have been able to interview with. But nine out of 10 still need to feel it but no one responds or they like a mostly they don't respond. No one's asked me to say like, oh, not interested. Move on. You know, if you want to be a one on one. So I think it's like security, like, you know, people feel insecure in their stuff. So they'd rather attack like, oh, this doesn't work or it's not gonna work for me. Yeah, and that's another big part where it's like, hey, it's gonna work for someone else but can't work for me because I'm unique. That's the other baby mindset. I always had to help people. It's like, you're not that special. You're not that unique. Because I have been asked so many questions in the last two and a half years about breaking into data science, data analytics and getting jobs and everyone thinks they have a unique thing. I talked to 50 year olds who are breaking into their first data job. They're like, look, I'm unique because I'm older and there's bias. I talk to PhDs. They're like, oh, there's bias. Because I'm seen as too academic. I talk to community college folks, because it's like, oh, there's bias. I didn't graduate from a four year degree. Every person Oh, I have bias because I'm coming from healthcare. Every person is looking for an excuse for why it might not work for them. And one of the easiest ways to you know excuse your own lack of doing things it's like oh, well it doesn't really work or it doesn't probably going to work for me. Why should I even try it? Well, let me say myself so hardy. Once you get over this mental hurdle hurdles, like anything can happen. You know, you can write to Gilbert and be like, hey, Gilbert, I'm gonna come on your podcast, like your book. And you'll respond, right? And like two years ago, I might have been scared to do that. But now I'm just like, Alright, whatever.

Gilbert Eijkelenboom:

Yes. And I think if you adopt this mindset, we see that it has nothing to do with us right writing the message. If if we dare to do so. And if we write a good message, and people don't respond, maybe it's just this person was busy and we don't need to feel bad, right? It's awesome. That's nothing to do with else. We don't need to feel bad about not getting a response.

Nick Singh:

Exactly. Yeah. And you want to feel bad or weird because you know that hey, some sometimes some will work out because I'm someone a quality of going for the projects and real skills. So if this company doesn't like me, or they don't really respond or work out somewhere else, that's exactly it. And that security though you cultivate a lot of this though, I want to again, emphasize the skill here is writing a good email, but it comes after you build a body of work, right? So I don't want to say like, oh, okay, so today I don't know data science. Instead of learning data science and building in a science project. We'll go write emails, right. This first starts with having some good thought or good experience good projects and contributions. Good skills that you build up. After that, that gives you the security humility to face failure and just keep going. Because you're not scared because you know, your work stands for itself.

Sharing his tips and advice about interviews

Gilbert Eijkelenboom:

100 percent, and earlier you mentioned the understanding human nature better and we'd like to tell the listeners do books if they're interested. So one is to Sell Is Human by Daniel Pink, that shows that you know, everyone is selling whether we know it or not, especially if you have kids. You're definitely selling, selling attention and selling what they need to do and what they need to eat. And then the second book is by Robert Greene laws of human nature. It's a bit of a deeper book, but if you're a bit into psychology, that you will definitely appreciate it and learn a lot about the laws of human nature. So we spent some time some time about, you know, how to get the interview. So now let's take the next step and to talk about you know, how to how to ace the interview. You wrote a book about this, so you probably have a lot to share about that. So what tips and advices can you share about interviews and how to do that?

Nick Singh:

There are too many. That's why I had to wade through every one page book, but I will maybe highlight one or two that I think are kind of interesting because again, there's just too many and DNA interviews cover SQL, machine learning, coding, statistics, probability. Usually there might be an open ended kind of case study businessy case study. So it's too much to cover exactly in this one podcast. But two things I would like to give tips that are many tips is one. Don't let the fact that data interviews cover so many things. Scary for preparing. Some people often feel like well, if they ask anything under the sun, how can I even study for it? How can I prepare? You know what I mean? Like they get so intimidated. It's like, well, if they could ask you anything. You know, let's just wing it. What can I do? I want to shift people away from that mindset, because you can absolutely prepare for these interviews because questions do come in certain patterns. And there are ways to practice those patterns. Let me give you just one quick pattern. Of course the book is booklet 201 questions, which map to almost like 40 different patterns that you need to know but let me just highlight the one one that actually you might find interesting, and that's actually been the topic of this podcast is this concept of my way of testing your statistics, knowledge and a lot of hiring managers ways of doing this is asking you to explain a concept like a confidence interval for a p value, like your five or like your middle schooler, or like you're not technical, some way of basically seeing Can you dumb down a technical concept into simpler words. This is a really popular thing because it tests whether you know the statistical concept at hand. And this actually works for machine learning concepts as well. You know, people ask you like, Hey, can you explain bias variance tradeoff in a more simple way? Or can you explain me how linear regression works, but pretend I will mean you know, high school level algebra? You know, I don't know calculus and all that but just explain it to me in terms of y equals mx plus b, you know, just high school level algebra one or two. That's a really common pattern. And a good way to practice these is to verbalize them yourself because there's a big big gap between like, oh yeah, p values. I know what those are. And then actually like, be like oh, wait, how would you describe the p value just in general? Because if there are nuances to these kind of concepts, right, like, how do you interpret a confidence interval or what does it mean to reject the null hypothesis? What does that actually mean? There is actually one says that people will explain it wrong. So that's one thing is to record yourself saying these things out loud, not just in your head. Second is I love Reddit. There's a subreddit called are explained like five, which is all about dumbing down technical concepts. So if you search up like Reddit, explain p values like I'm five you'll find some good results or explain Stochastic gradient descent, like five, you'll find some good analogies. You read it off in these articles, and you'll start to work on your own technical communication ability. Once you see how other masks people masterfully explained it themselves. And guess what read it has an upvote system. So you already know the good answers are on the top. Right. And that's how you can learn from the best and that's how you can yourself improve your technical communication skills, which is a win for the interviewer but it's also in like, right in the day to day job.

Gilbert Eijkelenboom:

Right. That's such a good idea. I run it down. I don't know this Reddit threads. Expanding games out is like on fire, but I think it's very, very useful. I'm going to look it up.

Nick Singh:

Yes, it's called a code. e li five, which stands for explain like I'm five by just one well, rather than looking on the subreddit and just say like Reddit, explain p values simply or explain like I'm five and you will find a lot of these threads and of course the whole point is the you don't have to be five to understand these things. A lot of them say hey, just pretend I only know high school level or middle school level math, right? No calculus, no linear algebra. How would you explain X and Y is the concept.

How does Nick recommend people use the star method in interviews

Gilbert Eijkelenboom:

Right. Good. Thanks for the tip. Something else I wanted to ask is because many people know about the STAR method right in interviews is so this is something you can recommend and how, yeah, how. How do you recommend people use that and what are some mistakes you see that people make?

Nick Singh:

Yes. I love star star for me, everyone. Interviews is this formula that you can use where you explain the situation task action result for every story you tell? I think it's useful. It's a useful framework. Why it's useful is because I've done so do mock interviews with really smart technical people who still make mistakes on something as simple as lol there's this four step formula. What happens is us techies we like to dive into the actions we talk to a part of star often will forget to set up the situation and task because in our head, we were steeped in this technical problem for months. So obviously, we know the situation has scar, but the interviewer they have no idea what you're talking about. They don't have that technical concept, even if they're very smart. They might work in a different business line or different industry than you did. So it's very crucial to explain the s&p part of the situation cause and finally, the result. It's another thing that techies often forget because we're so busy talking about the actions. We forget about the result, especially since a lot of times in the story. The result comes in at the end, and it's very easy to mess up your landing you know, or be like, oh, yeah, you know, I spent 80% of my time telling you about the situation, the results and the task at hand and the actions I took. I forgot to explain the result or like what was the so what? So what I like to do is tell people like Hey, start with the so what like repeat the result? So be like Oh, tell me about a time. I'll give you an example of an Amazon behavioral interview question. Tell me about a time you went against the status quo. Okay. They're looking for a little bit of contrarian thinking or they're just trying to understand like, when did you like speak up or do something different? If I was to answer that, I would say the result, hey, I thought that Facebook stories was bad for new users. And it did an analysis and I found that it actually was which actually changed the q4 roadmap for the Facebook stories team. Let me tell you a little bit more about the situation. See, I started with the results, like whoa, change the roadmap for this big product board. And he found something Okay, that's cool. And I think the situation Oh, we had a hunch that new users. They didn't really like Facebook stories, and that it might impact their new user experience because it was a complicated product. And new users on Facebook don't have any friends. And Facebook stories as a product has less inventory because it's way less popular than other modalities of posting. So it was just a unit that was at the front of their page that was just almost always empty. And it was always confusing. It's not what typical Facebook does, because the story feature was new at the time, in 2018. And it was almost like a clone of Instagram stories and Snapchat stories and it was a much poorly done clone that no one wanted to use really. And I had a hunch like Hey new users because my my team was all about protecting New Users and improving their retention rates. So I look at everything as a lens of like, if I were the user, what would I do? So I dove into the data paths was figured out is Facebook stories good or not? So I dove into the data, or a whole bunch of SQL queries. I found real bugs in the product. And I figured out that oh, yeah, actually there is a material impact on retention rates for new users. So that's how the result was. I was able to actually change the whole roadmap and kind of bring the Facebook story and leadership to big investment area. I was able to tell them like look, we got to pay attention. Users, because clearly they're behaving differently than regular users. And this is impacting both user retention rates. And it's not helping anyone the painful stories came because these new users are using stories. And that's the result. Of course, I'm a little rusty because the story I'm telling you is from 2018 which is now quite a few years ago. And yes, I rambled a little bit. But hopefully we started with a result made sense hopefully explained the situation tasks correctly. The actions were kind of brief, because I know that later on if you really care about my actions, you're gonna ask me, okay, hey, Nick, to tell me what SQL queries you write or what datasets you use, or what was the specific gap you found. That can only be asked later, right? And we'll dive into that and that's where I get to show off how good I am at SQL and understanding product gaps. But right from the get go. I'm just trying to explain you the story I keep it high level do my best to increase the star

How to build up mental index

Gilbert Eijkelenboom:

right and I like that because it makes it very concise and and if they ask about it, you go to this mental appendix and you take the information from there and then give give that information to to them and it also by starting with the results, right and piques their interest. It kind of sells why why they should listen well to this to the to the rest of the answer and and it hooks them in. So I really like that approach.

Nick Singh:

We like to clown on Buzzfeed. Right BuzzFeed style articles six, six things you'll never believe three is gonna shock you right, you know, but there's a reason BuzzFeed and that kind of style copywriting is so popular, right? So I mean, I don't want to sensationalize my answer but it really is like, I found i i figured out a big gap in Facebook stories that no one knew about. Stay tuned. Right. It's basically that's what I'm telling people like, oh yeah, I single handedly found this huge gap in this multi billion dollar investment project that affects millions of new users at Facebook, starting hundreds of millions of new users a year and no one even knew about it. Let me tell you how. I start with their design, pique your interest, and I go into it, obviously, look, I'm not going to talk like that in their view. But you can see that when you practice this as an exercise, you realize, Wow, your stories turn out to be a lot more compelling. And there's a lot more understandable than if I was like oh yeah, Gilbert, I went against the grain by the time I wrote a window function in SQL and I did all this and then I forget to even explain you like, what are the SQL queries doing and what am I finding and why does it matter, right. Something I also want to highlight Cobra that you mentioned was mental index, right? So I have this quick index of St. AR, what are the key four points I need to hit? Especially what's the result I need to hit? How do I build up that entire index? It's because I practice this kind of question so many times and I have the same you know, six stories that I kind of say over and over again, to answer a wide variety of questions. I advise listeners to this podcast if they want to improve their own oral communication skills in the interview. Practice needs perfect, but specifically, don't leave things to your memory. Because once you started having to like remember like, oh, yeah, what was the result? Like what was the key number? What was what you should just do is make a quick little thing in Google Sheets, where you have three or four big projects or big wins or big accomplishments, and then just write the S T AR and just try to boil it down. To one sentence. maximally two but keep it to one sentence. This exercise you to clear up your thinking so much, because if you know your four biggest accomplishments, and you know the one sentence way to describe a situation task action result when you're in an interview, almost any behavioral interview question can be mapped back to your forming projects. And the kernel of those things is always those four sentences. And the hard work doesn't happen during the interview, where you're trying to figure out like, you know, how do I tell this guy Philbrook hasn't worked at Facebook, who might not even know what Facebook stories is? How do I explain in the so why? Not enough? It's certainly not doing the face to face interview. I've done that beforehand. It might be pushy, and it takes me a good 30 minutes an hour to distill it two months project into four sentences. But that's where you come off as a much better speaker is when you do that pre work beforehand. So then I just have to remember the four sentences and not the full story.

Gilbert Eijkelenboom:

Yeah, because they're gonna need to come up during the interview. It's almost impossible, right? Because you're probably a little bit nervous or very nervous. You're thinking about so many things, you're going to be in your head and the more you think the less you are with the person in front of you. You're connected. So that doesn't work.

Nick Singh:

That's exactly it. And then people people always, you know, I have some more technical people who don't specialize in this kind of communication things. They say, Oh, Nick, good morning sound over rehearsed. Over prepared gonna come off and eNGenius I pronounced that wrong, but not Gen basically is what I'm trying to say. And it's like, look under the stress of an interview. Unless you're like a really slick salesperson or professional spokesperson or whatever. It is so hard to seem confident and composed and not nervous and deliver a cogent story around something technical that the other person might not have context on. So even if you prepare a lot, you're still going to do some slight variations and mess up. So it's like no chance that you're gonna like sound over rehearsed or over prepared is only going to help you. That's like the one main thing and the second thing is humility. Like, it's like, Look, I'm not that smart. I'm not that articulate. So I'm going to spend hours beforehand so that during an interview I don't have to be on the fly, articulate and crack jokes are like figuring out how to distill two months of work into 30 seconds, right? It's because I'm not that smart. And so you do the pre work and I think people can adopt that mentality. A lot of like, hey, look, I'm not that smooth talker. Let me go through the pre work, and you'll be better off for it.

Where can people connect and follow

Gilbert Eijkelenboom:

100% on a percent. So yeah, thanks. Thanks a lot for the conversations today where where can people connect with you or follow you?

Nick Singh:

Yeah, definitely. Folks can connect with me on LinkedIn, I have 130,000 followers. There. Just look up nixing. You can also follow me on Twitter. And mixing tech is my handle. I have about 20,000 followers there. I'm getting new on Tik Tok. And we're going to shout that out. If you see me you see me more embarrassed by it because I am not very good at dancing. And finally of course, check out the book ace the data science interview on Amazon, as well as sign up for a free account on a lever and you can start practicing real SQL interview questions today for free.

Gilbert Eijkelenboom:

Fantastic. I really appreciate your your approach your focus on helping other people is very close to how I like to approach business and life and I will put all the links in the in the show notes so that people can reach you and benefit from your work. So yeah, thanks a lot for a conversation today. I really enjoyed I learned some new things, you know, explaining before he was like I'm five I'm gonna look it up because it's very beneficial to see you know, how you can communicate things in a simple and persuasive way to others that they understand and appreciate. So thanks a lot for being here today. Is there anything lost you would like to share with with your audience or a big takeaway from today?

Nick Singh:

No, thank you for having me open and hopefully, people enjoy my little takes on psychology and humans because honestly, that's that was more of our conversation today, which is fine. I don't normally get to talk about that. So appreciate you having me on.

Gilbert Eijkelenboom:

Fantastic. Thanks. A lot, Nick. And we'll speak soon

Getting and Acing Your Data Science Interview

Recent Posts