Speaker 2: (00:01)
One of the best pieces of advice I received was the same show. Don't tell I really, I like delivering rather than talking about delivering. Don't talk about your knowledge of Apache air flow. Show it you know. Don't talk about your knowledge of Python. Show itÉ

Speaker 3: (00:22)
***[inaudible]

Speaker 1: (00:31)
what's up everyone? Thank you so much for tuning into the artists of data science podcast. My goal with this podcast is to share the stories and journeys of the thought leaders in data science, the artists who are creating value for our field, to the content they're creating, the work they're doing, and the positive impact they're having within their organizations, industries, society, and the art of data science as a whole. I can't even begin to express how excited I am that you're joining me today. My name is Harpreet Sahota and I'll be your host as we talk to some of the most amazing people in data science. Today's episode is brought to you by data science dream job. If you're wondering what it takes to break into the field of data science, checkout dsdj.co/artists with an S for an invitation to a free webinar where we'll give you tips on how to land your first job in data science. 

Speaker 1: (01:23)
I've also got a free open mastermind Slack community called the artists of data science loft that I encourage everyone in listening to join. I'll make myself available to you for questions on all things data science and keep you posted on the biweekly open office hours that I'll be hosting for our community. Check that out@artofdatascienceloftslack.com community is super important and I'm hoping you guys will join the community where we can keep each other motivated, keeping each other in the loop on what's going on with our own journeys so that we can learn, grow and get better together. Let's ride this beat out into another awesome episode and don't forget to subscribe, follow like love rate and review the show. 

Speaker 3: (02:06)
***[inaudible]

Speaker 1: (02:25)
what's up everybody? Welcome to the local spotlight edition of the artists of data science podcast. If you've been around me long enough, you've heard me talk about my grand vision for healing 

Speaker 1: (02:33)
transformed the city. I live in Winnipeg into the center of excellence for data science across all of North America. Our guest today is hands down one of the best data scientists in our city and someone who I've been trying to get to work with me for a long time, one of these days I hope to win him over. She's got expertise in data science applying a combination of skills that he's developed over the years in computer science, mathematics, statistics and economics to solve some the most pressing transportation problems of our time. He's the author of one of the most awesome blogs that I've ever come across for all things data science and beyond and provides a ton of value through his two newsletters, the data science newsletter and the spaced repetition newsletter. He's even been generous enough to create an open source AKI deck that allows you to download his brain throughout his career. 

Speaker 1: (03:18)
He's applied his problem solving and programming prowess to conduct data analysis, exploration and visualization to present research findings for his stakeholders. He also works on developing software tools to increase productivity and automate tasks related to data entry, data cleaning and reporting. In his spare time. He crushes cattle data science competitions having placed in the top 5% twice and has developed data-driven web applications including an online resume website and the backend API for conference organization web application. He loves the challenge and excitement of continuous learning, developing new skills and solving hard problems, especially those NP hard problems. He's earned a bachelor of economics and mathematics from the University of Manitoba and has gone on to complete a master's in economics at Western university in London, Ontario. His work experience includes a seven year stint as a senior research associate at PRA where he participated in and manage research projects involving a wide variety of methodologies and domains. He's currently a data scientist at the transportation division in the city of Winnipeg. So please help me welcoming our very special guest today *** Mark. 
Speaker 1:
Thank you so much for, for taking time out of your schedule to be on the show. Really appreciate you being here. 
Speaker2 :
Absolutely. Thanks so much for having me. 
Speaker 1:
So, so talk to me about the path that led you into data science. 

Speaker 2: (04:38)
Well, that started probably when I was in undergrad. I decided to major in economics, which I say was the beginning of my journey. And that led me into a math background cause it turns out economics is actually a pretty highly quantitative field and if you want to do graduate school in economics, which I did, it's really valuable to have a lot of math and actually a lot of pure mathematics. And in fact, some people recommended to me that I just major in math rather than economics. So I decided to do a double major economics math along the way. I took a lot of statistics and programming courses as well and that eventually led me to PRA, uh, where I worked there as a research associate doing qualitative and quantitative research. So I uh, self taught myself programming in my spare time because it was just something that interested me. 

Speaker 2: (05:24)
And uh, I had a lot of um, my eye on doing more programming data work. Eventually that led to doing more cattle competitions. I found ways to apply programming skill of my job, started a blog and then eventually I came across the job posting for my current job as data scientists at traffic signals branch at city of Winnipeg and given the skills I had built up over the previous seven years or so, working towards a data science career, it was just a perfect fit. I looked at the job description, it was pretty clear this is right for me, so I applied for the job, got the job and I've been there for almost two years now. 

Speaker 1: (05:59)
Awesome background. Have you got, it's actually quite similar to my educational background as well, having studied economics and statistics, you mentioned that you got a blog, which is actually how I've found out about you, so I was wondering if you could tell us a bit more about your blog and how that's helped you learn and grow as a data scientist. 

Speaker 2: (06:15)
Yeah, absolutely. The blog has been pretty huge for me and the decision to start a blog was a pretty clear and easy one because it had a bunch of clear benefits at the outset. So one of them, and I've, you know, it's, it's always nice when decisions are clear like that when it's an obvious decision to do something because there's just so many reasons to do it. Like one of the reasons was learning by teaching, it's pretty well established in the learning literature that teaching is one of the most effective ways to learn. We have to constantly learn as data scientists because it's such a broad field and so many skills to wrap up on. So writing a blog post about something you just learned about is incredibly valuable because it helps you learn. Another thing is just improving written communication skills like being able to write. That's another really good one reason to start a blog and I think it's a pretty underrated skill for a data scientist to have is being able to do right effectively. 

Speaker 1: (07:17)
Are you an aspiring data scientist struggling to break into the field or then checkout dsdj.co/artists to reserve your spot for a free informational webinar on how you can break into the field that's going to be filled with amazing tips that are specifically designed to help you land your first job. Check it out dsdj.co/artists  

Speaker 1: (07:43)
Yeah, definitely man. I mean I, I totally agree with that sentiment because when you teach something you get to learn it twice. Um, so I think your, your blog is definitely a Testament to that. Um, and you know, you're mentioning something very important here and that's communication skills that are necessary to really be successful as a data scientist. And I've seen you post on LinkedIn about Toastmasters and I was wondering if you could tell us a bit about what Toastmasters is, what pathway you chose and how you see your career as a, as a data scientist benefiting from being a part of this amazing club?

Speaker 2: (08:15)
Yeah well Toastmasters is basically just a group that you joined to improve your verbal communication skills. So there's lots of Toastmasters club clubs in Winnipeg alone. I think there's about 60 my club. We meet every Wednesday, a video dictum Toastmasters club. And so you get together with these in these meetings and they're pretty structured and they are designed to train basically every part of public speaking that you can imagine. Everyone has a different role and it trains some different aspects. So there's grunt master who's listening for anyone who says, um, or uh, or you know, those types of things that infect your speech and make you sound less competent. There's a timekeeper that keeps track of all the time. There's a table topics master who runs the impromptu speaking session. There's a place to give speeches. So it trains a lot of different aspects of speaking. And one of the great benefits is everything is, uh, is evaluated too. 

Speaker 2: (09:09)
So when you give a speech or you, when you do anything there, there's someone else that's listening and we'll give you feedback about how well you did. So I think it's beneficial for anyone, especially people that struggle with public speaking personally is something I struggled with immensely. Um, especially impromptu speaking, being asked questions without being able to prepare in advance. That just struck fear in my heart of the probably three years ago and going on a podcast like this actually would have been a very frightening thing for me then. But now it was actually kind of exciting. I look forward to it and I owe that to Toastmasters. Really effective job interviews, presentations, communicating results to management. I wish I'm a big advocate. I recommend it to anyone. I kind of wished I started going when I was 10 years old, so I can't recommend it enough. 

Speaker 1: (10:00)
Yeah, definitely man. It helps to have a safe space where you can practice these skills and know that whatever feedback you're getting is only going to help you improve and progress your journey as as public speaker. Um, I actually just recently joined Toastmasters myself, um, having found out about it through work. Uh, so if anybody's listening here and you don't know how to get involved in Toastmasters, the first place to start would probably be at your place of employment because a lot of companies have Toastmasters clubs that they sponsor. We'll pay for your free membership and everything that you can be a part of. Yeah, I agree with you. It's been tremendously helpful for me with respect to impromptu speaking and even, you know, going on and creating a podcast like this, so I'm glad. I'm glad that you're doing well with Toastmasters. I saw one of your speeches that you had posted on LinkedIn about spaced repetition. I found that to be extremely fascinating. Would you be able to tell us a bit more about space repetition and how it's helped you learn more effectively? 

Speaker 2: (11:00)
yeah sure well, it's space repetition is basically a technique for remembering things that you learn forever. Basically retaining things to memory that you won't forget or will be very unlikely to forget. It's been backed up by a lot of researchers, and it's been known for a long time to be one of the most effective ways to retain knowledge. Essentially, it's quizzing yourself on information at increasing intervals of time. So to take a silly example, say you want to the capital of Manitoba Winnipeg, so you quiz yourself on that. Okay, it's Winnipeg. One day later, you quiz yourself on that, still got it. One week later you quiz yourself one month, one year. That's a way of, you still retain the knowledge, but it's, you do so in a really efficient way because there's this huge spacing effect. So one of the tools that I use, there's lots of great tools for this. Now probably the most popular one is ***Anchia your ranking, which is essentially flashcard. It is‡  software that handles the spacing for you. So I have cards in my anchia deck that I just answered today that won't be asked of me again for another three years. And because of the spacing, there's people that have ten anchia decks with tens of thousands of cards. And the burden of reviewing them every day actually isn't that bad because they're just spaced out so widely. You still only need 20 

Speaker 2: (12:16)
you know, maybe 1530 minutes a day to review. So it was something, it's something I've used for a long time since my university days. Back in the day it was super memo, which are the most popular software at the time. I consider it a key to my success in university and I continue use it today in data science, you're always trying to ramp up your skills and you always run into this issue of, you know, you read a textbook, but do you retain the information that's there? Like you don't really retain it unless you do something like space repetition or you work on some project where you get to apply it or preferably both. I think it's particularly valuable tool for data scientists. 

Speaker 1: (12:53)
Yeah, definitely man. Because data science is like not an easy job by any means. Right? There's a ton of continuous learning and continuous education that goes on in this field. Um, never mind, just the amount of learning you have to do to even break into the field. Uh, so developing ways to help yourself learn and help yourself retain information is crucial. It kind of reminds me in a sense, in a way it's kind of reminds me of deliberate practice. I don't know if you're familiar with, uh, with deliberate practice. Um, can, can you speak to, to kind of the, the similarities between these two methodologies? 

Speaker 2: (13:25)
Absolutely. I think if deliberate practices like kind of uncomfortable practice that really tests your knowledge and it turns out that's what you want, you kind of want to feel a bit uncomfortable. Like you're pushing yourself when you're learning. Cause if it feels really easy, that's kind of a sign that maybe you're not learning so much. So space repetition, I mean a corollary to that is testing well-established to be one of the most effective ways to learn because you just, you're not just passively reading and when you passively read stuff, when you're just like reading through a textbook is really easy to convince yourself. You understand something when you don't convince yourself that you'll remember something when you won't. And as soon as you start quizzing yourself on it, you realize, Oh, I actually, I didn't understand that very well at all. Or I just completely forgot that. So even if you don't use space repetition, deliberate practice, Mmm. Or testing, you know, doing that kind of flashcard testing is really useful, 

Speaker 1: (14:23)
especially if you're trying to progress in your career as a data scientist. So often. Thank you for sharing that with us. Um, so yeah, this awesome blog post on the hidden power of compounding, which I thought was an excellent read. So anybody listening go to marketing equal ***[inaudible] com check out that blog post. I especially like the four ideas you laid out for ways to compound in an area that you want to improve. Can you tell us more about, about this compounding and kind of touch on those four ideas that you had about the Blog post? 

Speaker 2: (14:52)
Yeah. Well most people probably are somewhat familiar with compounding when they probably heard the term compound interest, which is a little bit of a boring concept of when you have a bank account and it grows at some interest rate over time. If you give it enough time, it'll grow too. Uh, counter-intuitively large. Um, uh, it like snowballs. So because of the exponential growth of the, of the interest. So if, um, you have a bank account growing at 7% per year and 10% as, excuse me, in 10 years, your money doubles in 20 years it quadruples in 30 years, you have eight times your original amount. So if you have enough time and you have constant growth like that, then get some pretty huge results. The main thing in my speech that I've learned from listening to some, um, people that I admire is that this doesn't just apply to bank accounts. 

Speaker 2: (15:42)
It's something that applies to it. Basically, there's two underlying factors that you need. You need growth and time, something growing at some percentage and you're able to do that over some period of time. So it applies to any area where that happens. So particularly skills like programming or data science, sales, communication. If you can get a consistent growth rate in your skills in those areas over time, you're going to see some pretty huge results. And who's to say that a 7% growth rate in sales is what you'd expect? I mean, maybe you can improve at 20% per year and you'd see even more fantastic results. I mean, in terms of like how to do it, anything that increases your growth rate or the number of years that you can stick to it is great. The ways I had a couple of like specific, so there's some obvious candidates like if you have a mentor, reading is a good idea. 

Speaker 2: (16:33)
Obviously taking classes, all those things will help you compound knowledge. I recommend having a plan is one thing because you need to be able to apply growth consistently over years, so if you don't have a plan you're probably not going to be able to stick with it personally. I have a Google doc, a simple Google doc that just lists the areas that I want to be improving in and I just check on that every so often just to make sure I'm making some sort of progress in those areas. Another way to compound I think is space repetition, which we just talked about because it's all about retaining knowledge over the longterm and that's obviously gonna have a huge effect on compounding of your skills. If you can retain what you learn indefinitely. 

Speaker 3: (17:19)
***[inaudible]

Speaker 1: (17:20)
WhatŐs up artist? check out our free open mastermind Slack channel, theartistsofdatascienceloft, @artofdatascienceloft.slack.com I'll keep you posted on the biweekly open office hours that I'll be hosting and it's a great environment and community for all of us to talk all things data science.   Looking forward to seeing you there.
 ***[inaudible] 
Speaker 1: (17:50)
it's definitely a fascinating topic or reminds me of this book. I just recently read, think it was called the compound effect and he had an example in there where he's talking about if you get one penny everyday and double it, then by the end of the month you get some inordinate amount of money. I think it was in the multimillions or something like that. Do you have any books that you'd recommend to our listeners around this topic about compounding? 
Speaker 2:
On copounding, I don't have any specific books to recommend. I guess one thing I would recommend is this guy Tyler Cowen, he, he's an economist and he has a blog called marginal revolution. He's the one that kind of helped me shift my thinking of about compounding, not just in terms of bank accounts and interest rates, but something that applies to other areas of life. So if you search Tyler cowen compounding, you can find some interesting stuff there. Awesome. Yeah, I'll definitely put that into the show notes. Cool. Thank you for sharing that with us. 

Speaker 1: (18:31)
So let's talk about some of these awesome data science projects that you've done. Your products have been pretty cool. I remember when I was first looking for data scientists and Winnipeg came across your get hub pen and the creativity you've put into your projects are just pretty freaking awesome. I was wondering if you could talk to us about what your creative process is like for bringing your project ideas to reality and if you have tips for up and coming data scientists who don't know where to start with their project. 
Speaker 2:
Yeah, well the key is motivation for me. For me, a good project should have, it should be beneficial in a lot of different ways and the more ways to better. So I'm looking for something that is relevant to a skill that I want to learn that makes it obviously more motivating to do it in the first place. 

Speaker 2: (19:13)
If it's, you know, even if the project doesn't work, then at least I've built up some skills that are relevant to me. Something I'm passionate or curious about. That's obviously going to have effect on the motivation. Something that other people are curious or motivated about. That's kind of a personal thing because I would want the project that I put out there to be of interest to other people just because that's a personal motivator for me. Something that's relevant to my job. Something that could improve job prospects or your marketability where you could show to a potential employer, this is what I did, this proves that I have some skills or motivation or whatever. So the more of those benefits, the better. In terms of like getting ideas for projects, I usually, I just write them down. I always have a note taking thing by my side. 

Speaker 2: (19:54)
So whenever I come up with an idea, I just write it down and I find that I actually accumulate a whole bunch of ideas that way, way more than I could ever possibly work on in a lifetime. 
Speaker 1:
So when you come up with these ideas for these projects, is identifying data sources ever a issue and how do you go about identifying where you can find data to start working on a project?
Speaker 2:
 It'll usually start with an idea and then I'll do some Googling to see if there's some data to back it up. One of my favorite types of projects to work on is data scraping projects. I think that's a really great project for data science work cause you're kind of creating your own data. You learn so much about the intranet, you've learned about how to work with websites, how websites work about internet protocols. 

Speaker 2: (20:39)
You learn about messy data because the data that you get from scraping is going to require a bunch of cleaning. So you've got to clean it. And at the end of the day you have this data set that maybe no one else really has because no one else has scraped it. Maybe the website owners have the data, but they're probably not doing what you're doing with it. Like one of my favorites projects that I've done was I scraped marginal revolution, Tyler Cowen's blog that I was just talking about. I scraped every page off there and then I did a, uh, you know, put it through like a data pipeline to clean up the data. I did a little data analysis on it and then posted on my blog. It was kind of interesting. You could see the posting habits of the different authors. They're like, it's, the blog is run by two guys, Alex tab rock and Tyler Cohen. Alex would always post at 7:00 AM all the time. So I did like a, a chart showing when he posted it was just a spike at 7:00 AM and Tyler Cowen who was like distributed evenly. So that was, and there was a little stuff like that. It eventually got shared on marginal evolution, which was really exciting to me. I've read reading that blog since 2005, there were like thousands of people going to my site. I was not expecting that. Yeah. One of my favorite projects of all time. I highly recommend scraping projects. 

Speaker 1: (21:44)
Any particular tool they are using for web scraping? 

Speaker 2: (21:47)
Yeah, I use Python. Um, my Python's my main programming language, so I use that and um, uh, it depends on the project, but beautiful soup is the biggest probably the most popular HTML parser. So I usually use that to parse the HTML. Sometimes you have to use this other library called selenium, which basically acts as a, um, as it's useful in the websites when there's a lot of JavaScript and it basically opens up the browser and interacts with the website as an end user. So those are the two main tools. The requests library in Python is another one to make the HTTP requests as well. 

Speaker 1: (22:26)
So how do they find out about your blog? Did they just notice it? A shit ton of traffic coming from a particular IP address and they just contacted you? How did this, how did this happen? 

Speaker 2: (22:36)
I think, uh, I think what I did is I posted it on Twitter. I posted a link to it on Twitter and I tagged Tyler Cowen and then he kind of responded with, I can't remember how he responded, but the next day it was on his website and other go, man, I think, I think I saw my web traffic and I'm like, Whoa, why, why is this thousands of people visiting my website and I go on his website and there's a link to it right there. So that was, uh, that was a pretty fun moment for me. One thing I mentioned, like with scraping projects, make sure you can, you can get in trouble with scraping projects if you do. You don't want to, you don't want to down a website or do like a denial of service attack on a website. So be careful, you know, get some advice, make sure you know what you're doing. But it's a great project to work on. 

Speaker 1: (23:18)
Yeah, it's always pretty cool when you get like noticed by people that uh, you know, are more or less are kind of heroes or idols for lack of better word. Yes. That's pretty cool man. I remember one of these projects I saw on your 



GitHub, uh, was about the trees of Winnipeg. I thought that's awesome. And can you talk to us a little bit about that project and, and kind of how did you discover the Winnipeg open data portal and you know, or do you recommend data scientists that are up and coming to use open data portals?

Speaker 2: (23:46)
Yeah, so I think with that project I was, at the time when I started the project, I was, I think I was maybe interviewing for my job, the city. So I was interested in city stuff and I was looking at the website and thought, Oh my catch the attention of people that I'm, you know, able to work with data. And so I came across this, yeah, I think I started searching open data, which is full of really interesting data. And I came across this data set of all of the trees and Winnipeg are all, they're all inventory. I think it's like 300,000 trees, at least on public property. I think maybe that, it's not all the trees, but all of them on public property. So I just, I didn't do anything too fancy with it, but I parse the data. 

Speaker 2: (24:36)
I posted it on my website with interactive maps where you could see the distribution of trees in the city. It looked at some interesting stuff, like what are the biggest trees and diameter, uh, what are the rarest trees? And each of them had a map where you can, where you can see exactly where they were in the city. And that was [inaudible]. That was basically the project. It was kind of fun. 
Speaker 1:
Yeah, that's pretty cool. Uh, I remember after seeing that project in your website, I was digging around the open data portal and I found a database that had a distribution of parking tickets for pretty much every parking ticket that had been given in Winnipeg for the last several years. What would you say is the most interesting or weirdest dataset that you've come across on an open data portal? 

Speaker 2: (25:34)
This is kind of a compound answer, I guess, but I haven't, it would probably be the tree data. I wasn't expecting that at all. It was, it had the different varieties of trees. It had a lot of different details about the trees. I was like, this is really amazing that this exists. And it really is surprising to me that it existed. So I'd say that was probably the weirdest. I'm sure there's much weirder when you go on the open data. Yeah, yeah. You realize what people need to keep track of and the type of data sets that are out there. Endless. 
Speaker 1:
Yeah, man. There's something for everyone, right? We've got a tree dataset for all the Denver fuel yaks out there, so enjoy that stuff. So, so talk to me kind of about your, your mental framework for decision making. I think that's how it was it like a blog post or I think I might've saw something on LinkedIn where you're talking about this and I thought it be fascinating. 

I was wondering if you could talk about how, how this mental framework for decision making has helped you navigate the ambiguity of starting a data science project at work?
Speaker 2:
 The way I think about decisions is really influenced by my background. So economists are always thinking about costs and benefits. There's an entire subfield in economics called CBA cost benefit analysis. That's kind of how the economists makes decisions and that's how I try to at least as much as a human being can. So basically you have benefits and costs if the benefits outweigh the costs do it. If you're in one of those situations where you have different options to choose from, pick the one with the highest benefit to cost ratio. Um, so it sounds pretty simple but obviously easier said than done within benefits and costs. There are probability and impact to kind of help you weigh how big the probability is or how big the cost is. 

Speaker 2: (27:30)
So ah, if a benefit is high probability, like it's likely to happen and high impact, like it's a very good thing, then obviously that's going to have higher weight. If the cost is high probability and higher impact, like a really, really bad thing that's going to have higher weight. So, um, it definitely has a lot of influence over my decisions, especially in my role currently at the city of Winnipeg. Mmm. I've had to make a lot of decisions in the role because I've actually, the first person with that job title hired a first data, first data scientists at the city of Winnipeg. There's more now, but Mmm. It's, there's some ambiguity because it's a new job. It's a new role. People aren't entirely sure what it's going to involve, like what exactly you're going to work on. So it requires some initiative and kind of an entrepreneurial view to kind of figure out how to best add value. 

Speaker 2: (28:32)
So one of the first things I did was just talk to people and I, that's probably good advice for anyone starting a new job was like asking people, uh, getting to know people, asking them what their role is, what problems you deal with, what kind of data they have. Have you thought about doing anything with it? Uh, what do you expect of me? Um, especially do it, you especially want to do this with the person that you report to directly, but generally people in the organization. So doing that to me is high benefit. You're, you get to know people, first of all, you understand their needs, you get ideas for projects and those ideas are based in reality. They aren't just concocted up in your brain. It's based on what people actually need. So it helps you avoid building the wrong thing or spending time on the wrong thing and the costs. 

Speaker 2: (29:16)
I mean, I can't think of what the costs of doing that is besides your time. And then, so that's, that was a clear decision to do that at first. And then after taking stock of people's needs and developing a huge list of potential projects, cause that's probably what will happen. There'll be a lot of potential projects and then then you can go through each project and get a sense of the benefits and the costs of each individual one. And then, uh, winner, it'll hopefully a winner will come out or a few winners in terms of like what you're going to be doing for the next little while. 

Speaker 1: (29:47)
Great advice man. Thank you so much for that. So it's always interesting when people ask you, Oh, what do you do for work? Oh, what's your job title? And you say data scientists. Have you had to kind of explain what a data scientist does to, to a lay person? You get that kind of like what data who
 Speaker 2: (30:05)
Yeah, it's a tough question to answer because it is a little bit ambiguous sometimes it's a pretty broad term and it can mean different things depending on the organization that you're entering into. Like I was the first data scientist hired, but there's a lot of other people working with data at the city for a long time. There's people in analyst roles, just programmers and developers work with data constantly. But the explanation I usually give is I, we have a lot of data and organizations have a lot of data and we want to be able to use it. Yeah. We want to be able to manage it. There's a lot of value and things that we could potentially do with the data. So we need someone, your role as a data scientist is kind of a steward of that data, of understanding the data and getting value out of it for the organization. 

Speaker 1: (30:48)
So before we jump in to our lightning round here, I wanted to ask you, what's the one thing you want people to learn from your story? 

Speaker 2: (30:55)
One of the best pieces of advice I received was, uh, I, I like the saying show, don't tell. So to me, I, I think it's much better to, I really, I like delivering rather than talking about delivering. So as much as I can, I try to do that. So don't talk about your knowledge of Apache airflow, you know, show it by writing an article about it. No, don't talk about your knowledge of Python. Show it by like building a web application or contributing to open source. I just think that's more effective and impressive and more fun and sell and satisfying to yourself. Um, you know, if you're building a startup , I've heard a famous, I think Paul Graham, the famous or someone Paul or other famous startup investors saying that it's so much more valuable to have a working prototype, then a PowerPoint deck when they're, when they're evaluating startups. It just says so much more than words to have an actual working prototype. Um, so I try to follow that advice as much as I, as much as I can. Like if you build interesting, useful things to people, um, if you deliver on important projects at work, uh, good things will happen. And you know, talking and talking about him doesn't necessarily do much good. And I mean, you, you want to get them in front of people. Obviously you're all that's important, but um, but do it, you know, delivering is what's most important. 

Speaker 1: (32:19)
That's right, man. Real artists ship. Right. That's awesome advice, man. I think all the work that you've done and everything that you've contributed to the data science community is a Testament that you are living true to that advice fan. So thank you very much for, for Oh, the contributions you've made to data science and especially here in our city. 

Speaker 1: (32:39)
So let's go into our lightning round real quick Yeah. Question one, you've kind of already answered Python or R.
Speaker 2: (32:41)Yeah, definitely Python knock. I don't have anything against our, I've used, ours are a little bit, but I, I'm definitely a Python guy Python for sure. Awesome. Uh, senior man pine on all the way. 
Speaker 1:
 What's a book every data scientist should read? 
Speaker 2:
I would say I, I, I'm big advocate of writing and being a good, uh, improving your writing skills. So there's this book called on writing well by William Zinsser. Oh my God, I'm not sure I'm able to pronounce his last name right. Williams Zinsser uh, it's a great book. Even if you just read the fourth chapter, first four chapters, it's really, really valuable and with lots of insights to take away to improving your writing.
Speaker 1:
 Awesome. I'll definitely, I'll add that to the show notes as well, and I think I might even get me a copy of that book. Um, so what's your favorite question to ask in a job interview?

Speaker 2: (33:30)
I actually haven't given that many job interviews. I have given a few, uh, but I think one of the best questions is tell me to just tell me about a project you worked on and how it's relevant to the position. It just gets so much. You get a lot of information about the candidates, um, and how their skills are relevant, how they're able to explain what they did, um, get a sense of their communication skills and their technical skills. 
Speaker 1:
That's awesome. That's a great question man. And great, great rationale behind asking that as well. Um, how about the strangest question that you've been asked in a job interview? 
Soeaker 2:
I was asked once, what's the difference between a geographic coordinate system and a projected coordinate system? I was a little bit taken aback. I mean the, the geographic, uh, data analysis was relevant to the position, so I don't, I actually wasn't very happy with how I answered the question, but I was kind of taken aback I wasn't expecting that one.
Speaker 1: (34:31)
Did you end up getting the job? 
Speaker 2:
I did, yeah.
Speaker 1:
 Nice. Nice. So certifications or self directed learning? 
Speajer 2:
I'm a huge fan of self directed learning, but I realized that certification you need to, you need to show people what you do, like you need to essentially a certification is just a way of showing people what you did. So I think you can do both self-directed, but showing the stuff that you've worked on. I'm not a big fan a, I mean I like online courses, like there's lots of information out there, but the certifications, I'm not totally sure it's that valuable to maybe have some more acronyms at the end of your name on your resume. It depends on the certification I guess, but I'm a big fan of self-directed self-directed projects that you can show and release out to the public so other people can see.
Speaker 1
How can people connect with you? Where can they find you? 
Speaker 2: (35:19)
You can go to my website, Mark nagelberg.com that's where I have my blog. There's contact info there, M a R K N a G E L B E R g.com. So my blogs there, I've got a my space repetition newsletter, which talks about productive learning, repetition, and just generally productive learning techniques. Um, and I also have a data science newsletter as well. And if you're interested in improving your public speaking skills, I'm president of any addictive Toastmasters club. So if you want to meet up there, that's great. We meet Wednesday nights a confusing corner. Just reach out to me if you're interested. 
Speaker 1:
Awesome. Awesome. Definitely man. Hey, well thank you so much for your time. Mark is really great having you on the show. I think our listeners are going to really take a lot away from your story and your perspective on data science. So happy to have you here. 
Speaker 2:
Look forward to hopefully someday having me back on the show, man. Right on. Thanks so much. Are pretty, has been great. 

Speaker 3: (36:14)
[inaudible].