SDS 331: Hacking Data Science Interviews for Graduates

Kirill Eremenko: Welcome to the SuperDataScience Podcast. My name is Kirill Eremenko, Data Science Coach and Lifestyle Entrepreneur, and each week we bring you inspiring people and ideas to help you build your successful career in data science. Thanks for being here today and now let’s make the complex simple.

Kirill Eremenko: This episode is brought to you by DataScienceGO 2020, our very own data science conference. We’ve already done three events in the past three years and we’re moving into our fourth year in 2020. And to give you a feel for what to expect, here are some stats from DSGO 2019. We had 620 attendees fly in from 25 different countries, 38 speakers gave talks, 150 plus business decision makers attended the sessions as well and get this 2,400 cups of coffee were drank during the networking sessions. So DataScienceGO is not just a place where you will get all the top data science skills that you need for your career. That’s definitely a huge component of the conference, but also is a great place where the community comes together to network.

Kirill Eremenko: At DataScienceGO, you will meet data scientists and professionals from companies like Accenture, AIG, Wells Fargo, MasterCard, Facebook, Google, IBM, Microsoft, Salesforce, Teradata, Amazon, eBay, Shopify, and many, many more. So this is a great opportunity to meet and network with your colleagues, to meet and start catching up with your mentor, or maybe to even meet the manager at the next company that you’ll be working for. At DataScienceGO 2020 we’ve been almost doubling every single year so we’re expecting about a thousand attendees at this next event. DataScienceGO is happening on the weekend of the 6th, 7th and 8th of November, 2020 and you can already secure your tickets today at DataScienceGO.com.

Kirill Eremenko: One more thing is that we actually have different tracks. So we found that this is a very important component for attendees and we have tracks tailored to your experience. So if you’re a beginner, there’s a beginner track, which will help you get the skills to break into data science. If you’re an intermediate practitioner, there’s an intermediate track for you to progress to advanced. If you’re already advanced, there’s an exclusive advanced track just for you. So whatever your level, you can find the right track, the right talks, the right workshops, the right sessions, and case studies and panels at DataScienceGO. So on that note, this is the best conference for you to attend to skyrocket your data science career. So make sure to secure your ticket at DataScienceGO.com today and I can’t wait to meet you in person in California in November, 2020.

Kirill Eremenko: Welcome back to the SuperDataScience Podcast, ladies and gentlemen. Super pumped to have you back here on the show today. Today’s episode is exclusively specialized, specialized for data science interviews for grads. In fact, our guest today, Harshal Sanap himself was a graduate just recently and he now has a job in data science, so he’s fresh. He’s being fresh to that process of graduating from university, of understanding, of applying for jobs, understanding what he wants, applying for internships, graduate positions, and then later jobs and preparing for interviews, going to interviews, making mistakes, learning from them and so on. So he’s just graduated and he’s just gotten his job. He’s been in his job for the past three months, three or four months. So all of this is all fresh in his mind and he was very kind to join the podcast and share these experiences to help you avoid those mistakes.

Kirill Eremenko: So if you are a graduate or recent graduate from a university or you’re about to graduate, even better if you’re graduating in a year or so, this podcast is going to be extremely relevant for you to help you find a data science job, to find those interviews, to prepare for them. Specifically the things we’re going to talk about. We’re going to start off by talking about how Harshal and I had a great time at DataScienceGO this year and he was actually volunteering there. So you’ll learn a lot of what he got to take away from that experience.

Kirill Eremenko: Then we moved on to the discussion about jobs and the first thing Harshal pointed out was to understand what is it that you want from data science, and he gave his experience. That is very personal thing for everyone, but I think his experience will be valuable. So Harshal gave four tips on how to apply for jobs. Then he gave five tips on how to prepare for the interviews and I actually gave one extra one to add to that. Then finally he shared five mistakes that he himself has made at interviews to help you avoid those interviews. So there you go. That’s a total of about 15 tips or even more that you’ll get from this podcast to apply to your job. Plus he gave plenty of resources which will be useful for you in new your preparation for the actual content part of the interviews.

Kirill Eremenko: So there we go. That’s what this podcast is about. Speaking from my personal experience, even if out of those 15, three of those tips are going to be super relevant to you specifically, this is going to be a great investment of your time because it can make or break an interview and your job in data science. So let’s do this and without further ado I bring to you data analyst Harshal Sanap.

Kirill Eremenko: Welcome back to the SuperDataScience Podcast, ladies and gentlemen. Super pumped to have you back here on the show. Today’s super special guests is calling in from San Diego. Harshal, how are you going, my friend?

Harshal Sanap: I’m doing amazing. Today’s weather is super sunny and it’s really warm and pleasing outside.

Kirill Eremenko: Fantastic.

Harshal Sanap: Thank you for having me. How are you doing?

Kirill Eremenko: Oh, I’m very excited to have you. I’m doing very well too. In fact, I’m doing amazing as well. Pretty early here in Australia, but well, not as early. It’s almost 9:00 but earlier than for you. I love this when we’re in different days. For me, it’s Monday. For you, it’s still Sunday. How cool is that?

Harshal Sanap: Oh, I did not know that Australia is ahead of us. Okay. Interesting.

Kirill Eremenko: Very much so. Very much. Okay.

Harshal Sanap: Yeah. Because I think I never dealt with any of my client role in Australia, it’s mainly has been Asia or Europe or like, so.

Kirill Eremenko: But you go back to Mumbai, right? Mumbai is farther.

Harshal Sanap: Yeah. Yeah. Right. So they are ahead as well. I think right now it’s like 3 or 4 AM Sunday morning there.

Kirill Eremenko: Yeah. Yeah. Well for us, it’s already almost 9. I think apart from Australia, Australia is even further ahead than Japan, even though Japan is called the country of the rising sun. I know Brisbane is one hour ahead of Tokyo and, but then there’s New Zealand and oh, I’m curious where Hawaii is in that sense. Hawaii.

Harshal Sanap: I think Hawaii should be in the same time zone as us because it’s like it’s really close from here.

Kirill Eremenko: Oh, okay. So it’s kind of like it’s also Sunday in Hawaii then.

Harshal Sanap: Yeah, absolutely. Yeah.

Kirill Eremenko: There’s only a few places that are even more ahead than Australia like New Zealand is one of them.

Harshal Sanap: Cool. Cool fact I learned today.

Kirill Eremenko: Yeah. Well yeah. Live and learn. Always. Always learning.

Harshal Sanap: Yeah, absolutely.

Kirill Eremenko: Speaking of learning, wanted to say a huge thank you Harshal for coming and volunteering at DataScienceGO this year in San Diego. It was amazing. Thank you so much.

Harshal Sanap: Yeah, my pleasure. Thank you for having me.

Kirill Eremenko: Your help was incredible. You were one of the top volunteers. How many volunteers did we have? How many did we have in total?

Harshal Sanap: I would say this time we have a huge population because there were like few exchange students from Korea. So there were like a lot of people, but I think last time it was just me and then one more girl from here. So yeah. I think in volunteers, we kind of grew exponentially I would say.

Kirill Eremenko: Yeah, no. It was crazy. This time we had like an army of Korean exchange students running around.

Harshal Sanap: Yeah, exactly. Because they were also, they were part of like learning English programs, so they were all super nervous to be around all some of the native Americans, English speakers. So it was a good experience.

Kirill Eremenko: Yeah, man. So you volunteered twice. How many DSGOs have you been to total, two?

Harshal Sanap: How many what?

Kirill Eremenko: How many of the conferences?

Harshal Sanap: Yeah, just two.

Kirill Eremenko: Just two. Okay. How did you find out-

Harshal Sanap: Two of them in San Diego.

Kirill Eremenko: How did you find out about DataScienceGO?

Harshal Sanap: It’s actually very interesting story. I think when you guys were looking, were going to do DataScienceGO conference for the first time here in San Diego, few months before that I think you guys started with the meetups in San Diego. So I think Maria or AKA Mavi and she reached out to me on LinkedIn and she was looking for someone to host the meetups here in San Diego and I think she was looking for a speaker during summertime or like early summer. Unfortunately, I was a little busy and I told her like, “Oh,” I’m like, “Look, I’m busy looking for internships and all those things.” So I told her I won’t be able to help her out at this point, but maybe in future, we’ll see. I think then she reached out to Ashlyn and then Ashlyn started hosting the meetups.

Harshal Sanap: But I think somewhere down the line I saw a little bit of advertisement about DSGO on LinkedIn and then I reached out again to Mavi and I had a call with her and I asked her if there is an open spot where I could contribute and she said, “Yeah, join us,” because I was here in San Diego as well, so she just got me in and here I am.

Kirill Eremenko: Nice. Very nice. How’s your experience been with volunteering?

Harshal Sanap: I would say the experience has been really good because there are two aspects either I could have come as a participant, but I think as a volunteer I got more information about the conference and I also got to know all the speakers individually. So I would say it was a more learning. I mean, I got to understand how to conduct the whole conference and how to organize it, but I also learned a lot from the conference because I think everyone at the conference, they were super nice and they allowed me to sit at any of the sessions that I was interested in. So it was a great experience and personally I got to know most of the speakers and I have their contact numbers, so hey, it was a good thing for me.

Kirill Eremenko: Fantastic. Fantastic. That’s really cool. Which of the sessions was a favorite this DSGO 2019?

Harshal Sanap: This time I would say I did not get to sit for a lot of data science [inaudible 00:11:49] but I sat for like most of the morning sessions and I think one of my favorite, I forgot the name of the speaker, but it was a little motivational talk, which happened on Sunday morning and he really pumped the whole crowd up and that was really good. I think he was talking about how to get things done and how to go after what you want. It was I think one of my favorite talks.

Kirill Eremenko: Yeah, I’ll remember the name. I mean, I actually missed that talk, but I heard it was very good.

Harshal Sanap: Yeah, yeah. So I think I remember talking to you about that talk on that date [inaudible 00:12:24] I think you came in a little late, but it was really amazing talk. Yeah.

Kirill Eremenko: Yeah. I think it was Antonio Neves. That’s-

Harshal Sanap: Yeah, Antonio, right. That’s correct. Yeah. That was one of my favorite talk.

Kirill Eremenko: That’s awesome. That’s pretty cool. Because we on purpose invite every time somebody who’s completely not related to data science to get a different perspective. First time it was Kyle Cease who’s a comedian and he gave a talk. Yeah, he gave to talk on following your heart, following your dreams. People cried in the audience, seriously like because-

Harshal Sanap: Oh my God.

Kirill Eremenko: Yeah, and cried and laughed and it was interesting because it was so different. Data science, it’s very technical. You’re using your brain a lot, thinking a lot.

Harshal Sanap: Right, right.

Kirill Eremenko: He spoke to the heart and he really opened people up. Everybody was giving each other hugs at the end. That was really cool.

Harshal Sanap: Wow, that’s good. Yeah, I would say at the end we are all people. It’s just that we happen to like data science. So yeah. I mean, if you pick a nerve, they are going to get emotional, so yeah.

Kirill Eremenko: It’s good. I like that mix. Last year we invited Pablos Holman who is not directly related to data science. He’s an inventor and a hacker.

Harshal Sanap: Right.

Kirill Eremenko: Do you remember that talk?

Harshal Sanap: Oh yeah, absolutely.

Kirill Eremenko: How good was that?

Harshal Sanap: The man who worked with Bill Gates. Yeah, I know. Everyone, I remember like all of the volunteers and your whole team was inside and everyone was like, “Oh my God, we need to be at this talk.” Yeah, it was really good.

Kirill Eremenko: Yeah. It stands to, now you kind of understand, sometimes we hear about these TED speakers and so on. Pablo Holman is a TED speaker and he’s got 20 million views.

Harshal Sanap: Yeah, absolutely.

Kirill Eremenko: You watch a TED Talk online, you’re like, “Oh, that’s pretty good.” But when you see him live, you’re like, “Wow, okay. I get it. I understand why people love his talk so much.”

Harshal Sanap: Oh, yeah. Absolutely. Yeah. So I personally think that the combination of a technical and little bit of like nontechnical aspects makes it a really good blend for a conference.

Kirill Eremenko: Yeah. Thanks man. You said you’ve got to connect with speakers and so on, what did you think of the community at DataScienceGO? For me for example, it’s unlike other data science conferences that might have fantastic content, here, in addition to the content, there’s actually I don’t know, like a sense of you know everybody. You’re like at home. Did you get that feeling?

Harshal Sanap: Yeah, absolutely. Because I think one of the best part of this conference is I could literally just go and talk to anyone and not feel as if like I’m talking to a speaker or someone who’s superior to me. I always felt that I’m talking to a peer or a colleague and it was really good bonding experience and once you bond well then you can connect to it and communicate well. So I think that was the best part for me at this conference.

Kirill Eremenko: Fantastic. Well, thank you so much. That’s really cool, some cool points. If anybody listening wants to come volunteer, we’re setting up a URL, datasciencego.com/volunteer. So head on over there and fill out the form and it would be really cool to have you as a volunteer and you will get to meet Harshal. You’re coming back to volunteer this year, right?

Harshal Sanap: Oh, yeah. Absolutely. I think I’ll be a senior volunteer now.

Kirill Eremenko: Yeah, yeah. The boss, the volunteer boss.

Harshal Sanap: Absolutely.

Kirill Eremenko: That’d be cool. That’d be cool. Awesome. Okay, well, on other topics, I want to give you a huge congrats, man. You got your data science job that you were looking for. Congrats.

Harshal Sanap: Yeah, thank you.

Kirill Eremenko: How does it feel? You’ve been there you said since August. So that’s what, three, four months now?

Harshal Sanap: Yeah, three, four months. I just started around the end of August. So yeah, so I would say my job is not purely into data science. It’s more of data, into data analytics. So my title is business intelligence analyst, but people also refer to us as business analyst or data analyst or just an analytical person. So it has been a great journey so far because one of the most important thing for me in a job was to get used to playing with the data. Because before this I did not… Never dealt with data in a huge chunk of data. Even though I’d worked in business intelligence, I’d never did data analytics using SQL or Tableau or Alteryx. So I look at this company as a foundation for me and I’m just happy to play with the data every day I go in and they have plenty of data to play around with. Yeah.

Kirill Eremenko: That’s really cool. That’s awesome. A great first step. So what industry is this company in, if you don’t mind sharing?

Harshal Sanap: Yeah. So I work with SmartDrive Systems, so we are into transportation intelligence. So apparently I did not know before joining this company that transportation is apparently a billion dollar industry. It’s like all the trucks that are moving from point A to point B, their safety is very important and there are a lot of things at stake. So a lot of companies are investing into how we can improve the safety and how we can improve the mileage efficiency for all the trucks and different kinds of fleets. So we basically have cameras and sensors on all of those trucks or vehicles and we collect data from them in realtime.

Harshal Sanap: Then my typical job would be to let them know, “Oh, on a weekly basis, driver X, Y, Z performed like this. They have this many collisions, this many near collisions, this many observations we noticed and this is how you can improve.” So we try to do this on our different sites, site level, company levels, just to roll it up on different levels of hierarchy so that we can understand how safe and secure that particular site is or that driver is and how we can improve the safety.

Kirill Eremenko: Very interesting. Sounds like a very responsible job. How do you-

Harshal Sanap: Yeah.

Kirill Eremenko: How do you feel about that?

Harshal Sanap: So in the beginning I was a little nervous because it’s like I would say in any data industry, you are liable for any report that you make because you are analyzing all the data and you’re showcasing it to someone and those people are going to make decisions on top of that data. So I was a bit nervous, but I would say some things just come to you with experience. So I would say I had really good mentors and my manager is super good. He has been in the data analytics industry for more than 20 years. So it’s like a very, to be honest, he has seen Tableau 1.0.

Kirill Eremenko: Oh wow.

Harshal Sanap: Yeah. So he has seen all the versions. So it’s good to have their support and from them, I slowly have started developing the confidence and the responsibility of the data that I’m dealing with. 

Kirill Eremenko: Fantastic. That’s very cool. Very important to have good mentors.

Harshal Sanap: Yeah, absolutely.

Kirill Eremenko: Is it hard to get up to speed with the industry?

Harshal Sanap: I would say it was a little bit hard to just get on the speed of the team because everyone is super experienced and I was just starting off fresh. But I just took it as a challenge and I just… I think I’m doing pretty well.

Kirill Eremenko: Nice. That’s important. That’s important.

Harshal Sanap: Yeah, yeah.

Kirill Eremenko: How big is the team?

Harshal Sanap: So the whole analytics team is almost 15 people.

Kirill Eremenko: Oh wow.

Harshal Sanap: So there is like one internal team which focuses on the internal activities, but my team is of four people and a manager. We focus mainly on the external side as if like if the client has specific requirement or they want to get this kind of data, they want to see this report. So we have so many clients or client specific custom reporting is done by us. Then there is a separate team of data scientists and they are like all PhDs. So also they have their different focus, which is a little more mathematical and statistical.

Kirill Eremenko: Okay. Okay. Very cool. So tell us about this process of getting the job because you finish your masters not so long ago, right? About in June, about six months ago.

Harshal Sanap: Right. Yeah. So I graduated in May and I would say to get a job after graduation, so here, I would like to focus more on getting a job after the graduation part. So-

Kirill Eremenko: Where did you start? So you graduated, did you already… Were you already doing some internship work in the meantime or were you starting looking from scratch? Tell us the story.

Harshal Sanap: So, oh yeah, absolutely. So I had an internship going on, but apparently when the internship was getting over I did not have the assurance whether it’s going to get converted into full-time. So I had to start with my hunt. I would say to be honest, looking for a job when you are finishing up with your graduation, it’s again a full-time job. I don’t know from how many people you must have heard this but it takes a lot of time and effort and a lot of persistence. So I would say it took me almost more than around four or five months to actually find something that I was looking for. Yeah, yeah.

Kirill Eremenko: Okay. So you graduated in May.

Harshal Sanap: Right.

Kirill Eremenko: You started looking, right? So you-

Harshal Sanap: No. I actually did… That would be a very bad thing to do after… I actually started looking in January or February, like that. 

Kirill Eremenko: Oh, like six months before.

Harshal Sanap: Yeah. Yeah. You would be surprised, some people actually started a year in advance. I would consider that I started a little late.

Kirill Eremenko: Wow.

Harshal Sanap: Because the early you start and the early you get done with it, so the more relaxed you are and the more you could focus on your graduation on just celebrating. Yeah, yeah.

Kirill Eremenko: Okay. All right. So very interesting. How did you start? So January or February, you’re still studying. What steps did you take when you say you started looking for a data science job?

Harshal Sanap: I would say the first step that I wanted to make sure was what exactly that I’m looking for because currently in the industry there are so many terms that are just flowing around. Anything related to data, some people call it data science, some people call it data analytics, some people call it business intelligence, some people call it business analyst, some people just call it analytics consultant. So there are so many things. So I have to first understand what exactly it is that I want. If it is data science, do I want to get too much detail into mathematics and statistics or not. So that was my first layer of understanding what exactly do I want. After that, I started preparing. I tried to understand what exactly are the factors that are most important for the interviews, what are the things that people are asking in the interviews?

Harshal Sanap: Because I realized that in the field of analytics, there are some common things like SQL, Python, Tableau, your data understanding. So I tried to understand, I just noted down all those things and then I tried to focus on one thing, like let’s say if SQL was the first part. So I would just like give one week just to do whole week of SQL, try to understand everything from basic select clause on all those basics until the advanced SQL and then practice. So this is how my journey was. Then once I taught that I’m ready, then I started looking for jobs or applying for jobs.

Kirill Eremenko: Okay, got you. So step one is understanding what do I want.

Harshal Sanap: Exactly.

Kirill Eremenko: Step two, what are people asking in interviews?

Harshal Sanap: Right.

Kirill Eremenko: Specifically the tools.

Harshal Sanap: Exactly.

Kirill Eremenko: Step three, applying for jobs. Okay, so let’s go through step one. Data science as you mentioned is so broad. Not only there are different terms for the professionals, from data analysts to business intelligence analysts to data scientists to analytics people, like you said.

Harshal Sanap: Right.

Kirill Eremenko: There’s actually also a huge scope of different types of work. You can be predominantly focused on data preparation, on modeling or data presentation or visualization, on storytelling, interpretations, building certain tools and stuff like that. That’s the scope of tasks within data science, and also on top of that, there’s also a huge variety of industries that you can go into from healthcare to transportation, from banking to finance and the science and whatever else you want. Pretty much these days, any industry can benefit from a data science, whether that be in marketing and operations and finance and so on. So the question is how did you go about all those things? There’s so many variables. How did you understand what is it exactly do you want to do?

Harshal Sanap: That’s a good question. I would say it’s a lengthy process. For me, when I actually came for my masters, all I knew that… Because before coming for a masters, I had worked with SAP in the business intelligence sector. So during that time, I knew the importance of reporting and playing with the data and I always knew that I want to learn more in that field. So when I came from a masters, my main focus was how do I create insights from the data? That was my main driving factor. I did not care that much about what industry it is, as long as there is data on one side and I’m creating value on the other side, it did not matter what tools are coming in the picture or what industry it was.

Harshal Sanap: So all those things, industry and the tools were secondary. My biggest focus was there is data on one side and there is something that I want to find some insight from that data which will be beneficial for that business. So that was my main focus. Then I went about learning all the tools and techniques that were required to be the middle man to translate the message from data into business insights. 

Kirill Eremenko: Okay. Interesting. In your story I heard a little bit kind of like that you were leaning, and correct me if I’m wrong, but it sounds like you were leaning more to the visualization side of things. Let’s say if in a crude way, we break data science into two areas. One would be modeling and doing statistical modeling, using packages in R and Python to create logistic regression, bayesian, inferences and so on, versus visualization using Tableau, Qlik Sense, Power BI, Alteryx, all these other tools to interact with data and actually help people see the data. Would you say you already knew you were leaning towards one or the other back when you finished university?

Harshal Sanap: No.

Kirill Eremenko: No.

Harshal Sanap: Honestly not, no. Because in my mind, my definition of data scientist was doing reporting and visualization and all those things which I really liked, and even though I was a little weak in the machine learning or building models, that part, I went ahead and I thought that I know all those part even though I was a little weak. But it’s like during the application process and during the interviews, it somehow got clear that this is the data science analytics or the data analytics part is where I’m more interested in and I’m just more confident about and I’m not that great when it comes to model building or optimizing. Yeah.

Kirill Eremenko: Interesting. So that’s what I’m trying to get to. So that happened not at uni but that happened through the job application process, that you understood, “Hey, I actually prefer this. This is more interesting to me, this visualization side of things rather than building logistic regressions and other types of models,” which could be exciting for other people.

Harshal Sanap: Yeah. Because-

Kirill Eremenko: Tell us how that happened.

Harshal Sanap: Yeah. Because I would say to be honest, at any university or are at any course you learn everything related to data science, right? But it’s up to you to decide how much do you actually know and even if you know, how much time do you actually want to spend doing that for rest of your life. So I think during the interview process, there are certain things, you just can’t fake it. Either you know it or you just don’t know it. So I think during the interview part I just started realizing certain things. When I was like during interview, I was super confident about certain things. But when it came to little mathematical part, I could say it, I could understand it, but I could not convey that information with that much confidence.

Harshal Sanap: So then I realized like, “Oh, okay, maybe I am a little more on the analytical part. I’m a little more as a business analyst or a business analyst kind of person who understands everything and who likes talking and who likes interpreting all the results.” So that’s when I realized what my strengths are.

Kirill Eremenko: Okay, but tell me this then. So I understand, you can go into an interview and you can, or a couple of them, and you can realize, “Whoa, look. I found myself. I realized that I’m very confident talking about business intelligence and dashboards and visualization and Tableau and all that stuff. I’m really confident with it. On the mathematical side and the modeling side, I’m not that confident talking about it and it looks like it’s not my strong point.” I understand the distinction between, okay, this is my strong point, this is not my strong point. But how do you extrapolate that and put it into this is what I love and this is what I don’t love. Because maybe your weak point or for somebody going in interview, they might find that that’s a weak point but actually they just need to learn more and then they’ll be good at it and that’s indeed what they love. So how do you go from strong, weak to love and want to do for my career or for the foreseeable future, and no I don’t, I’m not really that interested in building a career here?

Harshal Sanap: Okay. So I would say with that, I did a small test on myself. I just asked myself about my strong points. The reason I was strong at that is probably because I had spent time doing those things because I was genuinely interested in. All the weak points, I always had time but I just never dig deeper into it. Maybe because I was not that much interested in it even though it was there back of my head. But I just don’t want it to get too much into details of those things. For example, in data science we have L1, L2 normalization. So we know this is how it works. But there are some people are machine learning engineers, they can… They’d like to derive those things on the mathematical part.

Harshal Sanap: But I was just never interested in it and I realized that it’s better to… It’s like if I don’t want to know those things from the bottom up, like from scratch, then maybe I’m not the best fit for them. So let’s just focus on the things which I am really genuinely interested in and I think that’s how I figured it out.

Kirill Eremenko: That’s very cool. So that’s important that that there was that self inquiry where you stopped and asked yourself.

Harshal Sanap: Right. Yeah, absolutely.

Kirill Eremenko: Very cool. I think it’s a great example to follow for those starting out into data science. Data science has so many different faces.

Harshal Sanap: Right.

Kirill Eremenko: Just getting a data job, job as… I want a job as a data scientist. You might end I’m doing, like you say, something you don’t like, you’re not best fitted for. Therefore, you will maybe get a high paying job. Amazing. But you want to be happy in your role.

Harshal Sanap: Right, right. Yeah, it’s like I would say there are a lot of jobs and you just don’t have to set it… It doesn’t have to depend on the title. So just try to look for what the company’s offering in the job description and then decide accordingly if you are actually suited for it. Because I have seen some data analysts or business analysts job positions which had all the model building, machine learning and optimization part, which was equivalent to a machine learning engineer. So you cannot just go with the title.

Kirill Eremenko: Yeah. Sometimes the companies don’t know themselves what they’re describing.

Harshal Sanap: Exactly, exactly.

Kirill Eremenko: Yeah. Very, very true. Yeah. Okay. So what else? The other thing I wanted to ask to you is these interviews, so you said you were getting an interview here, an interview there. How did you get these interviews? Some people struggle with that. It’s even hard to get an interview in the first place.

Harshal Sanap: Yeah, absolutely. So I would talk about how you could get called from some of the big companies or like how, at least my personal experience with that. Honestly, it is not easy to get calls from any company, not just the big companies but any company because you have to go through a lot of resume polishing and I would say it took me almost seven to eight months or probably more than that to even get my resume noticed by the recruiters. So to start that process was when I actually started my masters program and I was looking for internship, that was the first time I actually did some, I drafted my first resume and then there were probably more than hundred revisions to that resume too.

Kirill Eremenko: Wow.

Harshal Sanap: Because it definitely makes a lot of difference. Some people say you got to put it as a PDF. Some people would say, “Oh, make it Word.” Or some people would say, “Use Calibri format.” Some say Arial format. There are so many choices and it just takes a lot of time to revise, revise, revise, and one day it will hit and you’re like, “Oh, this thing is working.” So you just keep applying with that resume. But after some days, you will have to change that too. So yeah. So my approach or my suggestions to apply for any companies or especially in big companies or to get a call, step one would be try to join in as an intern. So for that you got to start as early as possible. Let’s say if you want to start an internship at Google in summer 2020, their application process starts almost one year in advance. So probably in July or August, their process starts. 

Harshal Sanap: So you got to try to get your hands in as early as possible. The reason why I’m trying to secure your spot as an intern is because almost I think 80% of the people who join in as an intern, they get a full-time offer. All my friends, including myself, we all had full-time offers, which got converted from an internship. Even if it doesn’t give you a full-time opportunity, you have made all the necessary contacts to get back in. So I think that will be my first suggestion.

Kirill Eremenko: So let me understand this a bit better. You were an intern at SmartDrive systems before?

Harshal Sanap: No, I was an intern at a different company.

Kirill Eremenko: Oh, and you’ve got a full-time.

Harshal Sanap: Yeah, at somewhere else.

Kirill Eremenko: But then you chose not to take it.

Harshal Sanap: Yeah, I did not take it because of some reasons, because I already worked with my internship company for almost one year so and there were some visa obligation so I had to make a switch to a different company.

Kirill Eremenko: Oh, visa obligations. Okay, you had to switch.

Harshal Sanap: Yes. Yeah, because they were not sponsoring so I had to switch to some other company and my previous internship company was more of a startup.

Kirill Eremenko: It is hard in general for non-US citizens to get a job in data science?

Harshal Sanap: It’s not a data science job, it’s just job in general. Because I would say there are so many parameters. You need to find a job at a company, which is E-Verified, then you got to find a job which is relevant to your program of study, then you got to find job with the company who’s willing to sponsor you. Then you got to find a job within a specific time period from your graduation so there are so many factors. So it’s definitely hard but everyone can make it. That hard work is something that polishes you. So yeah, it’s definitely worth it. Yeah. So yeah, so that was just suggestion one.

Harshal Sanap: So tip two is try to apply for college graduate positions. Almost, a lot of companies and most of the big companies like Google, Facebook, Amazon, they all have graduate positions which open eight to 10 months before the graduation. So let’s say if you are graduating again in let’s say summer 2020, these positions open around August or somewhere during that time. A lot of people who apply for those positions will get a call because of the cohort size is very small. In the whole United States, or let’s say if the company is hiring in California, there may be a thousand students or probably less than that in that particular cohort, right? So there is a high chance of people getting a call. That is the one thing that really worked for me. I got a call from Facebook and Microsoft.

Kirill Eremenko: Wow.

Harshal Sanap: I did not convert it. That’s a different thing.

Kirill Eremenko: Why, why? So you went to the interview.

Harshal Sanap: Yes. I went into the interview. I will come to the part of why I did not clear it in the mistakes part. But as a college graduate position, there is a high likelihood that you will get a call from these companies.

Kirill Eremenko: Wow.

Harshal Sanap: Because as I said, the cohort size is very low.

Kirill Eremenko: Oh, man. You’re just like describing my life as well, and also making it feel less impressive because I did the same thing. I went for an internship when I was back at… During my masters I did an internship at Deloitte in data science and I did an internship at Ernst & Young in accounting.

Harshal Sanap: Oh, nice.

Kirill Eremenko: Hello to everyone who’s listening from there. Then I joined as a graduate as you said, I applied for the graduate position and the internship didn’t help with the graduate position. I joined as a graduate at Deloitte and yeah, and then after that, that got converted to an analyst role full-time. But back in the day I thought, “Oh, I’m so special. I got a call from Deloitte.” Now you’re saying everybody gets this call.

Harshal Sanap: I’m sorry. This is what-

Kirill Eremenko: That’s okay. That’s okay. The truth, the truth.

Harshal Sanap: But this is what I’ve observed. As I said, you will get a call because the amount of people that are applying for the position is pretty low and they have specific recruiters just to hire from the college. So the possibility of you getting a call is very likely. But again, a lot of students are at different point of their preparation of interview, so that could really affect. So with me, I got these calls when I had just started applying and just started preparing and I was not prepared at all. So I’m like, “Oh, damn.” Because, and this is the underestimating factor, I did not expect or anticipate to get a call. So I thought maybe I’m not even going to to get a call. I just applied and in few weeks they reached out to me. I’m like, “Oh, (beep).” Okay. So maybe I was just under-prepared. So I will talk a little bit more in detail later.

Harshal Sanap: But yeah, so this was like tip two. Third tip would be try to reach out to recruiters in career fair or conferences like DSGO or like there is other conferences like OESC, then GH, Grace Hopper conference. These are a few famous conferences here in United States that I’m aware of. You get to meet a lot of people and a lot of recruiters at career fairs as well as the conferences. There is a possibility that you will not find a job that you are looking for at that particular time, but you may find a person who knows some other person who has an opening and networking can be the key in that particular aspect. So don’t be shy, just go on there and say hello. You never know how that one connection will help you out in future. This is tip number three.

Kirill Eremenko: How did the networking help you?

Harshal Sanap: I would say I got my internship through networking and then I got to, whenever I was preparing for some interviews for some big companies and I did not know about anyone, so from networking I could ask some of my mutual connections for help. Yeah.

Kirill Eremenko: Okay, interesting.

Harshal Sanap: So just just to give you one quick example. For my internship, I think it was around… So internship generally start in before summer, around summertime like in May, May or early June. I started my internship search pretty late, again, some things I just don’t prioritize like the job.

Kirill Eremenko: Sounds like a pattern with you, man.

Harshal Sanap: Yeah. So started looking-

Kirill Eremenko: It’s a Christmas coming up very soon. Have you started preparing for that?

Harshal Sanap: Not yet. Probably on 24th. So for internship as well, I started a little late and I remember I was very anxious. It was April and everyone was telling that their internship is starting in May or early June. I’m like, “Oh my God.” I remember in April there was Just in Time career fair here in San Diego, as the name suggests, Just in Time.

Kirill Eremenko: It should be called career fair for Harshal.

Harshal Sanap: I’m like, “Holy (beep). That’s for me.” I remember I was literally walking out of that career fair like, “Oh my God, I don’t have anything here because all the jobs were for sales and marketing.” Then I remember I saw this company called [Sharp Meter 00:46:28] with whom I interned. They were hiring only for sales and marketing, but I just liked their logo and I just went and I started talking and I think I talked with their VP of sales and we had a good chat and he asked for my resume and I’m like, “Yeah, fine. Everyone just asks for a resume. No one gives a call.” I just gave him my resume and then after a few weeks he’s like, “Hey, we have a sales intern position open.” I’m like, “Come on man, I don’t want sales position.” Then I was like, “How bad would it be? I’ll just at least pick up the call.” Then I pick up the call and he’s explaining everything for data analyst position in the sales department. That’s why they were calling it a sales intern. I’m like, “Holy, (beep).”

Kirill Eremenko: You see. Sometimes they don’t know how to call these positions.

Harshal Sanap: I know. Yeah, exactly. Because apparently, they had a lot of data getting accumulated in the sales department. Not everyone were aware what to do with it. I think that was my first step. So this is again, networking, I got my internship through networking. I did not even see that coming. I had such a great experience. I learned Salesforce, Power BI, Python, so many things at that company. I’m thankful for that.

Kirill Eremenko: All that came from one conversation, random conversation you had, not even in data science fair.

Harshal Sanap: Exactly. Exactly. So that’s how I was like no matter where you are, just go say hello. You never know.

Kirill Eremenko: No, I get it now why you come to DataScienceGO. Being a volunteer, that also puts you like in a position of power. You’re like, “Yeah, I can guide you to that lecture theater. Oh by the way, do you have a business card?”

Harshal Sanap: Oh my God. I [inaudible 00:48:31] really my secrets here.

Kirill Eremenko: Oh man. Awesome. Awesome. Really cool. Well, great tips, great tips. So let’s recap on what do we have so far.

Harshal Sanap: So there was one more I missed on.

Kirill Eremenko: Okay, sure.

Harshal Sanap: Then the fourth one is, so these three were mainly you could do when you’re in a university, but other than that you could also try tapping into your alumni network or just like mutual, random mutual network. Let’s say, oh I want to work in company X, Y, Z. I see some person at that company who’s connected to one of my friend. So try to find that, try to just connect those dots and try to get a referral from them because referral does work and you will get a call and you will be prioritized based off a referral.

Kirill Eremenko: Absolutely, absolutely agree.

Harshal Sanap: Yeah, so.

Kirill Eremenko: I heard that about 70% of jobs are actually filled through referrals.

Harshal Sanap: Yeah. Yeah. Because I realized something, I don’t know how much of that is true that they have a separate portal, one for referral, all the applications that are coming through referrals so they are going to have a look at it first and if they’re not finding what they’re looking for from the referrals then they will look at a big pool of applications that are coming just out of nowhere.

Kirill Eremenko: Absolutely. It’s absolutely true. In Deloitte, that was exactly the process. If you knew somebody who submitted through the system, inside the system, so it’s a very different process. In fact, it was so important to the company. They value the referrals so much that they paid. They paid 5,000 bucks. Not kidding, $5,000 if you refer someone.

Harshal Sanap: Yeah, I believe you. Yeah.

Kirill Eremenko: All you have to pay tax on that. So in the end, it’s like 3,500 but still 3,500 bucks you get in your pocket just for referring a friend.

Harshal Sanap: Yeah. It’s literally like a broker. If you sell this one pretty good house, you get like 10%, something like that.

Kirill Eremenko: Crazy. Tell me a bit about this alumni network because I’m an alumni from two universities, one in Russia, one in Australia, but I never actually stay in touch with the alumni or never participate in any of these things. I don’t know. I just never got into the habit of it. Never saw value. Have you ever done anything with your university alumni?

Harshal Sanap: To be honest, I did not do anything proactively, but I did it when there was a need. Let’s say if I had some, let’s say on LinkedIn, I see this opening, let’s say Google, and I see that I have a mutual connection or like a random connection through SDSU, because I studied at SDSU here at San Diego and I see that, oh there is some person who did the same thing or we started from the same school. So I don’t know. Subconsciously I think that that person would have some empathy. So I will try pinging him like, “Hey, this is a fellow grad student from your school, a fellow Aztecs,” Aztec is a mascot for SDSU. So I’ll just start a conversation like that and then I’ll tell them like, “Oh, I’m just graduating.” Or like, “I wanted to know more about this particular job position. Would you have some time to talk about it? I just want to know about your experience so far.”

Harshal Sanap: Again here, I’m not asking directly for referring me. I’m just trying to know that person well. Yeah. So most of the time it was like the people were not interested in talking. They would be like, “Oh, just give me your email address.” So because they also know they’re going to get paid if this guy gets picked.

Kirill Eremenko: Okay. That’s [inaudible 00:52:38].

Harshal Sanap: Yeah, because it’s a win-win, so.

Kirill Eremenko: Yeah. Yeah. Interesting. Okay. Okay. Very good.

Harshal Sanap: Yeah. So I’ll just summarize sort of the four things that I said was first one is try to join in as an intern. Start early as possible. Second thing was try to apply for college graduate positions. Third thing is try to reach out to recruiters at career fairs or conferences. The fourth thing is try to tap into your alumni network or some random mutual network for some referrals.

Kirill Eremenko: Fantastic. Okay, let’s move on to the next step. So all right, you got the interview. How do you prepare?

Harshal Sanap: Okay, so here comes the big part. So first and foremost you need to read the job description very carefully and you need to understand what exactly is it that they’re asking for. Let’s say if a job description tells you that you need at least two years of SQL experience, then make sure that you know everything from basics of SQL, like let’s say basic joint or aggregation function to complex joints and window functions because they are mentioning at least two years of experience.

Kirill Eremenko: So do you need that two years?

Harshal Sanap: To be honest, it totally depends on the company. I did not have a professional SQL experience, but I did do well at some of the interviews which said we want at least two years of experience and something like that. So I would say you don’t definitely need two years of professional work experience as long as you can come up to the speed of a two year experienced professional, because currently from the internet you could learn so much. So I think everyone can come up to anyone’s speed. So as I said, if the position says you need two years of experience, you don’t, even if you don’t have it, you need to be mentally prepared to have that much amount of experience is what I’m saying.

Kirill Eremenko: Got you. So you kind of like, from that number of years they’re requesting, you can gauge how advanced do I need to be and then just get up to speed with that number.

Harshal Sanap: Exactly, exactly. Because some people would just say familiarity with SQL. So they will not ask you more than, “Oh, so what is the level of hierarchy of SQL function or like SQL script?” So something like that. So yeah, so that’s the number one thing. Try to read the job description very carefully. Then second thing is research most frequently asked interview questions. You could start off with Google or Glassdoor or sometimes Reddit also helps. So there are like a lot of people who are just like you who either went through interview process just-

Kirill Eremenko: With that particular company?

Harshal Sanap: Yes. With that particular company, you will most of the time find something either on Glassdoor or some individual random blogs. If you just Google it on, I mean if you just Google it, let’s say if I had an interview with Facebook, I’ll be like, “Oh, Facebook data analyst interview questions,” and you will find something and it’s not that-

Kirill Eremenko: Sorry, especially for big companies, right?

Harshal Sanap: Yeah.

Kirill Eremenko: They have standard process in. When I was going for my graduate interview or actually for my internship I think at Deloitte, I did the same as you said literally the day before. Obviously you should do it earlier, but I’m like you. I’m in the Harshal camp. I wait until the last minute. But I did just one Google search. I found this forum, which is quite popular in Australia, is called the Whirlpool Forums. There, somebody actually described the whole process and basically not that I got access to questions or anything like that, I just got familiar with the process in terms of psychologically.

Harshal Sanap: Yeah, right.

Kirill Eremenko: For instance, for me it was like you [inaudible 00:56:54] up. There will be a laptop. There’ll be like a room where 20 of you will sit, there’ll be laptops and there’s still… This exercise, we will get these emails and you need to reply to these emails, make sure to reply calmly and in order of that they’re coming in because they will just like … It’s a stress test. So I knew in advance that, all right, what should I be prepared for? But absolutely agree, man.

Harshal Sanap: Yeah, absolutely. It’s not just like the big companies for which you will get it, but you will be surprised that even for some small companies, you will find some insight. Just to give you an example, there was this one local company here in San Diego that I interviewed for and on Google somewhere, I found a blog which talked about them that they keep a secret aptitude test before the in-person interview even starts. I was like, “Come on. These people are maybe just joking.” But on a safer side, I always had that on back of my head. So they had given some sample questions and I just kind of prepared for those and I took some more sample questions from one of the website, which that person recommended and they did ask me to take an aptitude test and I’m like, “Holy (beep).”

Harshal Sanap: Wow. So it’s like you never know what will work out. Having said that, all the questions which you will find online are absolutely not going to be the questions which an interviewer will ask you, but you will get an idea of what to expect or what the level of expertise will be required.

Kirill Eremenko: For sure.

Harshal Sanap: Yeah. The third for preparing is try to reach out to people who are having the job title in the company in which you’re going to interview. Let’s say if I had a call from company X, Y, Z for a data analyst position, I will try to find some mutual connection on LinkedIn or through some other friend and see if it that company X, Y, Z if there is some other analytical person who’s working. So from that person I can understand how the interview process actually happens and that person would give me a new perspective and maybe that person would also calm me down to take it easy.

Kirill Eremenko: Yeah. That’s good.

Harshal Sanap: Yeah. I think I have two more. Yeah. The next one is obviously try to reach out to some mutual network connection. The last one, the most important thing is how to prepare the content. So this was more of getting to know and getting familiar, but you can… You would always have to put effort in the actual content. So for data analytics, I have few fundamental content topics. The first one is obviously SQL. Personally, I believe SQL is very important and essential because any company you go to, they’re going to have a relational database and as a data person, you need to, your first step is to extract the data. There is nothing more important than SQL, I believe. So you got to be very confident with SQL.

Harshal Sanap: Some of the things that helped me out were few of the courses which I took here at SDSU were focused on SQL. But I learned a lot online from Coursera SQL for data science. That course was really good. I also practiced from HackerRank and Mode, Mode.com. So again, I’m not sponsored by anyone. I’m just sharing my personal experience that helped me. So this was for SQL. Then second was Python. So Python, I think Python.org is really good. Analytics Vidhya was really good.

Kirill Eremenko: Yeah, it’s good. They also has SQL stuff.

Harshal Sanap: Oh yeah, yeah. But it’s like, yeah, some people will fine SQL stuff really good over there. But I think maybe, for me, for SQL, the Coursera thing worked well. So it’s again like a personal preference. So SQL done. Python done. Then for R, I learned a lot from the book ISLR, that’s Introduction to Statistical Learning in R. A lot of people also are finding Matt Dancho’s coursework really good and it’s just like people will focus on R.

Kirill Eremenko: Yeah. Did you meet Matt Dancho at DSGO?

Harshal Sanap: Oh yeah, absolutely. Yes.

Kirill Eremenko: How cool was that? So you knew about him before that.

Harshal Sanap: Yeah. I even went out for like food, like drinks and food with him.

Kirill Eremenko: That’s awesome. That’s so cool. That’s so cool.

Harshal Sanap: Yeah. So-

Kirill Eremenko: Yeah, Matt Dancho’s stuff is really good and you can find it. What is it? Business Science?

Harshal Sanap: Business Science, yeah. Business Science, yeah.

Kirill Eremenko: Yeah. Actually, I’m talking to Matt, so Hadelin and I, we’re talking to Matt about a new business opportunity and I’m planning on inviting him to the podcast maybe like January, February, 2020 so that we can discuss. But there’s like if this, if we pull it off, it’s going to be something really cool. We’re going to start a really cool new place where people can learn. But you’re right, Matt is amazing. His articles are great.

Harshal Sanap: Yeah, because I think it was DSGO 2018 where I sat at one of his classwork because I remember-

Kirill Eremenko: Or workshops.

Harshal Sanap: Yeah, workshop, which happens on Friday. So I remember sitting for his workshop where he was trying to predict defaults, credit defaults using R and it was super powerful and the way he taught it, he just made it look so easy. Wow. So I think he’s also doing a really good job there. So then the next part is machine learning. So for machine learning, most of the stuff I learned was from my classwork, but for others, there is a lot of content. One of the famous course is by Andrew Ng on Coursera, which is free. Then Analytics Vidhya has a lot of topics on machine learning which simplifies a lot of things. Even Randy’s website, Claoud.ML is also a really one-stop solution.

Kirill Eremenko: Randy Lao, right?

Harshal Sanap: Yeah. Randy Lao.

Kirill Eremenko: Did you meet Randy Lao at DSGO 2018?

Harshal Sanap: Of course.

Kirill Eremenko: That’s so cool. You met all your heroes.

Harshal Sanap: You would be surprised. A lot of people who were… Because I was like hanging out with Randy, they started asking me data science question. I was like, “I’m not there yet, guys, but okay.” So yeah. So that was for machine learning. The next thing is in general data warehouse and business intelligence because I believe even if you’re applying for data science or data engineering, a data warehouse and business intelligence concept are real important. So I think Ralph Kimball book called The Data Warehouse Toolkit is very good. The last thing is a metrics and A/B testing. For that, we have three courses are offered by Udacity. I think it’s in collaboration with Google. 

Kirill Eremenko: Wow.

Harshal Sanap: That’s a good thing. Now, if you think you are prepared with SQL, Python, R, machine learning, data warehouse and business intelligence metrics, et cetera, et cetera, and if you want to practice the interviews then you could go on to Lead Core. They have paid as well as free membership. There is also another website called Pramp, P-R-A-M-P. It literally stands for practice makes perfect. So over there, a lot of people in the software industry, they do Pramp because they don’t practice coding rounds, but I think they have also started a data science mock interviews over there as well. So I would say Lead Core and Pramp. They’re really good for practicing your interview skills. Obviously you could also try to Google most commonly asked questions for company X, Y, Z. So these are my five takes for preparing.

Kirill Eremenko: Fantastic.

Harshal Sanap: Do you want me to summarize it?

Kirill Eremenko: Yeah, yeah. If you could summarize them, that’s really good.

Harshal Sanap: Yeah. So just to summarize it, step one would be to understand job description very carefully, see what they’re exactly asking for. Second thing is research the most frequently asked questions to understand the level of intensity with which you could expect the questions. Third thing is try reaching out to people or your mutual network on LinkedIn or through other friends and see if you have some mutual connection who are working in the company where you are being interviewed just to understand how the process goes. The fifth thing is obviously the content for data analytics. It’s more focused on SQL, business intelligence, Python and I did not mention Tableau, Power BI and all those things because they are important as well. If it is more focused on data science roles, then you got to focus more on machine learning part and understanding data and logic and statistics and all those aspects. Oh, I also missed on statistics. Yeah.

Kirill Eremenko: Very important. Where did you learn statistics?

Harshal Sanap: Statistics, I would say I had a very good teacher in my program. I learned from there. I think I also had taken one course somewhere online, I forgot. Yeah.

Kirill Eremenko: Okay. Very cool. Very cool description. Thank you so much. I’m a bit surprised you didn’t, in all those things, didn’t mention any of our courses. Do you know that we do courses at SuperDataScience as well?

Harshal Sanap: Oh yeah, absolutely. That’s why I said I forgot to mention the Tableau part, my bad. I think I was introduced to Tableau by you.

Kirill Eremenko: Oh, okay.

Harshal Sanap: I’ve taken your course like the first time, I think it was January, 2018. It was my winter break. That was the first time I got introduced to Tableau and that’s when I started stalking you or following you.

Kirill Eremenko: Of course, yeah.

Harshal Sanap: Okay, I got to see this person.

Kirill Eremenko: Yeah, okay.

Harshal Sanap: So yeah, that’s when I knew, I think I got to know about you from Udemy.

Kirill Eremenko: Yeah, that’s right. That’s right.

Harshal Sanap: Your first Tableau course. I think my second course I took of you was the Machine learning A to Z, that was really good. Absolutely.

Kirill Eremenko: Oh, okay. Yeah, cool.

Harshal Sanap: It was you and Hadelin.

Kirill Eremenko: Yeah, yeah, Hadelin. Yeah.

Harshal Sanap: Yeah, you and Hadelin. Yeah.

Kirill Eremenko: Yeah, that’s cool.

Harshal Sanap: So I think those were the courses and that’s how I got interviewed.

Kirill Eremenko: That’s awesome, man. That’s awesome. I was just asking because I wanted to, in case you hadn’t seen them, that could be additional resource, but if you’ve seen them. Awesome. Awesome.

Harshal Sanap: I mean, I would say that’s how I got introduced and then I learned and then it’s like I forgot to mention. [inaudible 01:08:58] longterm back. Yeah.

Kirill Eremenko: All good. The other thing I want to say, that is a really comprehensive outline of, the first one was how to apply. Second one is how to prepare. I’ll actually just add that when you go to the interview, make sure you have a question for the company, right? That’s the worst thing. When I interview people who are applying for jobs in SuperDataScience, I’m like at the end I always ask like, “Do you have any questions for me? This is the one time when you should be asking them.” When people say no, how boring is that? You don’t care enough about the place you’ll be working at. Obviously, you should have questions.

Harshal Sanap: Yeah, absolutely.

Kirill Eremenko: But yeah, what I was saying it’s very comprehensive. You should write a blog post and publish it on Medium about all this. These are some really cool overview.

Harshal Sanap: Yeah. I tend to relays with all those things. Yeah. But, yeah.

Kirill Eremenko: Well, yeah.

Harshal Sanap: That’s why I was super happy when you start off like, “Yay, I’m going going to get off my content out.”

Kirill Eremenko: That’s true. That’s true. That’s very good. We will write up show notes for this episode, which people can go and download all this, all the materials you mentioned, get all the links so they don’t have to remember all the links now.

Harshal Sanap: Oh yeah. Probably maybe sometime I’ll check with you guys or ask you, maybe I’ll just, at one of the meetup, I’ll just present all this stuff.

Kirill Eremenko: That’s a good idea. That’s a very good idea.

Harshal Sanap: Yeah, absolutely.

Kirill Eremenko: Let’s quickly go over the mistakes.

Harshal Sanap: Oh, yeah.

Kirill Eremenko: You said you had some mistakes that you could share.

Harshal Sanap: Oh yeah, absolutely. So I think this is one of the most important part because obviously these are the mistake which I’ve done and most of my colleagues have done. If I could tell, like warn the people then they’re more cautious and they would not do it. So the number one mistake was don’t wait until the last moment to get the application out. It’s like most of the time, you will just lose onto an opportunity if you just wait too long. As I said, for most of the big companies, their internship slot opens almost one year in advance and the college graduate position opens almost more than six to eight months in advance. So don’t wait until the last moment.

Harshal Sanap: Obviously, most of the companies are hiring throughout the year, but there are certain things like internship or college grad positions, they get filled up super fast and they are like the easy way in. Because as a college grad, the amount of questions that will be asked to you are obviously going to be less than if it will be asked for an experienced professional. So don’t wait until the last moment and don’t try to… Don’t also wait for preparation until the end moment. I think what I did was I always prepared, prepared, prepared and then I started applying. But everyone can do it the other way as well. Just be prepared and then no matter which company gives you a call, you’re always ready.

Harshal Sanap: So I think with me, what happened with Facebook was I applied in December exactly in last year, early December and I got a call during Christmas time. I could have never anticipated that, right?

Kirill Eremenko: Yeah.

Harshal Sanap: So it’s always ready to be prepared and-

Kirill Eremenko: So you’ve got the call and then what? You failed on the call or you went to the interview and then that didn’t work?

Harshal Sanap: So I think I did not clear after the second round, yeah.

Kirill Eremenko: You did go to the interview.

Harshal Sanap: Oh yeah, yeah. I did go to the interview. So I think I just did some horrible things in the coding round.

Kirill Eremenko: What did you do horrible?

Harshal Sanap: So that’s the mistake number two. The mistake number two is don’t underestimate your potential. So most of the time, what happens is in from big companies, you will be… You will get a call and the recruiter will tell that your interview is set up with person X, Y, Z. He has a PhD in statistics from Stanford and by default it’s human nature, you have always heard about Stanford, MIT and all those big names, Ivy Leagues, and it’s human nature to feel just feeling inferior. I think I somehow underestimated me and I always thought that even if that person is going to ask me a simple question, it’s going to be complex. So I think that was one thing that I learned from my Facebook interview. Even though if I look at their questions now, I’ll be laughing.

Harshal Sanap: But at that point I was just like, I underestimated myself. Even though it was a simple question, I thought that it’s complex and I tried to use window functions even though it was not required and I just blew it off. So it’s again learning. It was I think my first interview when I started applying for jobs.

Kirill Eremenko: What was the question?

Harshal Sanap: I think it was somewhere on the lines of aggregation. I think-

Kirill Eremenko: In SQL?

Harshal Sanap: In SQL, right. I think it’s like most of the companies they have basic format like, “Oh, I want to know on a daily basis what is my X, Y, Z measure. So can you do that?” So they will just give you a bunch of transactions and they just want to do, they want to know a percentage of a measure or something along those lines. So I think most of the interviews that I’ve gone to, they will ask questions along these lines. But as I said, I was in the beginning phase and I was so much, I was inundated with knowledge. So even though it was a simple question, I was like, “I’m going to try to impress them with a window function,” and I blew it off, but it’s again, learning, so it’s fine. Yeah.

Kirill Eremenko: Yeah. Interesting. So that’s two tips. Any more?

Harshal Sanap: Yeah, absolutely. So I have a couple more. The next thing is don’t get stuck or obsessed with one thing. What I’ve seen with a lot of people is that they have certain sort of goals in their mind, which are most of the times company specific. They think that, “Oh, if I make it to company X, Y, Z then I will be successful. Then I will be worth,” and they just get stuck and they just get obsessed with that thing. Let’s say if they don’t make it to the interview, they get depressed. So I would say one thing, one tip is don’t get obsessed with a company or a position. Try to get obsessed with your job and try to find happiness and in the quality of work, which you will do regardless of the company or the industry that you’re working in and one day eventually your work will get you where you want to go. That’s my personal take.

Kirill Eremenko: Very, very solid tip. I like that.

Harshal Sanap: Yeah. Then the next one is don’t try to compare yourself with others. This is a most common mistake done in college. One thing that we, all the people who are in college or who are just graduating, the most important thing that they should understand is everyone is at a different level in their life or in their level of expertise. Some people are coming from engineering backgrounds so they are by default going to be good in coding. Some people are coming from a non-engineering background so they may be are good in presentation or some other parts. So don’t try to compare yourself with others and just try to focus on your own part. If someone is like succeeding at a higher pace than you, just try to learn from them and don’t try to compare or be jealous of that person because it’s just waste of energy and emotions. But just try to learn from that person and don’t compare.

Kirill Eremenko: But how is that relevant to an interview?

Harshal Sanap: I would say it’s very important because I would say as a grad student, an interview process, it’s like a rat race. It’s like everyone is trying to make it big, right? So let’s say if I get a call or if I make it to a big company, let’s say Google, and if one of my best friend makes it to Google and if I am not making it to a company which is as good as that or a company on the similar level, then I will always have a dissatisfaction within me.

Kirill Eremenko: Oh, got you. Got you.

Harshal Sanap: So that’s why I always say that don’t try to compare because it’s just going to make your life sour or the relationship sour. So that’s one thing. Don’t try to compare yourself with others, just try to fight your own war. Then another thing with finding a job or internship is try to not get burned down. So a lot of grad students, what happens at the end of graduation is they are super overwhelmed with a lot of things like their graduation project, their thesis, then they also know that if they are international, they also know that they need to find a job within three months of graduation. Then they have to find a job which is going to sponsor them.

Harshal Sanap: Oh, then they realize, “Oh, we have so much education loans.” So yeah, a lot of my friends, including myself, we got burned down because we started overworking and we started ignoring all the things that we used to enjoy. We stopped hanging out. We stopped going to a park or like a gym. So we stopped doing all the activities that would actually would make us happy and we would just work, work, work. We all had reached a point where we were like, “Oh, we can’t feel any things because we all got burned down.” So I would say even though finding a job or applying for interviews or preparing for interview, even though it’s a full-time job, try to balance it out with doing all the things which you enjoy because that is going to make you more focus and you will not feel overwhelmed with everything.

Kirill Eremenko: Wonderful. Great advise.

Harshal Sanap: Yeah, yeah. I think the last one is try talking to like-minded people. What I mean by that is you are not the only person who is applying for jobs and who is looking, who is out there. It’s not a solo fight. There are a lot of people who are looking for a job where you could reach out to them in your college or on LinkedIn. So this just kind of gives you a sense of not fighting a war alone. Somehow I feel, I don’t know the scientific fact behind it, but sharing does help you out and you just know that you are not the only person who’s being miserable right now.

Kirill Eremenko: Just thinking, but aren’t you competing with those other people. So why would you share and isn’t that shooting yourself in the foot?

Harshal Sanap: I would say what I mean by sharing is more of like how you are feeling in general, like how caught up are you in the job market. It’s fine if a lot of people, when they get a call from this particular company, they do not share until they make it. I’m not asking you to share with everyone. Let’s say if you get a call, like you got to tell the whole world. But you will always have that one person that you could rely on like buddy or you’re ally, try to share with that person. Everyone would have that person. So who’s not going to judge them or who’s going to be happy for you. So try to-

Kirill Eremenko: Like support, emotional, mental support person.

Harshal Sanap: Yeah. Yeah. You could call that. So because as I said, during, you or me, myself will not feel it at this moment. But I know when you are walking down the graduation ceremony and you have so many things on your head, people do get overwhelmed with a lot of things. So this thing definitely helps.

Kirill Eremenko: Yeah, that’s crazy. How are you feeling now? Now that you have a job, you’re all kind of like, you’re safe.

Harshal Sanap: I’m just chilling. I’m just chilling way too much now. I need to. Because I had decided that until end up this year, I just want to relax. Just appreciate whatever that I’ve accomplished so far. So even when I come for the conference, I did not have any goal in my mind. I was just there to help people out. If someone was coming and reaching out to me, I was just helping them out. I actually had a very funny story. There were a bunch of students that were explaining, they’re discussing something related to some Tableau or something related to SQL and they were stuck at some point and I overheard and I gave my input and they were like, “Oh my God, how do you know? We thought you were just a volunteer.” I’m like, “Go home, guys. I know everything.”

Harshal Sanap: So it’s like, yeah, I would say I’m doing really great at this point. Yeah. So just to summarize. Few of the mistakes to avoid is first, don’t wait until the last moment. Second thing is don’t procrastinate for application as well as preparation. Then don’t get stuck or obsessed with one thing. Fourth thing is don’t compare yourself with others. Fifth is don’t under or overestimate your potential and never miss an opportunity. Sixth thing is don’t get burned down. Seventh thing is try talking to like-minded people. I wish everyone a good luck with all their job process.

Kirill Eremenko: Fantastic, my friend. Thank you so much. Very, very thorough overview of how to apply specifically for grads. We will make sure we mentioned that in the podcast title so that it’s the people who know, who are graduating right now, listen to this. I think you’ve probably helped a lot of people by sharing. Thanks so much.

Harshal Sanap: Yeah, absolutely. Because I know that if I have been through this and I know there are so many grad students here in the States as well as in Canada and Europe, and if I could feel this, and a lot of my colleagues and friends, they went through the same thing, then it’s pretty sure everyone is going to go through the same thing. So I was like, “I need to go out there and share this content.”

Kirill Eremenko: Fantastic. Fantastic. Is it okay for people to get in touch with you if they have any follow up questions?

Harshal Sanap: Oh yeah, absolutely. You can share my LinkedIn. I’m pretty, I’m not that frequent at this point, but I will be frequent now.

Kirill Eremenko: Nice. Very cool. So we’ll definitely put that in the show notes.

Harshal Sanap: Yeah, absolutely.

Kirill Eremenko: Yeah. Very cool. Very cool. What are your plans for Christmas and New year’s?

Harshal Sanap: To be honest, I had just gone for a small vacation for three weeks on the East Coast.

Kirill Eremenko: Small vacation, three weeks? My friend, that is a luxury. I can’t afford myself ever. Three weeks, that’s crazy.

Harshal Sanap: Sorry. Not three, two and half. Two and a half.

Kirill Eremenko: Two and a half. Small vacation. What is your big vacation like? I think you kind of like misunderstanding this whole full-time employment. I know you’re new to it, but there’s no more three month holiday during the summer.

Harshal Sanap: Maybe I think out some Italian genes like you.

Kirill Eremenko: Probably, yeah. Oh man.

Harshal Sanap: The only reason why I did that is because I had a big Thanksgiving week, which was off. I took like another few days here and there. So I got almost two weeks off and then I went-

Kirill Eremenko: That’s a really cool ccompany. You’ve only been there four months and you’re already getting three weeks off. It’s crazy.

Harshal Sanap: Hey, because I put in a lot of work. Okay.

Kirill Eremenko: Yeah, that’s good. No, no, I’m not judging you. It’s good. It’s good.

Harshal Sanap: I’m just kidding. Yeah, yeah. It’s a really cool company. Everyone is super supportive. So yeah. It’s like as long as I’m getting my job done and everything is organized, it’s fine.

Kirill Eremenko: That’s very cool. You should bring some of your colleagues to DataScienceGO next year.

Harshal Sanap: Oh, yeah, absolutely. Yeah, yeah.

Kirill Eremenko: Yeah. Hit me up. We’ll arrange something. It will be cool to get your team there and maybe your boss, maybe you’ll find some more, if you’re growing the team, you’ll find some more people to join the company there.

Harshal Sanap: Oh yeah, sure. Absolutely. Yeah, so as I was saying, for Christmas there is nothing, specific plan because I just came back from vacation. Fun fact is on my vacation on the East Coast, I saw snow for the first time.

Kirill Eremenko: Oh wow. Congrats. That’s awesome.

Harshal Sanap: It’s so funny.

Kirill Eremenko: Yeah. Well, my friend. Thank you so much for coming on the show, for sharing all the insights. What a pleasure.

Harshal Sanap: Yeah.

Kirill Eremenko: It’s been great.

Harshal Sanap: Yeah, absolutely. Thanks a lot for having me on this show and I look forward to meeting you at DSGO 2020.

Kirill Eremenko: Yeah, November, 2020. I’ll see you there, man.

Harshal Sanap: Oh, awesome.

Kirill Eremenko: All right, take care.

Harshal Sanap: Thank you so much, man. Have a good weekend.

Kirill Eremenko: Thank you ladies and gentlemen for listening in on this conversation on this podcast. It went over time, but that was intentional because we had to get all those tips out there into the world so that if you’re preparing for your data science interview and you’re a graduate at a university or you’re about to be a graduate, that you’ve got all the information, all the value that Harshal had to share. In fact, he actually had some additional things to talk about specifically personal experience with some of the big companies. We just didn’t have enough time for that part of things. You can always hit him up, contact him. His name is Harshal Sanap. You can find him on LinkedIn.

Kirill Eremenko: Also, of course, all of the show notes, URLs, materials, everything that we mentioned will be available in the course notes including Harshal’s LinkedIn, the course notes are as usual available at SuperDataScience.com/331. That’s SuperDataScience.com/331. Get all your show notes there, including the transcript for this episode. My personal favorite take away from this episode was start early. So this is a good time. It’s January, 2020 now. If you haven’t applied for your internship or graduate positions for the summer in the US then this is the best time to jump on top of it or actually already a little bit behind. If you’re still over a year from graduating, a year and a half or a year, then this is the perfect time to start thinking about these things, applying for internships, grad positions and so on as per Harshal’s advice.

Kirill Eremenko: So that’s probably one of the biggest takeaways for me because I was also in the same boat and my brothers have been in the same boat. Everybody leaves it always too late. But of course there were plenty of other great tips Harshal shared with us today. So I hope you enjoyed this podcast as much as I did. If you know anybody who is going through data science or business intelligence analyst or data analyst interviews at the moment, then share this episode with them. This might actually inspire them. This might actually help them get the job that they’re dreaming for or maybe make a change in their approach and apply for different other jobs. This could really impact their careers. So if you know anybody, send them the link. It’s SuperDataScience.com/331.

Kirill Eremenko: Finally, as Harshal mentioned, volunteering at DataScienceGO made a huge impact on his career and his life and his connections. So if you’re in San Diego or in the area and you would like to volunteer at DataScienceGO in 2020 then head on over to DataScienceGO.com/volunteer. That’s DataScienceGO, one word, .com/volunteer, and join the team in November in San Diego to really make an impact on other people’s lives. Can’t wait to see you there and until next time. Happy analyzing.