Nick Singh: Ace The Data Science Interview

Nick Singh is ex-Facebook, ex-Google, and co-author of the new best selling book entitled Ace The Data Science Interview: 201 Real Interview Questions Asked By FAANG, Tech Startups, & Wall Street!

In this episode, Host Ted Hallum and Nick Singh unpack everything you need to know about the data science hiring process, to include: resume tips. building portfolio projects, cold-contacting hiring managers, the behavioral interview, and the technical interview!

FEATURED GUEST:

Name: Nick Singh

LinkedIn: https://www.linkedin.com/in/nipun-singh/

SUPPORT THE DATA CANTEEN (LIKE PBS, WE'RE LISTENER SUPPORTED!):

Donate: https://vetsindatascience.com/support-join

EPISODE LINKS:

Ace the Data Science Interview (book): https://tinyurl.com/ace-the-data-science-interview

• On Amazon, as of 04 JAN 2022, Ace the Data Science Interview is the:

• #1 Best Seller in Data Mining (Books)

• #1 Best Seller in Mathematical & Statistical Software

• #1 Best Seller in ODBC Networking

Nick Singh's Blog: https://www.nicksingh.com/

PODCAST INFO:

Host: Ted Hallum

Website: https://vetsindatascience.com/thedatacanteen

Apple Podcasts: https://podcasts.apple.com/us/podcast/the-data-canteen/id1551751086

YouTube: https://www.youtube.com/channel/UCaNx9aLFRy1h9P22hd8ZPyw

Stitcher: https://www.stitcher.com/show/the-data-canteen

CONTACT THE DATA CANTEEN:

Voicemail: https://www.speakpipe.com/datacanteen

VETERANS IN DATA SCIENCE AND MACHINE LEARNING:

Website: https://vetsindatascience.com/

Join the Community: https://vetsindatascience.com/support-join

Mentorship Program: https://vetsindatascience.com/mentorship

OUTLINE:

00:00:00​ - Introduction

00:02:28 - Nick & Kevin's motivation to write a book

00:10:59 - Why does the book lead with four chapters that aren't about the technical interview?

00:17:21 - Transmitting is not the same communicating

00:18:57 - Be audience centered

00:20:09 - Portfolio projects

00:38:06 - The power of cold-contacting hiring managers

00:49:00 - Behavioral interviews demystified

01:00:48 - Which of the technical interview chapters is the most critical for landing employment in the datasphere?

01:05:53 - Which of the technical interview chapters is the most difficult to master?

01:09:05 - Looking at the datasphere spectrum, what are the biggest differences between doing data in big tech vs. in support to the federal government?

01:17:04 - The world's a really malleable place. You can double your salary in a year...and completely change your life in two years.

01:24:06 - Make sure that you're learning things in the correct order. Good foundations matter.

01:25:43 - Consider community college courses as part of your upskilling path. They often offer excellent and affordable learning opportunities.

01:31:33 - Farewells

Transcript

DISCLAIMER: This is a direct, machine-generated transcript of the podcast audio and may not be grammatically correct.

[00:00:07] Ted Hallum: Welcome to the data. Canteen, a podcast focused on the care of feeding of data scientists, machine learning, engineers who share in the Kanban abuse, military service. I'm your host, Ted Hallam. Today I'm joined by Nick Singh. Mixing is the coauthor of this book. He is also a supporter and advocate for. And he's very concerned that you make it through the data science interview.

Hence his book, title, ACE, the data science interview. I first met Nick when his book debuted back in October. He and I met up for virtual coffee and after I talked to him about it and received my copy from Amazon and was impressed with what I saw if you're in our slack workspace and you went to the channel resources, underscore books, Then you'll be able to see more information about this book to include what's on the back cover and a link to where you can pick up a copy on Amazon if you're interested, but in the conversation that Nick and I are about to have you learn everything that you would need to know about the data science interview process everything from what is the most important part of the technical interview for you to master all the way to the hardest part of the technical interview for you to master.

And then some of the other things that are peripheral to the technical. Like the behavioral interview and how to cold contact hiring managers and recruiters to make sure that the hard work you've done on your resume and portfolio project actually gets seen by people who have a say in whether or not you get hired.

So I hope you enjoy the rest of this conversation. And here we go.

Nick. Welcome to the show, man. So good

[00:01:37] Nick Singh: to have you. Yeah. Thanks Ted. Excited to be on the data. Canteen. Just talk a little bit more about the data science interview process and job hunting. Yeah,

[00:01:46] Ted Hallum: absolutely. So I first became aware of your new book in October, I think, early October. And immediately I thought about the community of veterans and data science and machine learning, because I get these questions all the time. People are going through the interview process. They're going through the hiring process.

They're either trying to get their first job in the data sphere, or they're trying to, you know, move up the career ladder and everybody wants to know how can they best prepare themselves. You know, the technical interview for a lot of people is really scary, especially if they've never done one. So I kinda just want to start out by finding out, I know you have a co-author who helped, you know, the two of you collaborate on this book.

So what motivated you guys to write this book in

[00:02:30] Nick Singh: the first place? Yeah, it's not just the veterans community. That's confused about the data science job hunt process. It's everybody, you know, and that's what motivated us to write the book. So. 2020 hit people were having their job offers, rescinded internships were getting canceled.

You know, people were getting laid off and, you know, we thought that there was an opportunity to do help the data science community. And this was a back ID on the back burner for awhile because in other fields like software engineering, there's a book called cracking the coding interview that helps people with their software engineering job hunt.

And then there's a book called cracking the PM interview and then consultants, they have their own book called case in point or cracking the case interview. So I'm like, all right, where's the crap, you know, what's the equivalent in our field because actually our field is even more kind of nebulous than some of these software engineering fields where they have much more structured interview process and shit like data science at one company is called machine learning at the next, and it might be called a data analyst and the third place.

Right. So that kind of compounds the difficulty of even like. Clean structured advice because we're not even calling the same job, the same thing at each company. So we're just like, all right, you know, we've seen some of the same questions. We know that these kind of book exists in other fields. What can we do right now?

So that's what we, that's when we decided, Hey, we gotta get the book out and like start working on it. Cause there's a space here and people need this. And just on the side, we'd already been doing this kind of work. My coauthor, Kevin, he had already been helping he's a former data scientist at Facebook.

Then he went to be a quant on wall street. He had already been helping on the side, people break in to both wall, street and tech. Then on my side, you know, I worked at Facebook and Google. I've been, you know, reading resumes for my friends, giving referrals and posting on LinkedIn, running a newsletter just on the side, you know, just putting it out for free on my website.

You know, some of the best practices I've seen just cause I got really tired of like seeing the same resume mistakes over and over again. So just kind of culminated into, Hey, someone needs to do something about the data science and ML job hunt and like put more information out there. Why, why not it be us?

You know? So that's kind of how the book was born and yeah, that's how we got on.

[00:04:50] Ted Hallum: That's fantastic. So I got my own personal copy, which I held up there. So the audience could see what that looks like. I'm not currently looking for a job, I'm actually in a role that I love, but you know, this is a good resource to have on the bookshelf.

Plus, you know, one thing that really didn't Dawn on me until I had it in my hand, cause I really did buy the book, just thinking it would be a great resource to be familiar with, to recommend to the veteran's day sides of machine learning community. But because of the breadth that the book has to cover, because you're trying to prepare people for all the technical gotchas and yeah.

Sniping that can happen in interviews. It's an incredibly broad and concise resource for like virtually every topic in data science and machine learning. So if you were going to have one book on your bookshelf, it is kind of like the one to have, which is right.

[00:05:40] Nick Singh: Yeah. So this, this book is definitely not the one to learn data science or data analytics with, but it's definitely the one to like, practice where they're like concisely refresh yourself.

And just to allude to that, cause what Ted's mentioning we start with probability interviews, statistics. Then we go to S machine learning. Sequel coding product sense. And then these like open-ended case interviews that are kind of popular, you know, as a, just like open-ended challenges that companies will get.

So there's a lot of material in the book and, you know, it's like, we, we, we admit, you know, people will complain like, bro, you didn't teach us all of statistics in like five pages. You didn't, you know, how is your ML only 40 pages when there's like whole textbooks on machine learning. Right. But Hey, it is what it is.

So exactly. We tried to summarize it because, you know, unfortunately our field is one that's super interdisciplinary that requires a whole breadth of skills. And the interview process reflects that where, Hey, for a data scientist position at one company, you're going to be asked a lot of sequel and machine learning.

And the next there'll be more of a coding type question and a third, it might not even be so technical where they actually expect you to write sequel or code. Maybe they just hit you with these open-ended business and product. Taste studies, where does she know, how would you solve this problem? How do we handle this?

What do we, what metrics would you use to measure the success of our new initiative? You know, these are just more like, you know, trying to get a sense of how you think. So it is a wide variety of questions. And we, we tried our best and, you know, at least people are loving it so far and you know, there's always room for improvement.

So, but yeah.

[00:07:20] Ted Hallum: So I think with, with cracking, the, you mentioned cracking the coding interview earlier. I think the first edition that came out in like 2008, I believe they're on the sixth edition now. Yeah. So I hope you like updating this because I think people are going to be wanting, you know, data science evolves just as fast as the world of software engineering.

So I think that there are going to be, there's going to be a need for the book to evolve with the

[00:07:42] Nick Singh: field, for sure. For sure. So we actually secretly don't tell anybody this, but we've actually been updating the book. We've updated it like 14 times so far already, but Amazon lets us update it without.

People realizing it. So basically as we find a small typo or a bug or whatever, we're continuously updating it. So the book gets better pretty much every week. So is that like,

[00:08:05] Ted Hallum: if people get the Kindle version type of thing?

[00:08:09] Nick Singh: Yeah. I mean, that's where it's a little unfortunate where it's like, Hey, well we only have paperback.

So if you got the paperback earlier, sorry, but these are incremental improvements. It's not like we're like adding whole chapters or anything. That's

[00:08:22] Ted Hallum: awesome that the hard copy book can, can evolve that's

[00:08:25] Nick Singh: yeah, yeah, yeah, yeah, exactly. So yeah, people who get the book next week, we'll probably have a better copy than what we will watch.

But I mean, it's still mostly there, you know, I don't say this to be like, Hey, the quality that you know, but it's just like, Hey, we're always hungry to improve. Why it's improving. So anyone listening to it right now, it's like right now is the best time to, to get the book. But, you know, as, as, as the time goes we're changing less and less things anyway, just cause like, you know, it might just be a small like comma or whatever.

You know, not, not like core to the material, but you know, one, one, interestingly you said was, you know, data science, machine learning, it's a fast changing field, but actually the interview process doesn't change as fast as you think. So I've seen some of the earlier copies of cracking the coding interview and, you know, as she's added more additions Gayle McDowell, she's the author she's added more content and like cleaned up content, but like the core of like how people are being tested and software, honestly, hasn't changed much from like, you know, in like five or six years, you know, even though she's had two additions and I, I kind of anticipate the same and you know, if it does change, we're going to be here to react to it and like keep updating and material, but, you know, see.

People think that some new technology is going to kill sequel, but people have been doing SQL since the end of the nineties and will continue to be writing SQL, you know, and these statistics questions about the central limit theorem or hypothesis testing. Sure. There might be a new tool to run a Z test for T tests.

But at the end of the day, when an interviewer asks you about a Chi square test or Z test or T test, it's the same thing. And same way with linear regression, you might have a new package or library, but less often you're asked about like, Hey, in scikit-learn how do you run linear regression? And more you're asked like, Hey, tell me about some of the assumptions of linear regression.

Right? So as long as that doesn't change and it won't, it's not going to change. That's not going to change the book should still be in business. Exactly. So I don't, you know, I'm not saying this as like, oh yeah, the book will be in business. I mean, to say like, Hey, the skills you should focus on and learn, and that are.

More often than not are the timeless skills. So that's one other thing I just bring up just because, you know, it's so easy, you know, for people to hear what I'm talking about and be like, oh great. Why even bother with this? There's so many skills that I need to know and they'll change. And she's like, well, what, what's the point of even doing all this work?

And I just always want to give like a message of like, Hey, you know, these things are tough. There's a wide variety of things you would be tested on, but there's like incremental ways to like improve yourself. You know, I, I don't want to ever want to have lead people with a defeatist attitude of like, oh great.

I need to know 11 things that will change next year. It's not like that

[00:10:59] Ted Hallum: right now. Of course the meat of the book is focused on the technical interview, but I thought it was interesting. The book actually leads out with four chapters that are about things other than the technical interview. So. Just tell us about that decision, about what those four chapters cover and why you thought that was so critical to include in addition to the parts about the

[00:11:22] Nick Singh: technical.

For sure. Yeah. So as we mentioned, we have those technical interview chapters about like sequel and stat and ML, the first four chapters though, they're kind of more about ACEing the job hunt, not the interview process, but just like, how do we get more interviews in the first place? So we talk about resume tips and how to craft a good resume.

We talk about portfolio projects and why they're so important to build. We talk about how to cold email and cold contact recruiters and hiring managers so that you can really put your resume and portfolio projects in front of decision makers to get these interviews. And then finally, we talk about the behavioral interview aspect, which goes a little bit more meta than just like, Hey, here's how to answer behavioral interview, but more like, Hey, how do you position yourself as a job seeker so that your narrative aligns with.

They're looking for. So why don't we talk about this? Well, Hey, how are you going to ACE these interviews? You can't get an interview in the first place. And man, oh man. When I was at Facebook, a lot of my friends asked me for referrals and you know, I look at some of the resumes for my team and my goodness, what people make some crazy mistakes, really smart technical people who had all these right skills, who just sucked at presenting that information.

And you think even with these templates and you know, like diamond, doesn't like resources out there on how to write a good resume, you still see people mess up, especially because a lot of those resources might be for not for data science jobs. It might be just for like general people or, you know, an, our industry does work differently and technical jobs do work differently than what a resume would look like for a business world or like more, you know, other kinds of work.

So that, that's kind of why we put it out there. Cause it's like, Hey, we got to get you those interviews in the first place we want to ACE them. And the, yeah. And the second part was just like, Hey, I have a marketing meant, you know, I have some past marketing experience. So for me, and I've always been interested in marketing, like without school, just like, Hey, just thinking about things more holistically and so interesting to see really smart, capable people do really stupid things that are kind of myopic where, because they're not seeing like how it affects their personal brand.

So I'm not trying to be like, you know, some like personal branding mastermind where it's like, oh, how the most perfect LinkedIn and like have a official picture and like, you know, speaking catch phrases, not nothing like that. It's just like, you'll see people volunteer negative information on their resume, where it's like, Hey, like you have a terrible GPA.

You don't have to list it. You know, like you could just not have that line here. And if you had a good GPA, go ahead and list it, you know, and same way people will list random, random, random jobs. Now, if that's all you have, that's. But like, if you have some real experience or relevant experience, you don't have to mention the fact that you have totally irrelevant experience.

Cause that just drowns out. What's good about you, right? Cause of recruiter, they're scanning that in 10 seconds, 15 seconds, they don't have time to read that kind of junk. Right. And it only distracts them from understanding, yo, why are you actually relevant for this position? So that's another example of how people keep messing up the resume.

Another one is people read from left to right top to bottom. That's how we read in English. So let's keep our important information left to right top to bottom. Right? That means, you know, some people will say you have to list your stuff chronologically, but if the most relevant stuff. Was in your second to last job, not your most recent job, like there's ways to change the ordering or emphasis.

Right. And same with portfolio projects. Let's say you worked on three portfolio projects, but the second to last one is the media one, the one you really want to talk about, well, then get rid of your last project where nothing really happened or yeah. Or just move to your last project below your MIDI project.

Right. So some people will be like, Hey, like, you know? Yeah. So that's an example or like the lists they're volunteering experience, which is great. Like it's great that you're part of a community and that's all helpful, but. You know, you, you know, if that's drowning out space from like you talking about your portfolio project, where you're like, actually did Tableau and secret work, I hate to be that person, but like, that's probably way more important than your volunteering.

Like, and of course, people want to know that they're working with a good person, but that comes afterwards. That comes when you're talking face to face, like, what do you do outside of work? Or like, how are you as a person, but to get that interview in those 10 seconds, I just want to see like, yo does this person know SQL and Python and R and all these different technical skills and, you know, do they have past work experience at sort of relevant?

That's what I want to know in 10 seconds. 10 seconds is a very short time. So you'll see all kinds of crazy stuff with colors and infographics. Another one, Ted, have you seen this before? Where people rate their own skills where it's like, oh, SQL advance. I thought in a half, who who's to say, like, what does that even mean?

Like you're only, you know, calling yourself intermediate, like. Why do that? Why, why, why do that right. You know, and if, if you, if you truly don't have skills with something, then just don't list it on your resume, but it's not going to really help you to list a whole bunch of stuff. And then self-rate yourself.

So now people are like, well, you know, whose scale are you using? And then it's also like the fact that you're self rating yourself on stuff that's not important, you know, stuff that I think might be important. And then you just rate yourself as basic, just like, huh. You know, I don't, I don't really think it's a helpful now instead, if you had just positioned it as, Hey, I did this project with Python and then if you're pushed on it, it's like, Hey, you know, this project was just like a three-week long project.

It's not as much as the AR experience I had in this job where I spent one year on AR. So definitely Python is not as good as my Python skills, not as good as ours now that that's allowed that that's good. But you know, if you just want to list Python, your resume, contextualize it with a project which kind of leads into the next chapter of portfolio projects.

Cause I know, you know, Let's talk about resume. Like what have you seen tad? Like, is this kind of resonating? Have you seen some of the same stuff or

[00:17:21] Ted Hallum: absolutely. I've seen some of the same stuff. I think the biggest thing is people fail to think about their resume and their LinkedIn profile at just in the way that you would normally think critically about communication.

So as you know, my background, I came out of the military. I was in the army. And so when I think about this kind of stuff, I think about like radio communication, cause the military communicates a lot with radios. You've got, you always have a transmitter and receiver and communication in this case, your resume is not just about transmitting because if you just transmit and it's not received and that's not communication.

So it's, it's about having your signal be clear and strong enough that it's actually received and it's received intelligibly. And that whoever is on the receiving end, doesn't get fixated on some. Data point, they get fixated on what it is that you actually wanted to communicate to them. Exactly.

And so I think that is a key way for us to think about resumes and that when I see resumes that I feel like falter it's because they've not done that they haven't thought that way about the way the

[00:18:27] Nick Singh: resume is crafted. No, that that's so good. And who's your, your transmitting who's on the receiving end.

It's not someone who has like 30 minutes. Who's trying to nitpick you and see if you have every skill under the earth. It's someone who's looking at you for 10 seconds just to see, Hey, is this person worth calling or, you know, the hiring manager looking at later to just see if they want to interview, right?

So at that level, it's like, they're not going to click into your links that every year sob story, they're not going to like check out your volunteering experience. Like it's just kind of 10 seconds, you know? And and

[00:18:57] Ted Hallum: to your point about the receiver, be audience centered. That's the other big thing. I see someone, you know, puts in the effort.

They try to do one generic resume and then they want to send that out 5,000 times and they wonder what. They're not getting a response it's because you haven't made what, you're the signal you're sending captivating to who you're sending it to. Really, if you're serious about applying to a job and it's the job, you really think that you'd be good at, and that you want, you should take the time to craft a slightly tweaked version of your general template for that employer.

[00:19:30] Nick Singh: Yeah, for sure. Like, for example, when I applied to defense contractors, you know, I have a foreign sounding name, I put U S citizen right there on the top. Cause that's almost one of the first things they say, Hey, this is cleared work. I want them to know right from the get-go. Yeah, yeah, exactly. But I don't do that anymore because I'm interviewing a tech companies that that's not as relevant or important anymore.

Right. So it's just like, yeah. I had a different resume for different industries. Yeah, for sure. Yeah, but I mean, for people who might just like lack some of the experience on their resume, they're a career switcher, or they're just trying to figure out like, Hey, how do I present my skills more?

That's kind of what the next chapter is about, which is portfolio projects. I think the biggest thing I can just tell people about portfolio projects is to do them to completion and then show their complete. What I mean by that is we all have too many projects that like, we've sort of done. They're just sitting on some random repo that haven't been productionized or like some idea that you kind of worked on for a weekend, but didn't really get out there into the world.

And I think, you know, you're paid for results and if you can put your stuff out there, like, Hey, make a public Tableau dashboard and link to it, you know, and that doesn't require rocket science data skills, right. Tableau is just the visualization. And it's free and you can get started with putting a dashboard up there and you just have to like visualize data.

There's no machine learning or deep learning here. This is just like, Hey, can you cut data, find insights from data, poke around a dataset. Kaggle has all these free data sets about anything under the sun. I love my Indian cooking and they have a dataset of just Indian food recipes.

Like they have the most niche stuff. Right. And you could just like, visualize, like, what are the ingredients used in Indian cooking? That's like, that's a whole project that would actually be kind of interesting. Right. Cause you'd be like, huh, what are, you know, what spices are the most common. Right. They have it for sports.

They have it for healthcare insurance spending. Like you want to work in healthcare. Great. There's like a thousand healthcare datasets. You want to work in like GIS. There's all kinds of really interesting like geographic data sets or like satellite imagery, data sets. Right. They're waiting for you to analyze for free.

So I think people should work on these projects and then like put them out there into the world so that they have a link, whether it's a good hub link, a nice repo with a really nice read me, or it's a public Tableau dashboard where you just see your vis or you're trying to show off your sequel. So you write your sequel queries on, get hub and you show what the results were, or it's a medium blog post because you're just trying to show your results.

And like storytell the main key insights. And I'm trying to show them, then you get your code. Or if you are trying to show your nitty gritty crowed at the bottom of it, you know, put your coat up, put your notebook up on, get hub so that people can be like, oh yeah, this guy, Ted, he can actually run coat.

He's actually done something right? Because Our field favors people who do stuff, right. And credentials are great and having a grad degree, great bachelor's degree. But like the reason that bootcamp grads can do stuff for the reason self-taught people can do stuff is because at the end of the day, our field is one where it's illegible that I can actually show off like, Hey, I know SQL, you know what I mean?

Like, I don't know if I could show off, I know surgery. That's why we got med school and all this like licensing. And that's why in their interviews, they don't ask you like, yo perform surgery real quick. They're like, oh, you graduated from these schools. We know you have the skills because you have all these licenses and exams, but our field, because we don't have like a data science license or data science certificate or exam, you know, what we're looking for really is these portfolio projects that you want to build up.

So I think that's one advice that I would give for people, build them and do them to completion so that when you're sending cold emails to recruiters and hiring managers, it's a legible that you have this. So, what

[00:23:29] Ted Hallum: I hear you saying is it's it's evidence, right? So you need to think of recruiters and hiring managers as if they're like a prosecuting attorney who wants to convict you of being a capable data, analysts, data scientists, or machine learning engineer.

You need to provide them with enough evidence to make a a damning case. But in this case,

[00:23:51] Nick Singh: that's a good thing. Yeah, exactly. And in this case degree is great, but like still you can meet computer science. You know, I studied computer science in school. I didn't learn SQL. I actually, I learned CQL through a databases class, but that was optional.

So a lot of my peers at the university of Virginia, we've learned SQL without, sorry. We, we graduated with CS without database skills right now. Of course they could learn. Right. But I'm just trying to say like, Hey, like, you know, if a job requires SQL for business analyst, it's like, Hey, you studied computer science, but you can't even get this business.

Now, this job, because. You don't want a sequel, you know, so it's like, yeah, you can prove these things out today. And I mean, that's the beauty of the internet. You can learn these things for free. The data sets are for free. You can host on GitHub for free. You can host on Tableau for free. It's just, you know, the resources out there.

It's the willingness to do things that's scarce. Right. So if you're actually trying to do these things and you actually build, yeah. People are going to notice because it's that, that's the actual scarce part that people I

[00:24:48] Ted Hallum: talk to when I talk to people in our community, Portfolio projects. It's usually people who are trying to get started on their first or second project.

I get to typically get two questions that are, I think, connected. And they are one, where do I, where can I go to get a data set that's novel because I don't want to be that guy that does the Titanic thing for the 10 million of time. And then secondly, they're trying to just come up with an idea of like, what can I do a project on?

I think a lot of times it's finding a dataset. That's interesting is where your idea is going to come from. And you mentioned Kaggle as being a good source for datasets. What are some other sources that you like, or that you'd point people towards for cool datasets, where they get on?

[00:25:28] Nick Singh: I mean, there's other websites you can Google them around, but I'm telling you, Kaggle is crazy.

Kaggle has crazy amount of datasets. Now what you gotta do is you got to follow your passion slash you got to think intentionally with the dataset you use. Right? So if you're passionate about sports or let's see your passion about basketball, I can't, I don't I'm, you know, I'm not really into sports that.

So I can't even think about what are some interesting data sets in basketball. I don't know where people take a three pointer, like, you know, but like if you're in the sports world, you have a thousand basketball statistics you'd want to calculate, or you want to know about the golden state warriors and this and that like questions that I don't even think about it.

Cause I'm like, ah, I could, I don't know if that matters, like where you took a three pointer from and key. I know enough to know that that does matter. And people like make really cool this around, like where people do it, three pointers from. But anyways, the point is Kaggle has all kinds of data sets.

So I wouldn't even say that the issue is like, you're not looking at Kaggle, like the Capitol didn't have it. I think that people are just not being creative enough. And I think what you should do is you should follow your passion, right? Because passion makes you actually think a little bit harder about the type of work you're doing.

So I'll give you an example of a portfolio project that helped me get to Google and Facebook that I didn't call it. It was called wrap stock.io. So I love my hip hop music. Okay. And, you know, I grew up and around me, people are into fantasy football and fantasy basketball. They have these fantasy teams, the drafting, their players on too.

I was like, why does this not exist for music? Can I make a fantasy hip hop music label? Right. Cause in the hip hop community, we're always yelling like, oh, this person's the goat greatest of all time. Or, oh, this, this artists, they're a one hit wonder, they're going to, you know, little NASDAQ's, he's just going to fade away.

But Drake, you know, he's a, you know, Connie, you know, he's gonna keep going up and up. And the other people are like, oh little Wayne he's washed up. Like he was good 10 years ago, but we don't want to listen to him anymore. You know? So I made basically a stock market for rappers where I use data from Spotify to kind of price these artists in real time.

To show, Hey, who's doing well. Who's not doing well, quantify that. And they'll let you bet on that. Just like how fantasy football is kind of betting on, who's going to do well in the real world and sports. I made the same thing. Now you might be thinking like Nick, what a cool, crazy, interesting idea. But here's the thing.

I was a DJ. I used to be a DJ in high school and I still DJ kind of now on the side. So for me, betting on taste and like arguing about people about music and who's better is something that was core to me and my personality. I would argue all day about different artists. And I knew about Spotify and I was like, huh, just Spotify have a date API.

Oh great. They do. Hmm. You know, and I don't really care about the fantasy football, but like when I was in college, all my like suite mates and college mates, they would have their own little league. And, you know, I joined every now and then, but I was like, eh, I don't really watch the NFL. So it doesn't, you know, I don't really care about this, but I, you know what I do watch, I do.

Different art artists, releasing music and seeing like, oh yeah, this one, flopped, this album didn't. So I'm just trying to give you the story of like how the passion led me to do the project. It was creative. People resonated with it and they're like, whoa, this kid's kind of cool. They like took initiative.

They did a bunch of stuff. And I learned a bunch from project. Right. And I'm not trying to say everyone has to do something as novel as that, but you mentioned, you know, on Kaggle, the F one of the first data sets you see is like Titanic data set of like, who will survive the Titanic sinking. What are the attributes that kind of predict that you'll survive.

It's like, who's actually passionate about the Titanic. Nobody. Right. But I'm trying to say like, Hey, like, yeah. Don't, you know, don't tell me that, you know, capital doesn't have data sets on basketball. Right. And then also don't tell me if you actually like basketball, don't tell me you don't have 700 ideas when you look at that dataset.

You know? Cause I love my Indian cooking. And let me give you some ideas right there on the Indian cooking recipe. What are the most common immigrant. In Indian food too. Can you look at north Indian or south Indian cooking? And is there some interesting differences right there? Why do I know it because I'm passionate about Indian food.

I'm Indian. So I can, I know that there's a split between north and south India, right? I know that there's interesting ingredients. Now, let me give you an NLP project. Cause these are more like database. So this is a little bit more advanced natural language processing. Can I make my own Indian recipe given five ingredients?

Sorry, could I make my own freeform Indian recipe? Right? Could I generate some text based and trained it on the Corpus of Indian food recipes? Could I make my own recipe? You know, so much of what kind of GPT three is doing? Can I make a mini version of that? Here's another one. That's more prediction. If I give you four ingredients, what's the most likely fifth ingredient that's missing.

Hey, if I see ginger garlic, turmeric, you know, you're probably missing tomato. Cause I always see that maybe. Great flavor profiles, flavor profiles. Like there's like, you know, there's a crazy amount of stuff or, Hey, what percentage of recipes are vegetarian? I know a lot of people in India are vegetarian, right?

So I'm just trying to say like, okay, you're into barbecue. You go there. I'm certain there's like, Hey, what percentage of food is barbecued? Or, you know, there's, and this is just, we're just talking about food, right? So if it's, if you're a person and you have some interest in sports, music, healthcare, retail, analytics, it like, there's so many data sets that I, I just want to push back on, like, Hey, people need to think for themselves a little bit more and there's, there's crazy datasets.

And then let's say, you're not cool. And all you care about is getting a job, but you have a dream industry. You want to work in, you want to work in geospatial intelligence. You want to work in healthcare. You want to work in marketing analytics. There's tons of data sets about like marketing data and like, how can you analyze it to like drive campaign, spend better?

Cause that's a real industry. And if people have those data sets out there, right? So even if you don't want to do your hobby, I guarantee you some average show is not like looking at. Marketing campaign data. But if you analyze that, visualize it well, and then cold email that to a whole bunch of marketing analytics jobs and hiring managers, people are going to love it.

And that's the next chapter cold email. So you see how it's kind of working one way to another and it's like, I wish I wasn't so salesmen. You're like, oh, this leads to the other thing, but it's just, this is just the natural progression. You've done all this work for a ma you know, marketing analytics or for healthcare go email them, your project.

[00:31:55] Ted Hallum: Absolutely one thought I had real quick before we move on to cool contacting is you, you talked about the importance of having a passion about the dataset and a project. And one thing I think is important to point out because you also see, this is a strong corollary with people when, for the actual jobs that they're going to pursue.

Is that passion drives domain knowledge. Yeah. That is why, you know, you would Excel at doing a project about Indian food and cooking more so than you would sports because your passion has already driven you to know so much more about Indian food than you know about sports. And so naturally that domain knowledge is going to, you're going to know a golden insight when you see it.

And you're going to know how to seek up a golden insight out of that data in a way that someone that doesn't care about Indian food when they wouldn't even know where to start. And likewise, you know, once you do that project that helps you land the job. If you do land a job in some domain that you're not passionate about, that you don't have any domain knowledge for, you're gonna, it's gonna be an uphill battle in a way that it wouldn't be.

If you got a job working with something that you're just naturally into and that you go and learn about on your own time,

[00:33:05] Nick Singh: right? No, a hundred percent. And I think partly it's like people think that. Oh, I'm not very creative. Nick, come up with an idea for me and to sort of like, you are creative in your own passion, right?

Like I'm, I'm a creative cook or when it comes to music and production and DJ I'm creative, but it comes to thinking about sports strategy. I'm not creative because I don't really care, you know, but I think of myself as creative person, if I cared, I probably could come up with some like really interesting draft picks for my fantasy football or basketball team, but I, I could care less, you know?

But when it comes to music, yeah. I have some creative, you know, I make music, right. Like I'm creative there and that's why I have some insights there on like, Hey, like, you know, like if I know how to make music and I'm creative there, you know, maybe there's some way to quantify other artists that are creative, you know, can I find that through Spotify data?

Yeah. So exactly. You're right. Create, you know, a good project is creative and creative comes from domain experience and it doesn't have to be all nerdy, like domain experience in like retail analytics. Your domain experience could be just like.

And you're good to go. Yeah, exactly. So now you've done all this work. We've got to show people that you've done this work, you know, you got to get in front of their face. You gotta let them know. Cause you know, man applying online is a real mess. You're one with 300 other people when you're applying on LinkedIn or indeed.

And you're just another resume that honestly just gets filtered out because they don't sit there and look at 300 resumes, you know, it just never happens. So, so it's not

[00:34:39] Ted Hallum: like field of dreams. You don't just build it and they

[00:34:42] Nick Singh: will come. That's exactly, exactly. And you know, you get lucky and like, you know what, if you mentioned a lot of good keywords, you can get interviews.

It can happen, but I'm just saying, just take the extra 10 minutes to go see who's the hiring manager on LinkedIn, or try to guess, or Hey, if it's a small company, go reach out to the CEO, tell your story, you know, cover letters are dead, but telling your story is. So a good cover letter should tell you a story of like, Hey, why I'm perfect for this role, but I'm not going to open up a word doc and like read this like one page sob story about like, what happened?

And you're like, no, just send me a Chris' email of like, Hey, I'm me. Here's what I do. I built this really cool project that you might be interested in. I saw you had an open position for something that's very similar, you know, are you free to talk about it? That's it that's like five, six sentences. Right?

And you showed that you had real skills. Cause you mentioned you have this degree and this degree or this skill and this skill, but you also linked to your project, like with a real hyperlink. And if they click it, boom, they see this really cool to blow dashboard or they see your get hub with a really nice visual graphic.

And you know, sometimes what I do is I would screenshot a dashboard. So it's like, Hey, I'll give you an example of someone has a helping. They looked at Instacart grocery orders. Hey, what are people ordering from the grocery store? There were 300,000 transactions with like 3 million items bought. So, Hey, you know, what's being bought at grocery stores.

What are the most common things that are bought? What are the most common brands? So, one thing they did was they looked at the cookies. What are the most popular selling cookies? Are they Oreos? Are they Milano's? Are they chocolate chip? They took it. They visualize that this is like a very simple, like, you know, group by count.

You know, this is not some advanced machine learning. It's just like, Hey, just grouping all the cookies together. By the brand name, they put that in their dashboard, they took a screenshot of their whole dashboard. They set it there it's like clickbait. Like who's not going to click like, Hey, by the way, here's my top cookies.

And if you're curious about who's winning the Coke versus Pepsi war, here's the link to my dashboard that visualizes that too. And now everyone, everyone has opinion on copers Pepsi. So they click the link and they're like, oh cool. Yeah, Coke's winning. But you don't diet Pepsi does. Alright, and this is not rocket science AIML.

And this is like this person that was helping, you know, she's interested in cooking and like likes junk food. She's not the healthiest person, but she'll own up to that, you know? And that, you know, but like most people aren't and like most people are curious like, Hey, is lays better than Doritos. Like, I'm curious about that.

Right? Like, I don't even know the answer to that. So I'm just saying you put that kind of information in a email and you send it off to a hiring manager. You're getting their attention. You're showing like, oh, you actually re wrote SQL queries to query this data. You visualized it. And you want to go above and beyond because let's say you're trying to do a data sensor, ML, chop let's do some prediction work because this IRA, this data set originally came from Cabo where they're trying to predict what items are most likely to be reordered for.

They're kind of like recommendations thing within the app where it's like, oh, you ordered seven things. Are you sure you don't want. Bring up the milk from last time, you know? So that's the prediction work you can do. So I'm just trying to say this stuff. Doesn't have to be so basic of like, oh, it's just data is, but like I'm trying to do be a machine learning engineer.

Like what do I do with this? No, there's levels to it. But point is you put it in an email, you get in front of the hiring manager and you show your story. Even if you don't have every skill, even show that you can do this work, you know, you're up in your rate of interviews and it's not a foolproof method.

Right. But on LinkedIn, indeed, you're one in 200 applicants here. Maybe you're one in 10 to actually write an email. And they like you as a person, they see you're a veteran. They say you have skills or they see, Hey, you don't match every checkbox, but you actually care enough to like, not spray and pray your resume, but like you've actually built something kind of relevant to my company.

Yeah. I want to talk to you and that's why this works so

[00:38:48] Ted Hallum: well. The thing that I think is key about first building a project and then leveraging that project. And in your cold contacting effort is the whole hiring process is, is a triage effort, right? Hiring managers only have so much time. They're usually managing a team, that's actually building something.

So they usually come up with a list of kind of criteria that just says, if someone has all this stuff that is required and then this other set of stuff, that's desired, then it is virtually certain that they'll be able to do the job. And then they hand it off to a recruiter who may be slightly technical or may not be technical at all.

And then they go out looking for people on LinkedIn and the company's resume database that tick off all those boxes. Okay. Just because you may not tick every one of those boxes doesn't mean you can't do the job. Exactly. And so it's this cold contacting effort to help you bridge that gap, maybe between what you can actually do and what the recruiter thinks that they have to the boxes they

[00:39:53] Nick Singh: have to tick off.

And I want to mention like, you sh the cold contacting works because you're showing them, you can do the job. And that's where it's like, none of this will work if you don't have the portfolio of projects to back it up. But if you do, it's like, Hey, I see it's Ted homes get home and it's Tableau dashboard.

And he told me, Ted, you were really interested in the military stuff. You looked at military spending by country over the years and you saw how it spiked during world war two and this and that. And you can see this and that. Oh, wow. This is definitely Ted. He cares a lot about this stuff. That's really cool.

I want to talk to him, you know, and that's the beauty of it where it's like, oh, I like Ted as a person. His initiative, his passion and Hey, we're a defense contractor. This is actually kind of interesting to us. You know, something like that present.

[00:40:39] Ted Hallum: I actually have a personal story I can share. So this, this actually worked for me.

If you're listening and you're skeptical, I can at least tell you it worked for me. So in, in my current role, actually, so at my previous job, I was working along and one day I'd get a message on LinkedIn from a third-party recruiter who was trying to recruit for the position that I now have. And she said, Hey, you know, I briefly talked about how it's only like six seconds that they look at your LinkedIn profile or resume.

Just like I briefly looked at your profile and you look like you might be a good fit. Let me know if you're interested. And she sent the link to the job. So I went and I looked at the job and it did sound really interesting. And I thought, wow, I would be great for this role. But a couple of things they specifically said they were looking for five years of experience.

And at the time I had completed my master's degree and I had like two years of work experience in those areas, but not five. So I messaged the recruiter back and I said, I'm super interested. I think that I'd be a great fit for this role, but I do want to be transparent. I looked at the requirements for the job that you sent over.

And there are a couple of areas where I don't have as much experience as it says is required. But I think I can do it if you'd still consider me at which point she promptly ghosted me, never heard from her again, but she served the purpose of letting me know that the company and the job existed and that I'd be a good fit for it.

So I went with that little tip and I went out to the company's profile page on LinkedIn, and then I went to their employees and I scroll through all the employees looking for. The people that might be the hiring manager, or that would be a logical hiring manager for that position. I think I found maybe three people.

This sounded like they could be the hiring manager for that role. And I sent each of them a connection request with a note, very important. You know, don't just send a connection request where they have no idea who you are or what the logical reason for the connection request is. Included a note saying that, you know, a little bit about my background and that I thought their company was interesting.

And, you know, if they ever had any roles along such and such lines, please reach out the next day I was contacted by the hiring manager for the position I was interested in. And he's like, thanks so much for reaching out. I looked over your profile. You actually like, you'd be a great fit. Could you be available for a zoom call?

And so we did a couple of zoom calls. I had an opportunity to tell him more about my background, had an opportunity to show some work that I had done and. You know, after a few conversations, it was the job's yours, if you want it. And so, you know, if I hadn't done that, I would have been ruled out by someone who wasn't qualified to rule me out.

Exactly. As they're, they're just looking for five years of experience. Cause that's what the paper says and I had to, but the reality was I had made good use of my two years and the way I'd invested my two years actually made me qualified for this

[00:43:42] Nick Singh: a hundred percent man, a hundred percent. And I'll be, I'll be honest.

Like I saw you had a master's in business analytics, right? That's right. And now you're a machine learning engineer, right? That's right. If I'm looking at you for two seconds, as a recruiter, I'm thinking, Hey, machine learning engineering, you need a master's in computer science or masters in data science, business analytics, go, go be a business analyst or a data analyst, or like maybe a data scientist at best.

You're not an engineer. We need an engineer. Right? Like I'm looking at that, oh, you have one year experience, blah, blah, blah. But that's the beauty of it, but you've done the work. And I remember, and I'm just saying this, because I remember we had talked about the classes you did in your master's in business analytics.

I'm like, whoa, that's some really advanced coursework. And then since then you've done more things. So the, you know, that's the beauty of it, right? Like if I have five seconds, I'm just like looking at 200 other profiles, I'll just move on past you. And that's the, that's the case for most people. Like, you know, cause no one's a perfect candidate, no one meets all the requirements ever.

And if there were so obvious, they would already be employed in a different job, you know? And that's the beauty of it. Right. But recruiter can't see that, but a hiring manager can see that cause they can actually be like, oh wow, your tech stack actually kind of similar to what I needed. Or like, whoa. Even if you're tech sexually different, right.

A recruiter might be like, oh, this person's tech stack is totally different, but once you show them the project, they're like, whoa, the end result is exactly what we need. Really great machine learning model, and you put it in production and show. You did all kinds of different tools that we don't use at our company, but the end result is you put up machine learning model.

That's really performance out into production and it scales, well, it doesn't matter what tech stack you used, but a higher manager has that kind of clarity that a recruiter might not have. And that's the beauty of this whole tactic.

[00:45:30] Ted Hallum: Exactly. Well, and to bring a few ideas together here, come from when we first started talking to now, we were talking earlier about making sure that you communicate clearly that the most important things about you relevant to the position, get conveyed to the person on the receiving end.

You were just talking about putting things into production, and then you mentioned you know, I have a masters in business analytics, but I'm doing a machine learning engineering role. But I happen to have done an MSBA degree that. Maybe more technical than your average MSBA degree. Yeah. So the point is, you know, maybe you are trying to do a job where you put models in production, and maybe you do need to know a lot about infrastructure for the job that you're applying for.

If you did a data science boot camp, but that data science boot camp has a really strong emphasis on Kubernetes and nodes and clusters and deploying stuff. That way that needs to be like, what'd you say at the top lift of your resume, exactly. Make sure that you're highlighting your skills and abilities.

That might not be self-evident otherwise, because when I see data science bootcamp, I don't necessarily think Kubernetes, but I'm sure that's out there somewhere, right? The same way that I did an MSBA degree that had deep learning neural networks and, you know, things that people might not necessarily associate with

[00:46:47] Nick Singh: business analytics.

Exactly. And you listed that coursework there. In the order of like, oh, the first course you took in business analytics is deep learning. I'm like, what the hell? Like what, you know, I don't know. That's, you know, but I mean, and that's the beauty, right? Cause you know, and if you were actually going for a business analytics shop, maybe the first thing you write would be like financial analytics and the next one would be like managerial accounting.

And the third thing you'd be as like, Hey, like data is and like dashboarding, right. Then it's like, oh, wow. I mean, you're not going to be building models in a business analytics role. You are going to need to know some more MBA type skills and like, you know, so it's it's yeah, exactly. And that's where it's like, Hey, customizing the resume for what the job you want.

And like highlighting the skills that are most relevant way. Because the second I see, oh, masters in business analytics, I'm like, do they know machine learning engineering? And then I see your coursework and I'm like, oh, okay, great. This is, this is crazier than like, you know, this is like a CS degree, basically.

It just under a different name, you know, something like that. So,

[00:47:46] Ted Hallum: and that's kind of another, I think that's like the effect of us still, even though. In some ways this fields kind of come into its own and other ways we're still kind of in the wild west, especially like in the, I think in the education space.

So you look at a certain degree and it's not really, you can't know what that degree entails just by the name. You really you really need to know the coursework. And that's why, you know, you mentioned that I tried to list my coursework out on LinkedIn and put the most relevant stuff at the top. I would recommend everybody do that.

Because otherwise I think recruiters and hiring managers know that just seeing, you know, an MSBA degree or a master's of science and data science, or you mentioned that you have a background in computer science, these things are so from one school to the next, the contents of a program, they're not, it's not codified.

It can be radically different.

[00:48:39] Nick Singh: You can pass a program without building real projects, right? Like, like, I, I. Like all that stuff. I was telling you about rap stock dyo with a hip hop stock market. Like that was all like, just on the side that I'm like learning how to do web dev and put stuff up, but I could have graduated without doing any of that.

And like, not knowing SQL for example, you know? So no,

[00:49:00] Ted Hallum: I think the next one is behavioral interviews. I know a lot of people find these to be really scary. So yeah, I think it

[00:49:06] Nick Singh: shouldn't, it shouldn't be scary, man. It should just be like, they're just trying to learn more about the projects you've done and why you're a good fit for the role.

And usually the first question is, tell me about yourself, right? So I just want people to like practice this pitch because it's so important. It's the first thing they're asked and people will botch it, really smart people botch this all the time. And that's why the book exists because I didn't know people botch this.

Right. Because I'm great at talking about myself. You know, I'm in love with myself, narcissists very easy for me. Okay. You know, but for other people it's like. You're so awesome. You're not even talking like the way you just pitched yourself. I'm like, I know you're more awesome. Like I want to pitch you for you.

Like, you know, and it's like, people are really cool and people have really interesting life experiences. What they don't do is transmit that well, because you know, it's a little weird to talk about yourself in 90 seconds and like a sound bite. But I think what people should just do is when they're asked about, tell me about yourself, which is like the number one question.

The first question they're asked, sets the tone for the whole interview. It's, you know, talk about the past present future where the future is exactly the job you're interviewing for. Okay. Meaning, oh, the future is, I want to be a data analyst at a healthcare company. Oh, wait a second. That's exactly who I'm talking to right now, you know, or maybe the future is for you, Ted.

I want to be a machine learning engineer in a defense industry. Right? So what's the present for 10 present for you is. I did a master's in business analytics, but my favorite part was actually the engineering and ML part like database. That's all cool, but I did a lot of coursework and I really loved it.

And I even did the side project. And then what's the past for Ted? Hey, I served in the military. I did atmospheric sciences. You know, I have some background in science and math and technical stuff with military experience with real like operational work experience. So how you put it together? Hey, I started here where it kind of like data and math and got some real world experience.

And I liked the defense industry and I want to serve my country at present. I did my masters in business analytics. It was awesome, but my favorite part was a deep learning machine, learning the engineering side and my project. This was the best project. And that's why I'm looking for positions in the defense industry to combine.

With my new interest in productionizing machine learning models and maintaining that side of the data science world. It's like, whoa, this guy was born for it. Where are we going to find a vet that actually has experienced productionizing machine learning models, right?

[00:51:49] Ted Hallum: It's genius. You craft a narrative where you're the logical conclusion.

[00:51:54] Nick Singh: You're the logical conclusion. Exactly. And the thing is I left out lots of random stories about this and that. And I left out the fact like you have all this experience with like managerial accounting from your business degree. Like I skipped over all of that, right. So I have that same story for multiple jobs.

If I'm trying to be an evangelist, I say, Hey, I did the data stuff. Then I wrote the book. And my favorite part was going on podcasts, talking to people like Ted that's want to be an evangelist for your developer tool, but let's say I wanted to be a product data scientist, someone who's helping build better products.

I might gloss over the fact that I wrote the book. I'll just tell him like, Hey, at Facebook, I was helping make. And, you know, I wrote a book, let's pretend I didn't Plosser the book. And I wrote a book. It was great. But what I missed was actually going back to build real technical products, not informational products, but like real products.

That's why I'm here because I realized my passion really is with the work I did in the past. And that's why I'm here talking to you and sure, at a misstep where I learned a lot about marketing and positioning a product and getting product to market, but I really love technical products. And that's what will help your company do that?

You know? So did you know or let's say, I told you about machine learning engineering, right, Ted, Hey, you know, I studied CS, I got interested in data science work. I did a bunch of random things, but I want to be more technical and go back to my computer science, engineering roots, rather than data science or data, this type stuff, or marketing.

Like I want to go back in the field, build things and build things. Build pipelines and models, things I'm more intimately familiar with as a CS major than, you know, some of the database stuff where it's too interdisciplinary for me, you know, and that's a lie. Like I love the interdisciplinary aspect, but if I'm going to tell that lie, I'm gonna tell like, yeah, yeah, no, that was all missteps.

Like I just, I'm a true engineer. That was all a mistake. That's why I'm here. You know? And of course my story is not so convincing. Right? Cause it's like, Hey, you know, this, this author really want to be a machine learning engineer. No, but that's just, my background is really crazy, but you can just see how even then still, I can kind of tell you a story for all kinds of different jobs.

Right? I look at different industries,

[00:54:05] Ted Hallum: this approach that you're kind of laying out as a template that people can follow, because I think it's intuitive. People know when they're going into a behavioral interview, basically they know they need to sound interesting and compelling, but then in an attempt to do that, it's so easy to lose sight of the forest for the trees.

And they, you know, th they start talking about some random story, like you mentioned, and that all of us. And the goal of trying to sound interesting. They don't, or they don't sound relevant. By following this template that you're talking about, where you talk about your past, your present, your future, lay it all out as a narrative where you are the logical conclusion, you are the solution for their need.

You can't leave that interview without sounding interesting. So I think that's a perfect

[00:54:49] Nick Singh: strategy. Exactly. And that's, that's exactly it. That's why everyone's got to check out the books, shameless plug, but that's literally the chapter on behavioral interviews because it's, so it sounds intuitive right now and we give more examples of how to do that in the book and like more common questions that are like that go beyond this.

But you know, leaving y'all with a nugget, like this is how to go about, tell me about yourself. Cause that's one of the first things people do. And one of the most things to get. Right.

[00:55:13] Ted Hallum: Well, real quick, before we move on, you mentioned examples in the book. We were talking about cold contacting a second ago, and if I remember, I think there is an example of like, like a literal email that people could look at for that.

Right.

[00:55:25] Nick Singh: Okay. I put in some of the exact real emails I sent to get me interviews at Airbnb, Uber, and CloudFlare. And I put the exact email that got me my last job, which I got through a cold email to the CEO. So Ted, you're not the only one who had success with cold emailing. Like I, I did it too. That's exactly how I got my job, where I wrote to the CEO and told him, Hey, I'm perfect for the role within a day, they got back to me.

It's like, Hey, cool. I love your initiative. Like, yeah, this does seem pretty interesting. And it was a role. I was totally like, not qualified for that. I wouldn't have, I would have been filtered out. Like I was not a good fit for on paper, but once I told them the story of like, Hey, I know my resume on LinkedIn.

Doesn't mention it, but I'm a DJ and I have sales and marketing. Oh, and I'm also technical. Cause I worked at Facebook. You can see that, but you don't know that I was a DJ and like more three-dimensional than just like, oh, and I ran a startup, which you know, on my LinkedIn is not so much. Cause it was more like a project, but I tell them about like how I grew it to 2000 users and I have this like really sharp marketing skill of like, Hey, I've built things and like, know how to position them and get users for them.

So I'm not just like an engineer will just sit there and code whatever someone else tells me to. I think critically about products. That's when I combine these kinds of things where it's like, Hey, you don't see this. You think I'm just saying normal engineer, but really I'm more than that. This is why I'm like a unique person for your role.

Right? And I, this is like six sentences and I linked to my old rap stock that IO and I told him 2000 users and you know, I'm already at Facebook, but one thing you don't know is I'm also DJ was saying, so how'd that sales and marketing background, like from like hustling and going to that to like $8,000 I made in.

From being a DJ. Right. And then that role was this really, because it was an early stage startup. They needed someone with like technical skills, but also some business schools with a lot of initiative. Right. And there's nothing more scrappy than like, yo I ran my own mobile DJ business from thin air. And then I started my own startup from thin air and, you know, so it resonated there.

So I'm just trying to say like, basically, if you're think about the jobs you interview for intelligently, you build your projects intelligently for those kinds of jobs, and then you reach out to the right people. It all connects, you know, and of course, when I built my hip hop stock market, I didn't know.

I later on to a, more of like a marketing for a data science company role, you know, that's what I did at safe craft. I didn't think I would do that. I thought I'd just stay technical forever, but it just kind of lined up. And when I was doing deejaying, I didn't think that was my start of my business career.

I was just deejaying. Cause it was fun. You know, that's the thing you just got to build more projects. And if you can be intelligent about what you're building. To make it line up with things. That's great. But even if you don't, what's awesome about life is things work out or things it's, it's, it's hard to connect the dots going forward, but once you've lived it it's like, whoa, tech, like you worked in the military, you did business analytics, and now you're here.

It seems so logical that you would end up here. Cause you were like a deep inside. You're more of a CS engineer type person. That's, you know, just done a whole bunch of other stuff, but got you to that position. Right. But you might not have known that when you're studying atmospheric sciences, that one day you'd be a machine learning engineer, but

[00:58:39] Ted Hallum: momentum has a way of continuing to propagate itself in a certain direction.

Right. And the, and the further you go in a direction and the more speed you pick up, the more likely you are to continue on that trajectory. One last thing I would mention just from all the conversations I've had with different members of the veterans and data science and machine learning community, I think maybe the one hangup that that would give someone pause.

Or that I've heard people say it gives them pause about doing the cold contacting type thing. Is they they just feel like maybe it's aggressive or like they're being salesy, but I think that's the wrong I would just add. I think that's the wrong perspective, because again, let's go back and restate.

These are very busy people. And so if you can effectively communicate talking about communication again, that you're in the ballpark for what they need, when they receive that message from you. They're not going to think, oh, this super aggressive person trying to throw themselves on me. When they look at what you're communicating to them, it's a relief or at least potentially a rally because they're, they're having to expend a lot of effort trying to find the right candidate.

And if you can make the case to them effectively, that. Or can be what they need. Then the solution is just fallen in their lap. You've taken that burden off them of having to go sift through the sea of candidates to find the right one. And you've just said, here I am. And so you're you, you've kind of,

[01:00:09] Nick Singh: do

you make their life easier as long as you're relevant, which is why we say like, this is not going to work and you're totally like, not in the ballpark and you don't have any projects and you're just like bothering people, that's bothering people.

But if you actually can tell the story in about six sentences with a real link to real work, and you looked something up about that person on LinkedIn to make it just slightly personal, like, Hey, I know you're the manager for the demand forecasting team. I used to do supply chain forecasting for the military and check out this really cool project I did.

Can we talk, boom. That's it let's four sentences, right? Exactly. So as

[01:00:48] Ted Hallum: far as the technical chapters of the book, that the deal with the technical interview, you know, all the skills and all the technical chapters are good, but for somebody who's just coming out of the military, you know, maybe they were a, an E four or a Sergeant or whatever the case may be.

And they're looking for their first entry level role. They just need to break out and get that, that first position, because we all know the first position is the most difficult one to get of all of those technical chapters, which is the most critical for just landing that breakout opportunity.

[01:01:21] Nick Singh: Oh man.

That's, that's really hard. Right? Cause everything works in concert. I think it's really hard. But I think the one, I guess I, the, the chapters don't really map like that where it's like, oh, this is like the easiest skill to have. Right. Cause it's like, if you, if let's say you consider SQL really. You still need other things to kind of combine with it.

But I would probably say SQL SQL is probably the first way, like first wire first place where it's like, oh, this is something that not everybody knows. That's kind of like, yeah, that makes

[01:01:51] Ted Hallum: sense. After go do stuff with data, you need to be able to get data CQL is kind of like the most common language to retrieve data with.

[01:01:59] Nick Singh: Yeah, exactly. And like, Hey, even if you don't know any modeling or statistics or anything, because as long as you can get the data, that's all you're already in business in the sense of like, Hey, well, I can just do some very, some very simple, like, Hey, what's the most popular this, right. That's a SQL query, you know, and that's not rocket science right there to learn.

So I think SQL would be a great place to start. But I think, I mean, we don't have a chapter on Excel, but I think just like getting good at Excel, like getting really good at Excel is probably the first thing I would advise people to do and then layer it on with. But yeah, we don't have like an Excel chapter because you know, our book is called EISA data science interview.

And for data science, they're not really often testing or you know, Excel capabilities. That's more for data analysts. And even then it's usually like applied to a problem or something like, it's not just like, Hey, tell me how to make a pivot table. It's usually like, Hey, we're giving this problem. How would you find the top 10 this?

And it's like, oh, I make a pivot table. I'll do it like this or this

[01:03:04] Ted Hallum: now w w quick, random thought earlier, we, you mentioned don't mention random jobs on your resume, unless all you have is random jobs, but what you also just mentioned, Microsoft Excel. And what I would say is if you're looking for your first breakout data related opportunity, it's hard for me to imagine any job out there, except for, you know, maybe just like, if you're in the trades or something where you're.

Touched Microsoft Excel, virtually everybody uses Microsoft Excel. Microsoft Excel is a, you know, at least the basic level, a data tool. So look back, think back, think hard to, you know, maybe you were in that job for two or three years. Well, what were the one or two experiences where you did have to use a spreadsheet?

What did you do with it? Did you, you know, create a pivot table? I think you mentioned at the tables earlier and if so, like focus on that, make it sound like, you know, leave the reader with like your two years of that job. You were in Excel building pivot tables. Don't I wouldn't say lie, but definitely emphasize the most important parts and the most relevant parts of those job experiences.

[01:04:12] Nick Singh: Most of these jobs in the field, their data quality issues talk about like that. Right. And that could be a story there. Right, right in there in that it's like, Hey, I was dealing with data quality issues and. Right. That's why I realized, Hey, we need data structure and like databases and like better ways to structure data or ways to like take unstructured texts and structure it.

Oh, wow. That's why I like data science. You know, it's like, that's your story in a really convincing story. Right. So there's always, you know, if you're interested in this field, there's gotta be some kind of spark already, right? Like, so just have to dive into it and see what it is. And like, let's say you don't have that stuff.

Do those projects to make up for it, you know, list your projects first, make those projects meaty. You know, you have a job that's irrelevant on the weekends or nights be working on those portfolio projects. That's, that's the only way. Right?

[01:05:05] Ted Hallum: So you've mentioned PR probably the most, in your opinion, the most important for getting your first job technical chapter might be the one on SQL.

So that's getting the data and I think one of the most important things you can do. Quickly driving insights is visualizing data. So I think that those two kind of go hand in hand dessert D is the book touch on

[01:05:28] Nick Singh: Vietnam? We don't really talk about visualization. Something we might want to ask later, but in interviews it's one of those visualization is all about show, right?

Like, I mean, it's literally there, right? So that occasion you're right. So it's sort of like, Hey, we don't have a chapter on that. Cause it's like, Hey, your job needs data is they'll ask you, where's your dashboard. You know, they're not going to ask you, like, how would you visualize this that often? Yeah,

[01:05:53] Ted Hallum: sure.

So I think my next logical question we talked about, you know, the most fundamental, critically important chapter of the technical chapters, which one do you think is the most difficult one to master?

[01:06:07] Nick Singh: Yeah, that's a good question. I think I think the, the product sense chapter can be kind of. Because it's a very nebulous scale that really takes time from real-world experience to get better at like, the book definitely helps you improve, but it's sort of like, Hey, how do you build product and business sense, unless you're building products or working in a business, it's hard to develop those skills.

Right? So in a book we do our best and it definitely improves your baseline level, but there's just something of like, Hey, at some level you, unless you've been in the trenches, you can't answer these. Right. So it's like, even if I don't have a really intense SQL job, I can learn the most advanced sequel right now for free online.

That's more advanced than what a job will ask me. But on product sense, it's sort of like, Hey, unless you're like in the trenches doing those kinds of things. So I think that's definitely something to do, but, you know, I wouldn't let it people, I wouldn't let it bother people because that's one of those that come through just real-world building things in teams to see, Hey, What does it take to actually build a good business or go good product?

And how does data fit into it? That just comes from rural we're working. I think another thing people could do to just like Uplevel themselves is just to read more about businesses, read information, read these technical blogs from companies like stitch fix and Facebook and Uber who talk about the real world problems.

They're solving with data, how they're fighting fraud, how they're optimizing the Uber search pricing algorithm, how stitch fix recommends close to people, how do their recommender systems works? How does Airbnb collect marketing data and analyze that marketing and data it's all there. And I think what reading those technical blogs is really good for is because they lay out not just the technical details, but like the business problem.

Like, Hey, we're Airbnb. We had this big problem at our company because you'll find that whatever problems these companies have your company, probably when it hits a certain scale has the same problem. Right? So. Uber wrote about how they use like graph based machine learning models to fight fraud. It's like, Hey, your company probably also has a fraud element to it too.

If you're working in finance or consumer technology, they're just going to be bad actors. There's going to be fraud. So it's sort of like, how do you build this repertoire of like, Hey, how do people generally fight fraud? That's not something you're going to read in a textbook. It's something you're going to see in technical blogs.

And that's where, when you go for a risk analyst position or you join like a machine learning team, that's like credit card fraud. It's like by reading more about fraud in general from other companies, that's where you start to show off like, Hey, I know something more about the business and the science of fraud than just, Hey, I have these specific technical skills, but I don't have any domain experience.

[01:09:05] Ted Hallum: Yeah, absolutely. And we, and that kind of goes back to the earlier, when we were talking about projects, Passion drives domain experience and domain experience is what allows you to actually do a meaningful project or build a meaningful mafia or whatever the case may be. Yeah. So, you know, of course I used to be a veteran and I transitioned towards the world of data and we talked about momentum and how momentum puts you on a certain trajectory, or you can think about it kind of a different way.

It'd be like a gravitational field. If it's, if you're thinking you want to escape, it can make escaping kind of a challenge. Because if you, if you get too far away from where your background is you're maybe issuing some of your most beneficial assets that could be along for the ride. But I've often wondered.

So since I went into the, into the data sphere, all of my roles have been for companies that support the federal government and the department of defense. It was just, you know, very natural for me. Yeah. You can take advantage. I could take advantage of all my previous skills and experiences and abilities by going this route.

But I have wondered, you know, from the world that I'm in, where I'm doing data, how does that compare and contrast the actual job and the interview process and everything to like big tech and where a lot of people, I think where people mostly think of data science and machine learning. And I want to talk to you about this because you have a really interesting background and that I believe earlier, you said that you'd worked at both Google and Facebook, right?

Yeah. And I think I saw in the book where you had interned at Microsoft, so you have plenty of exposure to big tech, but then you also had a role where you did data science for a prominent, or at least a well-respected firm that does data science in support to the federal government. Yeah. So you could speak from both sides of that coin.

Yeah. And I think a lot of our listeners would be really interested to hear kind of your perspective on that.

[01:11:06] Nick Singh: Yeah. I mean you know exactly how you said it like led into your background to be in this industry of defense and everything. It's a good industry, good work-life balance, high pay, good job security, a good mission that you can usually talk about or not, sorry, not talk about, but more like you can point to like, you know, you know, what's happening with the work you're doing.

Yeah, working in big tech, you know, there are some like, you know, you know, it's not the most popular thing in the media. You know, he gets in trouble, sometimes pays good work-life balance can be hit or miss. The types of challenges are definitely different because you're working often at a really big scale, which is pretty interesting with some cutting edge tools, but then also sort of like, damn it, I'm working on tools that are like specific to my company, versus like, you know, at the defense contracting firm, we actually used a ton of open source tools which we didn't, because at Facebook we had in-house tools for everything, you know, so I didn't get to use, you know, I had to do things the Facebook way.

So this is there's nothing pros and cons, but I mean, I definitely think at school for people to work in defense contracting, all, especially if you're in like in an org, like I grew up here in Northern Virginia, that's like most of what the industry is. And then you have this really good job security, especially you get cleared.

You don't kind of have that kind of system at these tech companies, you know, every time you interview you to go through the conflict again, and it's really tough. And actually that pisses people off because it's like, damn it, like I have so many years experience, but these interviews are killing me versus Hey, in defense it might just be like, oh, well, we see you've worked at Northrop Grumman and you have this right clearance.

You probably can do the job here. You know? So yeah. There's pros and cons. But I definitely think like, Hey, if you're trying to break into the industry, there's nothing wrong with going to government contracting firms. If that's where your experience lends itself best. And like, Hey Ted, now that you've done, that kind of work at a defense contracting firm and you have years of machine learning engineering experience, you can go to another industry.

Does that make sense? Like, it's a good on-ramp and I, I, you know, where it's like, not like, oh, this is all all you can do. I don't think there's anything about that. It's just. Like, Hey, it's, you know, cause sometimes it's really hard, like, cause go, these Facebook has really hard interviews, right. So if you say like, Hey, I'm just going to go directly to that.

You're going to get crushed because I'll get crushed. You know, everyone gets crushed. Like I, I wrote the book cause I got crushed now. Cause I'm so good. I wrote the book because Hey, I was good enough to get these kind of jobs, but still that came from a lot of failure and a lot of other companies where I was like, huh, how do we interview better?

Right. If I interviewed perfectly, I would have not written this book. Right. I'm not writing a book on how to eat Indian food. I'm a pro at that. Like I don't even think about it. I just eat my Indian food, but I don't re you know, but I wrote this book because I was like, ah, dammit, like how do we actually help people do interview prep?

Like I used to struggle with this too, but I figured it out eventually to work at some of these companies, but I used to struggle too. So I would just say like, Hey, think about it as like an on-ramp like, yeah. So, and you know, at some level it's like, Hey, take whatever jobs the best you get, you know? It's the best job you can get as in health care, go to healthcare.

Right? Because

[01:14:25] Ted Hallum: now for any of our listeners who maybe that's their goal, you know, they're like, no matter what, it may take me 15 years, but one day I'm going to work for Google or one day I'm going to work for Facebook and I'm not going to stop until I get there. Yeah. You mentioned that, you know, the, the interview process is a gauntlet and it's very difficult and you, I think you used the word crushed.

So what would you say would be some tips for remaining resilient for people that are just hell bent on pursuing that and getting crushed over and over and over? Because I suspect it's, it's not, you get crushed once it's you get crushed many times until you get good enough at that process to land a

[01:15:04] Nick Singh: position.

Exactly. No, that's exactly true. Right? Momentum is key, right? So it's like, Hey, well first let's get the machine learning job. And then getting the industry a care about that, build these skills and then interview. And, you know, it's a, you know, it's just take it one step at a time, because if you just like, yo, how do I interview at Google?

It's like, oh, I don't have the education, this, this, this for it. It's like, well, should it be even worth trying? But like, no, it's, it's, we're trying, it just takes longer. But I think that's why in the mind hack it's like, Hey, you know, people ask me, how do you write the book? And I'm like, or like, it wasn't worth it.

And just sort of like, if I knew how much work it was, I wouldn't have done it. But the point is when I went blindly naively into it, I was like, ah, this is not terrible. And then I kept going and going, going, but then you get towards the end. It's like, damn it, this was a lot of work, but I'm already so far.

Let's keep going. Does that make sense? But if I known in the beginning, I wouldn't have done it so same way. That's where it's sort of like, Hey, the message here. Isn't like, wow, Google interviews are really tough. You need to know a whole bunch of stuff and you need to do a hundred things. The message should be, let's start as a data analyst, let's start as a business analyst, let's start with just Excel and keep going up and up and scale.

And up and up in like types of pay or responsibility or the capabilities you have, you know, because otherwise you'll just mentally crush yourself.

[01:16:26] Ted Hallum: Sure. Well, so Nick, as we start to wrap up I think back to Donald Rumsfeld and he used to have his four different types of knowledge. One of them was unknown unknowns.

So those things you don't know, you don't know. I didn't write the book on ACEing the data science interview you did. So we've mostly talked about the questions that I could think of to ask you as we start to, you know, prepare to leave our audience. What are the, what, what are the questions that I didn't know to ask that I should have asked from your perspective as the author of this book,

[01:17:04] Nick Singh: right?

I think, I think it's, it's not so much like. Question about the interview process, just this part of career advice that, you know, I always try to tell people about that. You know, it, you know, is that, I guess world's a really malleable place and it's built by people who are no smarter than you are not.

Right. So I mean, to say like, a lot of people are like, oh, I'll give you an example in your own story, you wrote so nicely to a recruiter being like, hi, I'm really close, but I'm not perfect. Like, do you want to, you know, you know, you wrote such a thoughtful, nice email that really showed your thought and you were like, ah, I'm so close and you got ghosted, you know, and you just like nothing, right.

For the same position that you actually end up getting and working. Right. So it's not even like, oh, sometimes they'll ghost you, but other times it's like the same position that happened. Right. So what I'm trying to say is like, you know, people think like, oh, maybe I can't work at Google or maybe I can't be at ML engine.

You know, I have it, you know, I just have an associates degree. How am I going to be an ML engineer? And it's just sort of like, wow, the world is a malleable place. And I might've heard like people overestimate what they can do in a year and then underestimate what they can do in 10. I honestly think people overestimate what they can do in a month and then underestimate what they can do in a year.

Like, I don't know anyone who worked for a year at a time on a 10-year goal, honestly like, Hey, you could learn sequel in three months. You can learn Excel at a really good level. In three months, you can learn to blow really well and get certified in Tableau in three months. So I'm not even trying to be like, oh, you're, you're overestimating what you can do in a year.

I'm saying people are thinking like, yo, all right, I'm gonna work really hard for three weekends. But football season just started or Hey, or like, you know what, you know, I've gotten really busy or, you know, I gotta take care of the kids. It's sort of like, oh my God, A year in a lot of people, like a year is not a long time.

Right? Like you did your master's program at nights over a year and a half or something like that. And then you take it 18 months and then you're taking pre-recs for that maybe six months beforehand. Right. But you change your whole education profile. You went from not doing the data science stuff to leave it as like a deep learning person in like two years.

Right. But Ted, I don't want to age you, but like, if you told me you were two years younger than you are now, I wouldn't blink. I wouldn't be like, whoa, like a really like, or if you told me you're two years older than you are not much, you know, like two years is like, like this, but you literally changed your life in two years.

Right. And I'm trying to say two years ago, I was writing code. Now I'm have a best-selling book as an author. And I don't even like to write, you know, I didn't go to college, learn all the CS and math stuff to be a writer. I, that was not even in my purview. Right. I, this was kind of an accident. I didn't try to be a career.

You know what I mean? I was working good jobs doing the thing. I didn't think I would, you know, I didn't think I'd leave Facebook to do any of this. This is a weird story of how this all happened, but I'm just trying to say that was just two years ago, I was doing that kind of work, you know? So I'm just trying to tell people like, Hey, like people think like, oh, this is the week I'm gonna become a data scientist next week, next week.

But it's like, whoa, if you just forget all of that and you just bring it down to SQL Excel to blow and you just set it to like, yo, I'm going to make one project around basketball. Cause I love basketball. I'm going to ignore all the hype and all the most cutting edge techniques. And I'm just going to try to understand the game of basketball from this one day to set with this one tool that that's where we got to focus on.

Because once you start doing that, it's crazy how fast like life can change in just two years. Right? Cause I didn't think I'd be. And here I am, and I don't know what I'm going to do next year, either. Right? And same way you didn't, you maybe two or three years ago, you weren't didn't know you'd be an ML engineer.

You didn't know what life would be like. So I think that's what I just want to leave people with is just sort of like, don't get so hung up on, like, here are the specific things I need to do right now. And like, I have this like big journey and like there's 300 pages in this book. And like, there's like nine things I have to learn.

It's sort of like, Hey, let's start somewhere. Cause you can really change your life. And the other thing is about the whole malleable thing is like one, you can change your life faster than you'd expect, sorry, slower than you expect. But in the grand scheme of life, it's way faster. It's like, Hey, you can like change your whole career, get double your salary in a year.

People don't know that you can do that. But people do that all the time by learning data skills. And I think the other thing is like, Hey, it's in your country. Like nothing is stopping you from getting this free data set on cattle or doing any of these things. These are all free. Coursera is free. MIT OpenCourseWare is free.

Now of course there's value to getting your degrees and there's value to being structured courses and having instructors and paying for all of that, for sure. But I'm just trying to tell people like, Hey, you can email anybody in the world today. You can learn anything in the world today. You can build a project as advanced as you want.

Thanks to the cloud. You can do something crazy, like huge scale just from your bedroom. It's all there. Now. It's just

[01:22:06] Ted Hallum: the willingness to AWS account. You can scale up to whatever you need.

[01:22:09] Nick Singh: Yeah. Yeah, exactly. You can train some crazy large models. You can do all kinds of stuff that before 20 years ago it was like select few research scientists at like E-bay or something could do.

Like that's what people in their like second or third machine learning class can do. Now. I just 20 or 21 is what people used to do. Twenty-five years ago at these, like, you know, at Yahoo, for example, to do like web searching. Like I can build my own search engine decently, well, in the next few weeks, you know, so that's just kind of the message I want to leave people with it's like w remember that the world is malleable and remember, you know, remember that story of how Ted you send that cold email for a job that they rejected you for and you got it.

How I sent a cold email to ju this kind of random job, I wasn't qualified for how I wrote a book out of thin air, even though I'm not an author. And definitely people in the military, they might be, feel constrained veterans or something. Like they might feel constrained like, oh, I've only done these certain things.

It's like, well, you can change your life fashion. And you can imagine, and it's, it's on you. It's not like, Hey, there's like 87 gatekeepers. It's like, whoa, you can start today.

[01:23:17] Ted Hallum: That was certainly my experience. Once I decided I wanted to make that pivot because. A natural passion for me, it snowballed very quickly.

And I think, you know, I wouldn't say while I love data science and machine learning. I wouldn't say they're for everybody. It's hard for me to imagine not liking it, but I know some people are, it's just not their thing. And so I would say that everything you said, at least in my experience was very true about how quickly you can pivot and build your skills.

And I would say that if you get into it and you're passionate, that's certainly going to be the case. If you get into it and it feels like hard work, that might be a red flag. I would, I would just add that too, because it doesn't get better. You have to like to learn, you have to enjoy learning this stuff because we talked earlier about it's changing all the time.

[01:24:06] Nick Singh: No one comment. I want to just say as though don't people, people look for excuses to give up all the time. I just want to push back on the red flag thing. Go talking to someone with more experience and then make sure you're, you're learning and a thing that is actually how do I say what should be being learned at that time?

Right. As in, if you don't have really strong statistics background, you're trying to learn machine learning. I would hate that too, because I don't have the pre-recs and I just making no sense. I'm like, what are these Greek symbols? I hate this. Right. But I don't actually hate it because I hate what machine learning can do and this and that.

I just hate it because I don't know it. Right. So like, I, I don't have the pre-recs. Right. So I've definitely worked with enough people. Like, I didn't like CS because I took too hard of a CS class. At one point in my life where I was just in a, the people around me were way smarter than me. It was this weird mix of circumstances where I was like, you know, and real young.

So it was just slurred. I was like, I was getting clowned on for being bad at coding. And I was like, yo, how are you guys doing this? And everyone's just showing off. And I'm like something wrong with me. Am I just not cut out for it? I was like, no, these people are just showing off. They're on the next level and it's just like a, Hey, my fundamentals aren't there.

Right? So once you start slow, so I just want to keep that there, like, Hey, if you just hate learning in general. Yeah. This is not going to work for you. Like none of this is going to work for you, but if you just feel like, Hey, you don't like it. Cause it feels really stressful and hard. See if there was an easier version to get to, because maybe that's the issue.

And that, that too often is people like jumping in machine learning when they don't even know statistics one-on-one or they learned it like eight years ago.

[01:25:43] Ted Hallum: Right now that reminds me of something that came up when you and I did our virtual coffee that love for you to kind of, as a closing note, expound on, you talked about pivoting, you talked about getting prerequisite knowledge so that what you want to do actually makes sense, and it's not artificially more difficult than it should be.

On this podcast, we've talked in the past. Degree programs and we've talked about bootcamps, but one thing we've never really touched on, which I thought, man, why have we not talked about this? You mentioned community colleges and picking up courses there. I think that's way under utilized way undervalued that people just don't a lot of times think about, but community colleges are in close proximity to virtually everyone.

And I think that's something that people could better leverage

[01:26:28] Nick Singh: for sure. And I'll give you an example. I was helping someone learn R and they hated it. They were struggling like crazy. And I asked them, you know, you know, and they're just like, I don't want to do any of this. This is too hard. And then when I broke it down, I realized this person was a learning their first programming language.

And they were learning R which is a programming language, but there's one thing to be like, I know five languages I'm learning are. And the other is like, I'm learning a programming language in general. I'm learning how to talk to a company. And then I'm dealing with our syntax, but like, do you see how there's a little difference?

You know what I mean? Right. And then their statistics was a little bit weak. They'd taken a stock class a long time ago. So now we're trying to build a simple, simple linear regression model in R and this person's hating it because the syntax is not making sense. The programming is not making sense, the mass aren't making sense.

And why would we even want to do predictions and make a linear regression model? That's not making sense cause they haven't done modeling in Excel. Right. And at that point it clicked in my head. Oh goodness. You know what, if you had known Python and JavaScript and now you're learning are at least some of the syntax would make sense.

Like, Hey, this is a function. Something. And you've worked with function before now. Same way. Let's say you've never done programming, but you'd done modeling and Excel. We're trying to just predict something forecasting in Excel. You're trying to forecast supply calendar, something like that. Hey, at least you understand what this modeling piece is.

Even if you don't understand code, and let's say you understood statistics, maybe you don't really know what modeling means in the real world or why we predict things, but Hey, at least, you know, when you build a linear regression model and it spits you out, like, Hey, is this a good fit or not? You know, about hypothesis testing and what it means for like, Hey, this was a good fit.

And what a R squared value means, right? So maybe you don't know the coding aspect. Maybe you don't know the modeling aspect, but at least you've heard of the word called R squared. You know? So I realized at that point, this person's hating it. Not because they hate data science or anything. It's just like too many things coming at once.

And that's where the community college aspect is so good. So I told this person like, honestly, forget about this, our stuff. If you had just taken a stat 1 0 1 class that you did a long time ago in college, but you've rightfully forgot about, cause you haven't touched in a while. You just took that one class and it came top of your mind.

And if you're able to take a second class later on Python programming one, or like any, any language Java, anything, then you would at least understand what the hell is this function doing? Oh, how do we write code? Oh, I know what a goodness of fit is. Or I know what an R squared is. Oh, okay. Now I'm understanding.

Or even though I've never learned R and both people will be learning are right. It's just that one person is actually learning a crazy more amount of skills and it's getting more WellMed versus the other. And that's where I think the community college thing has just slept on because people are like, okay, I'm going to join a bootcamp and learn everything under the sun, right.

From the get go where it's sort of like, Hey, even if you learn a little bit of programming or it skills before doing we bootcamp or doing a master's or you learn a little bit more stats before you start getting into the actual harder. It pays dividends. Right? And that's why some people are always just like, Hey, how do we get a master's in this stuff where it's like, Hey, maybe the correct thing is to do a prereq of stat and programming, just like two classes in each then, you know, should I get a master's?

And I think community college is a great way to kind of learn that in a structured way. And I know this material is there free and online. So you don't have to do it, but just sort of like, Hey, some people need the rigor. Some people need the structure and there's nothing wrong with you. If you can't wake up early in the morning to learn statistics, you need a class to like, kind of push you towards a set curriculum.

Like that's normal. You're just a human structure. Yeah. That's normal. That's how people work. And I always like to think like, oh, I'm above it, but it's like, even now I'm like, yo, unless I have someone on my butt telling me what to do, you know, I get a little lazy, you know, and then this is how it is. And I came up earlier that housing stat 1 0 1 in community college, I'm not going to be like, yo, I'm not going to do the homework homework.

I'll do that.

[01:30:46] Ted Hallum: Sure, but like the whole, the whole book is good advice. That is fantastic advice. My takeaway is if you're pivoting into this, don't set yourself up for failure. You're not a wizard. If you try to build a levitating building, you're going to fail. You have to have a foundation and it has to be built in a

[01:31:04] Nick Singh: logical order.

You look for red flags like, oh, I'm struggling too much. You're going to get it. Like, it's not easy for anybody at any level. Right? So it's like, if you're just like, always just like, Hey, I'm going to go until I struggle. Boom. You're always going to quit because this is a struggle. It's just, how can we be more thoughtful about the struggle intentional about the struggle?

Make the struggle, not as extreme as it could be. Like have a strategy, have a strategy throughout the struggle. Right? It's still going to be hard that none of us is easy. Never I've said this is going to be easy to do between

[01:31:33] Ted Hallum: hard tenable and hard. Impossible. Yeah, exactly. Cool. Well, Nick man, thank you so much for coming on the show.

This has been awesome. We're going to have a link to ACE the data science interview in the show notes. So if anybody would like to get a copy, you can get a copy. If you're in our slack workspace, back in October in the resources underscore books channel, you'll find a summary of everything on the back cover.

And so you can kind of check out more details on the book there and Nick, I really appreciate you coming on to help our veterans today, science and machine learning, ACE, the data science interview. Thank you so much for

[01:32:08] Nick Singh: your time. And people can find the book on Amazon. You can hit me up on LinkedIn nixing.

I have like 69,000 followers on the platform I post every day about career advice and job hunting. I've also got a blog on nixing.com with some free resources for the job hunt. So yeah, just check me out in variety of places.

[01:32:27]Ted Hallum: Thank you for joining Nick. And I's we chatted about AC in the data science interview. If you'd like to check out his book on that very topic, check out the show notes. There'll be a link there where you can get more information or pick up a copy for yourself with that until the next episode, a Bijou clean data, low P values and Godspeed on your data journey.