Kirill: This is episode number 35 with data science course critic David Venturi.
(background music plays)
Welcome to the SuperDataScience podcast. My name is Kirill Eremenko, data science coach and lifestyle entrepreneur. And each week we bring you inspiring people and ideas to help you build your successful career in data science. Thanks for being here today and now let’s make the complex simple.
(background music plays)
Hello and welcome to the SuperDataScience podcast. Got a very interesting guest today. Today we’re talking with David Venturi, and David is a fascinating person because he, at some point in his life, understood that he wants to study data science, he wants to be a data scientist. And he actually enrolled in one of the prestigious universities in Canada for a data science degree, for a data science Masters. But within two weeks of studying there, he dropped out. And he didn’t drop out because he was not satisfied with his career choice and he wanted to do something else, no. He dropped out because he realised that what he was learning wasn’t sufficient enough. And guess what he did instead? Instead, he created his own Masters programme based on courses available online.
So he looked at different data science Masters programmes that are offered by different universities, and then based on that, he compiled his own programme using the courses that he can purchase or find for free online. And I thought that it was an ingenious idea. It saved him $30,000, and also of course, he was free to pick the best quality of content that was available to him.
So we’re going to talk a lot about that, and if you are interested in learning more about data science, and getting to the space, and finding the best courses on data science, then this podcast is for you. David will give you some tips and tricks on how to do that, and moreover, he will show you his own guidelines. So he actually has some very interesting articles published around the best courses on data science, and he’s continued to do that research so you can actually copy what he’s doing and learn data science on your own online. Because ultimately, what’s the end goal of learning? It’s not that piece of paper that you get from a university. It’s the skills and knowledge that you get from the training. And if the training is better online because it’s more diverse, you have access to more interesting instructors, and you can do it at your own pace, then why not do it online?
So I’m super excited for you to check out this podcast, especially if you’re passionate about learning more and more data science tools, techniques, and methodologies and structuring your own pathway into the field of data science. And without further ado, I bring to you David Venturi.
(background music plays)
Welcome everybody to the SuperDataScience podcast. Today I’ve got a very interesting and exciting guest with us, David Venturi, calling in from Toronto, Canada. David, welcome to the show. How are you today?
David: Hi Kirill. Thanks, I’m good. I’m good.
Kirill: Awesome. How’s the weather in Canada?
David: Yeah, it’s fine. It’s warmed up. We’re in February now, it should be usually around -10 now, but it’s above 0 Celsius right now. It’s nice.
Kirill: Yeah. Ok, cool. Cool. Well yeah, so very, very nice to have you on the show, and David, for our listeners, how I got in touch with David is because David wrote this incredible article, I just was very surprised and excited to read it, where he rated the courses on data science, and it turned out that according to his rating, my “Data Science A-Z” course was the top course in his opinion. So I was very excited about that and wanted to invite David onto the show, but not just because of that, but also because David actually is creating a Masters in Data Science for himself. So maybe let’s start with that. So how did you come about getting into the space of data science?
David: Yeah, so it’s kind of a long-ish story. So I was working in Western Canada. My background is I’m a chemical engineer and I was working for a company out there, and I decided to go on vacation for my birthday, and I visited my friend, who works as a software engineer in San Francisco, just for fun, just a fun vacation. And he picked me up at the airport there. Now he had mentioned that he interviewed at a company called Coursera, which your listeners may or may not be familiar with. It’s an education company that offers free online courses from really, really good universities, Stanford, Johns Hopkins, and such.
And throughout the weekend, we were talking about how I wasn’t exactly ecstatic with where my career was going as a potential chemical engineer, and by the end of the weekend, he and I were convinced enough that I should try some of these online courses. So I dove into a programming course right when I got home, and fast forward six months, but I ended up flying to a computer science programme at the University of Toronto, which is one of the best programmes in Canada. After two weeks at that programme, I almost felt like I was regressing a little bit almost. Like I was so in love with the online method, being on the pod videos. The selection is amazing. And so I dropped it after two weeks, and a huge reason for dropping it was because I could focus on something that I liked better, and that happened to be data science.
A lot of reasons there, but basically it was I had complete control of the courses I took, it was $30,000 cheaper —
Kirill: Just $30,000 cheaper!
David: Yeah. Not pocket change. And yeah, this was what I loved. I’m a huge data geek, it’s been what I’ve been doing my whole life. That’s what I enjoy. And all of it is right there in front of me to do.
Kirill: And was it a hard decision to make, to drop out of university? That’s obviously one of the prestigious universities, it must have been hard to get in, and to actually secure the spot. Was it a hard decision to make?
David: Surprisingly, it wasn’t. It was one of those gut feelings. It was just, “this doesn’t feel right.” I had already been at a university before, which was a huge factor in the decision. I had university experience. But it was just the quality of the course. I wasn’t scared at all about sacrificing quality. So no, it wasn’t very hard.
Kirill: Ok, ok. Gotcha. That’s really cool. And now as you say on your LinkedIn profile and other places, you’re creating your own Masters of Data Science through online courses. Tell us a bit more about that. How are you going about doing that?
David: Basically, what I did was I didn’t want to just go in there and mix and match courses. I wanted to have a data-driven approach. So what I did was I looked at what universities offered in an actual traditional Data Science Masters, or even Data Science Bachelors programs. I took notes on what subjects they offered. As you know, data science is a very multidisciplinary discipline. So, take the best stats courses, the actual data science process like wrangling, cleaning, machine learning, etc.
So, I had to somehow find a way to find the best courses online in those subjects. My method was looking up these subject areas on online course review sites. There’s a few popular ones out there. One that I work for now is called Class Central, and they have thousands of reviews. So, using the data from that website and just a general guideline of what a data science program should look like, I then made my own iteration, a highly customized Masters programme which I’ve been doing for about a year now, like 30 courses in total—I think I’ve done about 25 now, so nearly done, yeah.
Kirill: That’s really cool. And is it expensive? Is it costing you $30,000 to do this Masters program?
David: No. I’m estimating it’s about just over $1,000. Some of the courses I am doing, they aren’t completely free because I think a huge part of learning is having supervised grading, so some of them you have to pay for the grading on Udacity, for example, which is one of the main course providers I use.
Kirill: Gotcha. Yeah, that sounds like a pretty cool idea. You could maybe even put that on the Class Central website. You could create this Masters like a shell and then get people to sign up for it and then just complete those courses, show your evidence of completion, and then you could issue a Masters degree based on that.
David: Yeah, that’s something to look at. That’s a good idea. It’s something to dabble in, for sure.
Kirill: Yeah, that’s awesome. And how are you feeling about it? Because you mentioned that when you were studying at the university degree, you felt that you were regressing. Are you feeling different now? Are you feeling that you’re in control, that you’re actually learning the things that you want to be learning?
David: Yeah, definitely. There are some challenges with the self-taught method. It’s hard to stay disciplined sometimes with time management with no deadlines. There’s no professor or classmates or even pressure on you to finish your assignments this weekend or get this course done by this time.
Yeah, I’m ecstatic with my choice to do the DIY approach. The selection is amazing, the freedom is really great. I went on a random trip to India at the end of December just because I could. You know, I didn’t have exams. It’s not for everybody, but I think the online method and learning from videos where you can pause and search on Google when you’re stuck, and just kind of go your own pace, I think that benefits me as a learner the most. So that’s probably the number one reason I’m happy with my choice.
Kirill: Gotcha. I think this whole example is going to be inspiring for a lot of our listeners because a lot of the people who are viewing my courses or listening to this show, they’re kind of sometimes hesitant. “Will this be a credible thing on my resume? You know, if I go and complete a degree, it’s something I can put on my resume and employers are going to recognize it. It’s going to get me jobs and open doors for me. But if I complete an online course, people are going to look at it and they’re not going to really value me as much as if I had completed a proper degree.” What are your thoughts on that? Do you think it’s still a credible way to get education? Are you worried about not being able to get a job after you complete this Masters that you’ve designed for yourself?
David: That’s a really good question, one that I often get of the comments of the articles that I write. So with respect to the actual learning process, I think that the selection of what’s available—it’s literally just courses that are offered at digital universities or by experts like you. So the actual learning process and absorbing the information, I think it’s just as effective if you do it the right way, if you’re disciplined.
Kirill: I personally think it’s even better than at university.
David: I would agree, yeah. But the aspect of getting a job, that’s a real issue. I can’t say that it’s going to be as easy using these online courses and going to a company that has a massive HR department. In the current stage that that we are in, for online courses at least, you’re probably going to be limiting some of your options.
But if you do it right, if you make a portfolio, if you prove yourself and get a social media presence and network, it’s definitely doable. There are people that have done it and I’m excited to do it soon. Yeah, I think there’s challenges, but they can be overcome if you’re the right candidate, which is probably standard for any applicant really.
Kirill: Yeah, I totally agree. And this resonates with what Josh Coulson said in podcast #33. Of course, as you say, the large companies—you might be sacrificing or limiting your options just because of their very templated approach to hiring. You know, they check these things and they might miss talent, but that’s their problem. It’s their fault and it’s something they have to improve.
But ideally, a manager who is hiring for their own team or an entrepreneur or a CEO, when they’re hiring, they actually want not just to hire the skillset, they want to hire passionate people who are excited to learn, who are driven to find out new things, to solve problems and so on. That’s probably the most important thing.
So, if you’re doing a Masters of your own, and in addition to that you can start creating an online presence, which I see you’re already doing, like you have at least two articles that you’ve published and everybody can see them on your LinkedIn, and you’re probably doing other things and people can use Twitter, LinkedIn, blogs and so on to create content and put their ideas out there, opinions, and create a following.
And then through that you start networking. You start networking early on. Say you’re one year into your two-year Masters, and you start networking and you get some connections, and you won’t even need to apply for jobs by the end of it because people will already know who you are, and people will be anticipating, people will be waiting for you to join their teams, because they know that you’re a driven and motivated person. So, I think you’re on the right track. I think your opportunities are out there waiting for you already.
David: Yeah, I agree with that for sure.
Kirill: Tell us a little bit about the structure of the course. What do you think are the most important aspects from data science? Like, the way you’ve structured your Masters degree and the courses that you’ve taken, what did you find are the most important tools that other universities out there include in their programmes, the tools that you’ve chosen to learn about, maybe techniques and methodologies? What should somebody trying to copy your approach also focus on?
David: Okay. So, I think for data science specifically, I think a huge thing is the data science process, which I get into in the article that your “Data Science A-Z” course was focused in. Data science is inherently a multidisciplinary subject. You’ve got stats, you’ve got math, you’ve got just general data management things, several things, and universities can teach these things because these courses are modules, they’re separate, they can link things together.
So I think, starting off, you definitely want a good course that has the data science process at its heart, like someone leading you through and saying, “Okay, this is why you do this here. This is why cleaning this dataset this way is important this way,” rather than just learning the tools to clean datasets like Python, like using a good tool. If you want to learn the motivations behind it, and I think that’s a huge key. So rather than taking the best stats course, which is you just learning the pure stats basics, you want to have that in a data science context. These courses out there, they exist – you know, stats for data science and programming, like learning Python for data science. You want to have that all linked together.
Kirill: Okay. That’s very, very solid advice. So does it really matter what tools you specifically learn, or is the process more important?
David: That’s a good question.
Kirill: I’ll tell you why I ask this question. Because when I was creating Data Science A-Z course, I did a survey of all my existing students back then, and a lot of them said that—I said, “I’m creating this course. What do you guys want to know? What do you want to know about data science?” And I already had students doing my Tableau course, or my Tableau course and some other course, maybe. So I had some people interested in data science.
A lot of them said, “We want to know stats and this and that and data cleaning and so on,” but an overwhelming – not the majority, but an overwhelming number of students said that they would like to know R programming inside this course, and some said Python. But what you probably notice from the course is that it doesn’t have R, and it doesn’t have Python. And a lot of people later asked me, “Why did you structure a course and you didn’t include two of the most essential tools in data science, R and Python?” My answer was simple: “Look, I would love to include R or Python, but I don’t want the people taking this course to get bogged down in learning this programming language and spending 5 hours or however long from the course just focused on that, because that will really turn them away from data science and they won’t be in that state of flow when they’re learning about the actual concepts behind it in a sense and, like you say, the whole process, the end-to-end process.
And that’s why I use a simpler tool, Gretl, which is kind of a visual tool, it has a graphical user interface. It’s very simple to build models in Gretl. It’s not really used in the industry, as in employers don’t really use it, but it really helps you get into the nitty-gritty of what data science is and what it’s all about.
That’s why I asked this question. When you were taking this course, and I’m assuming you’ve looked through it or maybe even taken it fully, did you feel that you need to have R and Python right away, or is it more important for you to first understand the process of data science and then kind of build these other tools as add-ons onto your foundational knowledge?
David: When I was doing your course, I already had Python and R back then. But yeah, going into what you’re saying, I think your choice was correct. Using an easy to grasp tool, that is definitely more important, to communicate the process rather than the tool. I would say that you made the right choice and it’s reflected in how I rated your course on the “Best Intro to Data Science Courses” article. Knowing Python and R is not 100% necessary. It could help if you could offer two versions of that course using both tools or maybe even expect students to have a background going into this course, but as a catch-all intro, I would definitely rather it be an easy to grasp tool.
Kirill: Okay, gotcha. Thank you. And out of R and Python, since you have a background in both, what are your thoughts on where is the world going in terms of R and Python? Should people who are just starting out now, listening to this podcast, should they learn R, or should they learn Python?
David: That’s a really good question. I don’t have a hot take answer on that one. I know there are people that love both, hate one, there could be some that hate both, I guess. But I don’t have a good answer for you, unfortunately. I’m sorry.
Kirill: Okay. All right. But from the university programs that you’ve looked at, do most of them offer R, or do most of them offer Python? Or is it like an even split?
David: Yeah, so a lot of them, I’d guess 75% choose Python. Most stats courses use R, which makes sense since R is a stats dominated programming language for the most part. But yeah, people usually choose Python because it’s easier to grasp than R for the relatively intro courses for data science.
Kirill: Yeah. Personally, I really like working with R. I like the RStudio, and just kind of like how everything is nice and neat and the vectorized operations, but what I’m seeing as well is that Python is kind of more versatile because it has that extended use case to it that you can actually use it to develop applications, and when it comes to machine learning and artificial intelligence, that’s when Python kind of supersedes R.
Okay, so that’s pretty cool. We’ve chatted a bit about the course and a bit of your background. What I wanted to also ask you is who has influenced you, apart from your friend?
By the way, it was really interesting to know that you’re a chemical engineer because you’re the third chemical engineer on the podcast. We had Ruben Kogel in episode 1, Ben Taylor in episode 29, and now you’re also a chemical engineer so I have this tally of chemical engineers versus physicists, and so far we’ve got two physicists, three chemical engineers.
So apart from your friend in San Francisco, who else or what else influenced you to choose data science or even to stick to your choice after you made that decision? Obviously, as human beings we always think have we made the right choice, are we doing the right thing, “Will I have a successful career here? Is the world going in this direction?” What has motivated you and strengthened your understanding that data science or machine learning or artificial intelligence, all these things that you’re learning, that they are indeed the correct path for you?
David: I have to say that a huge thing, other than going to my friend in San Francisco—that trip almost didn’t happen and I could very well be somewhere else right now. I almost bailed on that trip like a day before because work was too busy. That was 100% the most influential thing. It’s a life changing event, truly.
But other than confirming my suspicion that data science would be right for me, other than compiling my Twitter feed and surrounding myself as much as possible with influential and insightful data science and machine learning people, I can’t really say there was one specific person. It was just that I knew myself and I knew that data was basically what I was doing in all of my other jobs. I just wasn’t titled data analyst or data scientist. I was always digging deep into data, always working with data in my spare time and studying sports analytics. That’s what I was doing for fun. Like, it was mostly knowing myself and seeing other important and influential people doing similar things to me and knowing that it’s going to have a huge impact on the world going forward. And that is pretty much it.
Kirill: Okay. That’s very interesting. And how do you go about surrounding yourself with influential people in the space of data science? I think a lot of people listening to this podcast could benefit from that. I think everybody should surround themselves with influential people and learn from them and bounce ideas and network. How do you go about doing this?
David: Yeah, so the one major thing for me is—I can’t say I know them personally, but I really enjoy and find podcasts to be a beneficial resource. Your podcast, for example, and there are several other data science related podcasts that I do enjoy and listen every morning over breakfast. We always want to learn. We always want to progress and things like “Linear Digressions” is one of the first ones I listened to.
Kirill: Yeah. I totally don’t mind if you share which podcasts you listen to on data science. Maybe people will pick up some cool ideas from you.
David: “Linear Digressions” is the main one I started with. Also, A16Z is a venture capital firm in the States, they have a really good podcast. It’s not data science specific, but they do focus a lot on AI and machine learning. That’s really interesting. There’s one called “This Week in Machine Learning” which is more technical, but I find that I learn a lot on that one. There’s not necessarily one major one, but—
Kirill: You grab bits and pieces here and there.
David: Yeah, exactly.
Kirill: All right, cool. Thanks for sharing that. So, you listen to podcasts. Any other ways you get opinions from influential people?
David: Other than filling my Twitter feed with people and interacting with them, I’ll reach out to someone who had an interesting article on Twitter, send them a DM and chat here and there, or LinkedIn messages or typical meetups—have you heard of this meetup thing in Australia?
Kirill: Yeah, I’ve done a few meetups, but not in data science. I’ve done meetups with people who are learning French or—I think that’s the only one I’ve done. (Laughs)
David: Yeah, I didn’t do too many of those, either. But yeah, there are so many options now.
Kirill: Interesting that you say that, because here at SuperDataScience we really want to connect people and we’re working on a project that’s going to facilitate interaction between people who are studying or doing data science and their peers in their city. That’s going to be very exciting. What do you think? Is it beneficial for data scientists to meet up in real life or IRL, as they say, to catch up and have chats?
David: I think so, yeah. If you can work on a project or like a Kaggle competition or just solving a random programming project related to data. I think it’s fun making new friends. I think it’s important. I don’t think it’s absolutely necessary, but I think it’s positive for sure.
Kirill: Yeah, gotcha. Okay, cool. And since you’ve started into this Data Science Masters of your own, what has been the biggest challenge for you?
David: Good question. I would probably say staying positive about your progress. Doing a self-directed program, it comes with its challenges because there’s no defined path. You can always do more. For example, today I studied 6-7 hours so far, and it’s—when is it appropriate to stop? You know, I would like to start a career soon. Like, how much is healthy with no deadlines, if that makes sense? In university you have deadlines, but here you’re just kind of self-directed and it’s easy to put pressure on yourself to do more than you can. So, you have to keep your work-life balance at a good level, and I find myself struggling with that.
Kirill: So, 6-7 hours? That’s quite a lot. Are you retaining all of that information? Excactly, how do you know when you’ve studied enough, even just for the day?
David: For the day, it’s usually when it’s just time to go to bed. (Laughs) That’s for retaining information. It’s been proven in studies you have to take breaks. I find that as long as you take breaks, you can do 13 hours straight, as long as you’re taking breaks in there. It’s less so being worried about retaining information rather than planning time to go to the gym. Or planning time for friends and family, if that makes sense.
Kirill: Yeah, totally. And how do you choose the courses? Do you study one course at a time? Or are you studying two or three courses in parallel?
David: That’s a good point. I choose to do it singular. When I was in university, I found that challenging. Did you find that challenging, to balance five courses at once so you have to compartmentalize your mind a little bit?
Kirill: That’s a good question. Like, in my Masters degree, definitely not because it was just Masters of Accounting and Finance. It was pretty simple stuff. I didn’t spend that much time. In my Bachelors degree, I found that it was actually good for me to have multiple at the same time because they were so different. Like one, let’s say, was physics, other one was chemistry, other one was mathematics, and then another type of mathematics, and another type of physics. Like, if I had been studying just the one, like sequentially, to fit four subjects, for instance, into a semester, I would really have to study every single day this one subject and I’d get sick of it. I wouldn’t have the variety. I liked the variety of different topics and to be able to switch my brain over. So I didn’t find it strenuous during my Bachelors degree to have four at the same time. But you’re saying sequential is better for you, right?
David: Yeah, it’s more of a personal preference thing here. I prefer to do one. I almost have somewhat of a completion complex in a way. I like focusing on one task.
Kirill: Okay, that’s interesting. And then how do you make sure, once you’ve completed a course, how do you make sure that you don’t instantly forget everything? Because that’s what normally happens at university. Or maybe that’s what happened to me. You study for a semester, you prepare for the exams, you know everything, you smash out the exam, and two days later you don’t remember a thing. Or you don’t remember half the things because you were so stressed out about it and you put all your maximum energy into it. And then moreover the next semester approaches, and you move on to new subjects, and that knowledge slowly fades away.
You can’t really afford that in data science, especially in preparing for a career. You need to keep your knowledge fresh and even more so on the cutting edge of that subject. So how do you make sure that, if you’ve studied R programming and now you’ve moved onto visualization, that you maintain that knowledge about R or whatever you learned before?
David: One thing is just taking detailed notes. That’s like an obvious one. But when you’re doing more programming stuff, like keeping your code on GitHub, for example, you still have that easy place to go back to and see what you did. You could easily just watch several videos of a course and, like you said, just forget it. So keeping documentation is important.
And one thing that’s less documentation oriented is more about how you’re studying. So even though I do try to finish my courses as fast as possible, I’m never just trying to get them done for the sake of being done. I think I read a couple of weeks ago – I think it was released from one of the big online learning platforms, maybe edX or something. Maybe Udacity. They said that students that take more attempts at learning, for example, a certain quiz, rather than just getting the quiz right, you actually experiment in a quiz and see what you can break, like what is right and what’s wrong, take your time learning at certain checkpoints rather than just rushing to them. I think that approach to learning will help you retain that better.
Kirill: Yeah, I totally agree with that. I remember finding myself—and I still do that when I’m even solving a problem, like a mathematics problem, or doing something in coding and I know what the correct approach is, and then I have an idea, “Oh, actually it’d be interesting to see what happens if I do this.” Even though I know it’s going to waste my time, it’s not the result I want and it’s probably not going to be a part of the end product, I still do it just to try it out and then you get cool ideas from there and it helps you learn better when you actually let your curiosity take over. You’re right, it helps retain the knowledge as well.
So the next thing I wanted to ask you: Can you walk us through your day? Because you’re the complete master of whatever you do, you don’t have to go to a 9 to 5 job, you don’t have to wake up at certain time, what sequence does your day go through? What time do you wake up? What do you do first? Do you go studying? Do you go for a walk? How do you pass your days in a way that it doesn’t become mundane, routine and you don’t get bored?
David: Okay, so my ideal day—I’d usually wake up around 9, and then over breakfast I’ll have one of those podcasts playing in the background there. So I ease into my day with someone talking to me about data science or whatever I’m interested in that day. And then I usually like to dedicate 40-45 minute chunks of pure studying and then 10-15 minute breaks for about 3-4 hours in the morning. Then I’d go for lunch. And then depending on the day, I’ll mix in going to the gym or—that’s actually mostly it. (Laughs) I live a pretty boring life.
Yeah, I think that one thing that’s helped me maintain my sanity throughout my whole life is sports, so I do like doing an organized sport. Currently I play water polo. Just being in a big 100-person team or something. I’ll get the social interaction there, or somewhere, and then usually I’ll come back and study before bed or work on projects before bed.
Kirill: That’s really cool. And yeah, I agree, social interaction is very important a lot of the time. Especially when you’re doing something that you’re completely in control of, it’s very easy to lose track of that and that’s not good to be all by yourself all the time.
David: No, definitely not.
Kirill: Okay. And what’s a recent success that you’ve had? Something that you can share with us, something in your studies that you’re proud of, something that you’ve accomplished that’s a major accomplishment?
David: So, one that you’re familiar with, and definitely the thing I’m most proud of, is Class Central. I work for them part-time and it’s the data science career guide which is a part of a larger idea for Class Central, which is career guides in general. Basically you’re picking the best courses for whatever, it could be data science, it could be software engineering, it could be history. So, I’m kind of in charge of that whole product there, and so far we’ve released three articles for the data science career guide.
One outlined the best programming courses for data science, one outlining the best stats courses for data science, and then the most recent one, which your “Data Science A-Z” course is included as the number one choice, is the best intro to data science courses, where you’re applying the data science process. And those courses, thanks to a little help from our friends at Free Code Camp, we have a big publication on Medium.
They picked up those and brought about 200,000 views and approaching 5,000 recommends. And it’s really nice to see people really appreciate the content and the work you put out there. They’re learning stuff, they are taking suggestions. It can be overwhelming. There are so many options out there, but it’s really nice to see that people are actually taking these courses and learning and bettering themselves and finding their passion, hopefully. So that’s one thing I’m most proud of.
Kirill: That’s awesome. Yeah, I can completely agree with that. These articles are quite amazing, so everybody listening to this, we will put the links in the show notes. For some reason, there’s only 2 out of 3 linked on your LinkedIn.
David: Yeah.
Kirill: I think the programming one isn’t there. So, either through David’s LinkedIn profile or through the show notes for this episode, definitely check them out, and reading them you can learn quite a few things. And David, what would your recommendation be on how to best use these articles? What should people do with these articles to get the maximum benefit and actually start learning data science as you recommend?
David: Yeah, in each article there is probably 25-30 courses in each one, and it’s hard to pick the best one. Several of them are relatively close in terms of actual star rating, like several of them are above 4 out of 5 stars and some of the reviews are very subjective. So pick what kind of style you like the best. If you want to learn a certain tool—like you were saying before, if you really want to learn R, maybe pick a course that teaches R rather than one that teaches something else. Just tailor to yourself. I’m just recommending the ones that I think are the best for most people, but there are definitely lots of great choices in those articles.
Kirill: Yeah. Guys, check them out. There’s definitely some very powerful articles there, and lots of discussions, Seriously, 200,000 views? That’s a lot of views. That’s awesome! And to wrap up, I’ve got a couple of interesting questions. What is your one most favourite thing about data science, something that you’re super excited about?
David: Yeah, so my favourite part is the predictive aspect. I don’t know why, but finding new insights and predicting things that you can’t just interpret by yourself or just get an idea of without looking deeper into the data is something that gets me going. So I would say predictive aspect.
Kirill: That’s awesome. I totally agree with that. And from what you’ve seen in your journey into data science, where do you think this whole field is going? What are the most important things people should focus on to prepare themselves for the future of data science coming in 2, 3 or 5 years from now?
David: Yeah, one thing that I’m currently working on and that I didn’t realize was really important going in is the software engineering aspect. Just being able to make your code reproducible and having the best practices for testing your code and debugging. All of those software engineering skills that I think a lot of new data scientists may not have going in if you didn’t study computer science, for example. So that may not be what the future is, but I think that’s a really important part that can’t be overlooked.
Kirill: Yeah, okay. Thank you, that’s very valuable. And I think as well, that with the advancement of things like artificial intelligence and machine learning, data science skills combined with software engineering skills are going to be more and more valuable because people will start wanting — you’ll still be able to perform analytics, that will still be there. But this space of creating something that involves data science and machine learning, but is an end product that is something that you’ve coded from start to finish that can be just launched and used on its own and work as separate standalone program, I think that will be become more and more valuable, so that’s a very good point there that you’ve mentioned.
Thank you very much for coming on the show. You’ve shared some very interesting thoughts and I think a lot of our listeners have something to think about, to ponder on, and something that will inspire them to learn more about data science, take more control of their education, not be afraid to do these online courses as opposed to just courses at universities. And how can our listeners contact you, find you, follow you in case they want to find out more about your career, read some more of your fascinating articles and just get in touch with you and maybe ask you some questions?
David: Yeah, thanks. My website is davidventuri.com. All my social links are on there. I use Twitter and my e-mail is there as well. Those are the two main ones. Those articles are hosted on my Medium blog, which is also linked on that website.
Kirill: Okay, we’ll include all of those links. And are you happy for people to connect with you on LinkedIn as well?
David: Yeah, for sure.
Kirill: Awesome. Guys, reach out to David, get in touch and network with him. And one final question for you today, David: What is your one favourite book that you can recommend to our listeners to help them become better data scientists?
David: My one recommendation is an intro to machine learning book. It’s called “An Intro to Statistical Learning with Applications in R.” That last part isn’t necessarily that important, being R. It’s just the tool they choose to use. I find it to be a really easy intro to machine learning. It’s well reviewed on Amazon and it’s probably the best intro to machine learning book, from my research at least. I’ve enjoyed that, and I would say if you’re looking to get into machine learning and you’re maybe daunted by the technicality of it, I’d say that’s a good start.
Kirill: Fantastic. Thank you very much. “An Intro to Statistical Learning with Applications in R.” Guys, check out that book. And once again, David, thank you so much for coming on the show. It’s been a great pleasure and thank you for sharing all this knowledge with us.
David: Yeah, thanks for having me. It was great to chat with you.
(background music plays)
Kirill: So there you have it. I hope you enjoyed this podcast and hopefully you picked up quite a lot of interesting things. As you can tell, David has looked at myriads and myriads of these online courses. He’s studied them, he’s examined them, he’s critiqued them, he’s selected from them and therefore all of the insights that he shares inside his guidelines are extremely valuable because they’ll save you the time of looking for the best courses available.
So definitely check out their website class-central.com where you can find the best online courses from top universities and top instructors. Also, make sure to follow David on Twitter, connect with him on LinkedIn, check out his blog. We’re going to link all of these in the show notes at www.www.superdatascience.com/35 and there, in addition to that, you will find all the resources, all the books and courses mentioned, as well as the show notes transcript.
And one final request at the end today: Do you know somebody who is learning data science? Do you know somebody who wants to break into this space? If you do, then send them this episode. Send them this episode and help them understand how to do it better and help them learn from David’s experiences, mistakes and wins and help them build their own pathways into data science and create a career for themselves. Help out your friends, send them this episode. They can get a lot of valuable information from the things that we’ve discussed today. And on that note, thank you so much for being here. I really appreciate you and I can’t wait to see you next time. Until then, happy analyzing.