Kirill Eremenko: This is episode number 301 with Data Scientist Ayobami Ayodeji.
Kirill Eremenko: Welcome to the SuperDataScience Podcast. My name is Kirill Eremenko, Data Science Coach and Lifestyle Entrepreneur. Each week we bring you inspiring people and ideas to help you build your successful career in data science. Thanks for being here today and now let’s make the complex simple.
Kirill Eremenko: This episode is brought to you by DataScienceGO 2020. This Sunday we just finished DataScienceGO 2019. A very exciting time. I met a lot of people. We had over 600 people actually attend. Actually not just registered but actually checked into the event. We also had dozens of exciting speakers, multiple workshops and lots and lots of fun meeting and networking with each other. So if you are up for DataScienceGO 2020, then make sure to grab your tickets already. This week we have a super early bird promotion going on. You can get your tickets with a whopping 80% off. Just head on over to datasciencego.com and you can grab your tickets there.
Kirill Eremenko: The event is going to be bigger, faster, stronger, better next year. It’s going to be even more exciting than this year. So you don’t want to miss out. Secure your tickets and come and meet us next year so we can all grow and learn together. Can’t wait to see you at DataScienceGO 2020. Grab your tickets today at www.datasciencego.com and I’ll see you there.
Kirill Eremenko: Welcome back to the SuperDataScience Podcast ladies and gentlemen. Super pumped to have you back here on the show today. Today is a very special episode, not only because we are kicking off the next hundred episodes, this is number 301, but also because this episode was recorded at DataScienceGO 2019. So, DataScienceGO 2019 just ended. This is, today is, I’m recording this on the final day and a few hours earlier, I sat down with Ayo and we had this fantastic chat. So in today’s episode what you will find out, first, Ayo will share some takeaways with you from DataScienceGO 2019.
Kirill Eremenko: So if you are unable to attend this year, this is going to give you some really cool insights into some of the takeaways that he got. I think they’ll be very valuable for you too. So there were three main topics we discussed, which were productization of data science products. Number two was data science teams and you’ll find out about three types of data science teams. The innovator team, the partner team and the engineer team. Then we will talk about building character. What you do when no one’s looking and also about resilience and how that is important in a career.
Kirill Eremenko: So those are the three main topics that came from DataScienceGO. Also, you will learn about Ayo’s background and how he went from a project manager, from a project management backgrounds. So he’s PMP certified to a Data Scientist and what it took, how much he sacrificed, what incredible careers he sacrificed to make that move into data science. You’ll find out why. Also we will talk about, towards the end, we’ll talk about project management in data science. So there you go. A very interesting podcast coming up. Can’t wait for you to meet our guest today. So without any further ado I bring to you, Data Scientist Ayobami Ayodeji.
Kirill Eremenko: Welcome to the SuperDataScience Podcast ladies and gentlemen. Super excited to have you on the show. Today we’ve got a very special guests, Ayobami Ayodeji. Ayo, how you going?
Ayobami Ayodeji: Good, good.
Kirill Eremenko: Good. Where are we Ayo?
Ayobami Ayodeji: We are at DataScienceGO. Which is amazing. It’s been an amazing experience so far.
Kirill Eremenko: Day two, right?
Ayobami Ayodeji: Yeah. Well actually day three.
Kirill Eremenko: Day three, oh yeah. Were you at the workshops?
Ayobami Ayodeji: I was at a workshops. Yeah.
Kirill Eremenko: Oh, yeah, cool. You were at the Tableau workshop as well?
Ayobami Ayodeji: I was at the Tableau workshop. Yeah.
Kirill Eremenko: That was really good. There was quite a lot of people.
Ayobami Ayodeji: That was interesting for me, because I actually do use Tableau literally every day at work. So it was obviously, you wanted it to be basic. Right? People that have their experience, but-
Kirill Eremenko: And I warned everybody. I was like, this is going to be a basic workshop. If this is too easy for you, go to the second workshop that’s [inaudible 00:04:38].
Ayobami Ayodeji: Yeah, it was good. I think it was a good thing I was there because I was actually able to help people around me [crosstalk 00:04:46]. People who have never used Tableau before, so that was fun. It’s always nice helping people. Right?
Kirill Eremenko: I love that. I love that. I was expecting like only around 20 people to show up. I think we had like what? 40, 50 people?
Ayobami Ayodeji: Yeah, over 40 people.
Kirill Eremenko: Of course when you have a room of that size, even though I tried to do a basic workshop, there are people who are going ahead, falling behind, I really enjoy that you, Ogo and some other people we’re jumping up and actually helping out others. Jonathan was also helping out and that was really inspiring. So instead of really just sitting there and being like, “Oh, okay, I’m way ahead of everybody.” Everybody was helping each other. That was so cool. What made you do that? Why did you volunteer to help others?
Ayobami Ayodeji: Honestly, one of the big things I love about data science is that everyone is willing to share. Like, just think about open source, that one word, there’s so many organizations putting in a lot of money into developing new products and just sharing it with the world, right? Like you have people with a GitHub profile writing a lot of different really innovative code and just sharing it. So just the fact that data science in general is a very sharing profession.
Ayobami Ayodeji: That’s one of the things I like about data science and that’s one thing I see myself doing. I’m not an expert in data science. I just got into the field not very long ago, but being able to help people even at this level, makes me feel very good.
Kirill Eremenko: That’s awesome. Why do you think that is? This is, not to say there’s no other fields like this, I’m sure there are, but I don’t often see in general a community that is, everybody wants to help. Why do you think that is?
Ayobami Ayodeji: I think a big part is, a lot of people in data science don’t have a computer science background. I feel like it’s a really good entry point for people that are not from the technical background to get into a technical kind of role. So in order to do that, people need a lot of help. I feel like a lot of the organizations that are promoting data science and trying to democratize data science and even with the individuals that are doing that probably got a lot of help from other people in the first place. Right?
Ayobami Ayodeji: So it’s kind of a classic case of painting forward. You know you got something good and you actually want to help out other people as well. I think that is one of the big things. Also it just feels really good helping. Actually putting any work into something and getting other people to use it just pretty much increases the impact that you’re able to make. Because a lot of data scientists, their main goal or one of the big goals is to create impact. So the more people that use your code, the more impact your work actually has.
Kirill Eremenko: That’s true. That’s true. You get a sense like eventually a sense of contribution when you help someone feels good.
Ayobami Ayodeji: Yeah, it does.
Kirill Eremenko: Interesting. So, DataScienceGO, how are you feeling?
Ayobami Ayodeji: Really good.
Kirill Eremenko: It’s three days.
Ayobami Ayodeji: Yeah, three days. Learning a lot for sure. I’m learning quite a bit of technical stuff, but also some of the subscales, right? Learning a little bit more about some of the things that you need in data science. I think one of the big ones for me was how to easily put data science products into production. There’s actually been a few lectures about that, pretty much focused on that. Then also-
Kirill Eremenko: One of them was by John Pitch from Amazon.
Ayobami Ayodeji: Yeah. From Amazon. Yeah. He was talking about one of the new Amazon product that allows you to pretty much simplify the process of production lies in your models and stuff. There was another one when we just had someone from Google, he talked about-
Kirill Eremenko: Someone from Google was talking about also production?
Ayobami Ayodeji: Yeah. So, and he said something about Kubernetes and some of the technology. He was pretty much encouraging us to go out there and do some more research on how to production… Like automate the process of production that lies in your code. So yeah, that was pretty good.
Kirill Eremenko: Definitely taking notes?
Ayobami Ayodeji: Oh yeah, definitely. Lots of notes. Lots of notes.
Kirill Eremenko: That’s important. You’re still making notes.
Ayobami Ayodeji: Yeah. Actually, that’s actually one thing I like to mention to people listening on this podcast. In the past I didn’t really use to take a lot of notes and I noticed that not a lot of people actually do that. I actually noticed that in this conference as well. Not a lot of people were actually taking notes, but notes are really important. Because the next day you’re trying to think about what you learned, but honestly, there’s no way you can remember everything, right? But when you take notes-
Kirill Eremenko: How many talks did you attend yesterday or on Saturday?
Ayobami Ayodeji: A lot.
Kirill Eremenko: At least like seven.
Ayobami Ayodeji: Exactly. How do you even start? How do you even begin to remember the stuff that you’ve learned? Right? So taking notes is really important. If you only take one thing out of this episode, always take notes. It’s so worth it.
Kirill Eremenko: I couldn’t agree more. It’s not only is it something to go back to and look at, but also while you’re writing, you remember better. Your retention. If you don’t take notes, I heard somewhere that like is about 20%, as soon as you start writing, it goes up to like 50 to 70% retention.
Ayobami Ayodeji: There’s actually this concept of dual encoding of your learning process. So, for example, if you’re reading something and at the same time you’re listening to it or you’re listening to something and you’re writing it, it’s two different ways for your brain to… It’s probably maybe two times. So maybe a lot more than two times I should be retaining what you’re trying to learn. So one of the ways to improve your learning or speed up your learning is to learn it using two different ways or multiple different ways. Taking notes, just like you said, is one of the ways you can actually do in code while you’re learning.
Kirill Eremenko: Okay. That’s very cool. All right. So productization of data science, that’s an important topic. A lot of data scientists don’t really think about that. It’s like, okay, project insights, done. But actually if you build a model that needs to go to production… I always struggled with this word at the start when I learned it years ago, production, basically means… What does production mean in your role?
Ayobami Ayodeji: Well, production is pretty much putting it in a kind of state where it can be used over time. So, for example, depending on what the business case is, for example, if it’s a machine learning solution within a software, being able to take that model that you built and put it into a software. Or if it’s for an organization, being able to put in in some kind of servers so people can actually use that model to make decision over time. [Crosstalk 00:11:49]
Kirill Eremenko: Sometimes data science is one of analysis, you give insights, great. But if you develop a model that needs to be, for instance, a churn model or a segmentation model that needs to be used by the business daily or weekly or monthly, that needs to be put into production environment. Basically a server that is governed by the IT department and your model is sitting there. They make sure it’s working, the data’s going in, it’s coming out, everything is working fine, it has its run time during the night and things like that.
Kirill Eremenko: It’s a whole different story, right? Doing the project is, I don’t know, it might take you two weeks, but they don’t take another two weeks to put that-
Ayobami Ayodeji: Into production. Exactly. Another thing that I really learnt from this experience was the fact that another thing you want to consider apart from making it production ready is also making it in a way that you can actually redeploy. Because a lot of the time your model is probably going to degrade over time. So having that strategy to redeploy or update your model is also really important.
Kirill Eremenko: Yeah. Thinking about that in advance.
Ayobami Ayodeji: Yeah, exactly.
Kirill Eremenko: So what kind of strategies did you learn for redeploying for updating models?
Ayobami Ayodeji: Well, I can’t really think of one just up in my head, but I think one of the first steps is to recognize the importance of doing that because obviously over time the models degrade, but also… So I think one of the speakers actually talked about coming up with a system or a pipeline to whenever you get new data, put it into the infrastructure and making in a way that you can actually retrain your model based on the new data. So there’s various ways you can do it. You can do like a batch process.
Ayobami Ayodeji: So say, for example, you get X amount of data, you can retrain using that batch or you can actually make it live if you want. So there’s different strategies for doing that. Obviously data science is a research field. You need to figure out what the best strategy is for your use case.
Kirill Eremenko: Yeah. For sure. Got you. Not a lot of people don’t think about that either, that the model’s deployed but over time…
Ayobami Ayodeji: Exactly. Oh, and also a continuous monitoring. So say, for example, how do you know when to redeploy it though, when to improve it, right? If you’re working in an organization where you’re building hundreds of models, how do you know which model to focus on to redeploy? So coming up with a way to monitor your models in a really easy way, make a visual. Figuring out which models you need to focus on is another key thing. So that’s something else you need to consider when you’re making these data science projects.
Kirill Eremenko: Yeah. Okay. What are some of the reasons why models can deteriorate? Did they mention anything about that?
Ayobami Ayodeji: Yeah. Sometimes the data, the fundamental structure of whatever the data source is. Maybe, for example, you’re looking at customer data, if the customers changed over time, the model that you had that you trained on isn’t really applicable to the new situation. Right? So that’s one of the examples. So whenever there’s a change in the data set or say, for example, you need to add more features, then that’s another reason, right? So you get new features that pretty much changes the structure of the model. Yeah, there’s a lot of different reasons why models deteriorate.
Kirill Eremenko: Yeah. So if you’ve changed the features, new customer, so old customer’s going away, new customer’s joined, they might have different or you might still have the same customers, but what I’ve seen before or like I’ve discussed with data scientist when I was in industry was when… This is just a general change in the preferences of the population, right? Like for instance, your customers are the same. I worked in superannuation, like a pension fund where your customers don’t change that much or if they’re with you, they’re with you for a long time unless you really provide bad service, which is rare.
Kirill Eremenko: So customers really rarely move around there. But the thing is there might be a general change in preference of like, okay, that was a trend before. I don’t know people would want to behave or purchase certain things. Now it’s just normal. Like fashion has changed or lifestyle-
Ayobami Ayodeji: Sentiments changed. Exactly.
Kirill Eremenko: Legislation could also change, right? If your industry’s affected by legislation like for instance in my industry, if they changed the date, the age when somebody goes to pension, all of a sudden now you have more customers or less customers or they behave different again. So lots of reasons can affect you model. Well cool. What other talks did you enjoy?
Ayobami Ayodeji: I also liked the talks that were revolved around a data science team. For example, some of the things that it’s expected from a data science team or the different structures of data science teams. So I learned a lot about that. I think one of the big things that I’m trying to focus on in my career is to implement better ways of doing data science. So my background is in engineering and product management. So just part of the process of trying to figure out how to leverage my strength in within this industry is, so I’ve put my new experience learning how to take different things and put it together into one deliverable or one project goal.
Ayobami Ayodeji: I kind of want to be able to do that kind of stuff with data science as well. So I learned about different types of data science teams and different expectations, different ways to communicate with management and different ways to promote data science. How to be a better data-driven organization. There’s a lot of things that you don’t consider. So you decided, okay, everyone’s making decisions based on data, right? But how do I even start? How do you take your organization the way it is right now and make it a more data-driven team? Data driven organizations.
Ayobami Ayodeji: So some of the things to consider, for example, you need to make it really important and get people to understand that the data is a field, right? You don’t want to lose the data, the old mentality of trying to get rid of as much data as possible to conserve space. In this day and age, it’s really, really cheap to store data, right? So data is an asset, so getting your organization to understand the importance of data and then also finding ways to democratize the analysis od data. Getting more people more familiar, more comfortable using data products.
Ayobami Ayodeji: So, for example, Tableau, right? How do you train, get people into using tools like that by themselves, not just depending on the data science team, but actually democratizing the analysis of data. That’s another big one. Then one of the big things that they mentioned was the importance of having it come from the executive level because you need support from the top. Data science is a science, right? A lot of the time at the beginning you have no idea whether or not the product is going to be successful, right? Or you don’t know how long it’s going to take.
Ayobami Ayodeji: So having that support from the top pretty much takes some of that risk and some of that pressure from you because you don’t always have to answer, okay, so what did I do last week? On the other hand, you need to consider the concept of minimum viable products. That also helps, right? It also helps build confidence because if you’re able to come up with some kind of simple model earlier on, coming up with a strategy to build on that also helps. But being a data driven organization is really important to have that support from the top because that goes a long way.
Kirill Eremenko: Wow, very cool. You mentioned that… It looks like a lot of [inaudible 00:19:58].
Ayobami Ayodeji: Yeah, a lot to offer, for sure. I’m looking forward to going over my notes again, but.
Kirill Eremenko: That’s awesome. You mentioned two… Oh, not two, you mentioned a few, you learned about a few types of data science teams. Could you share a bit about that?
Ayobami Ayodeji: Okay. So one of them, for example, is, that she mentioned was the innovator data science team, which is more-
Kirill Eremenko: Is this Michelle Kein?
Ayobami Ayodeji: I think it just happened. I don’t really remember how many today. Actually it was actually still happening when I left. But so the innovator data science team involves research, right. So that’s pretty much, I think that’s where a lot of the PhD data scientists come in. So if they’re responsible for coming up with the next generation new algorithms or new kinds of products from some of the latest guard themes coming from the papers. So once all of the biggest skills that they need for those or like advanced machine learning skills, machine learning and stat skills and also cloud computing skills.
Ayobami Ayodeji: Maybe they need a little bit less of data visualization, storytelling. They need a little less of being able to work with the product management teams, but they need to have that strong technical skill and to be able to innovate. Then there’s the other one, which is I think is the partner structure where you actually work pretty close with the product managers within the organization and you’re responsible for providing sometimes ad hoc but also designing things like A/B testing for…
Ayobami Ayodeji: So say, for example, the product manager is trying to figure out what feature should we implement within this? Right? So being able to help them with things like A/B testing making predictions to help them make the decision. So this is, I think this is probably where most of the data science teams are. You’re sitting right next to the product managers and supporting them with their requirements. Then I think the third one was, I think it was more like a data engineer project management team. Where you work closely with the engineers. So the software engineers, for example.
Ayobami Ayodeji: So your partners are the software, maybe the software engineering managers, so you work with them closely. You build models that are easy to deploy like I mentioned before. You’re probably more involved with software development part of things. So, you need to pretty much think about your organization and what your needs are and then based on that you can build a data science team.
Kirill Eremenko: It sounds like a talk more for executives or leaders. Well, what value did you get out of it?
Ayobami Ayodeji: I got a lot of value because like I said, I’m trying to see how I can build… For my product management experience, how to better leverage data science within my organization at a high level. Right. So learning about the different ways you can have data science teams was actually really good for me. I think a lot of executives, I think more and more executives are reorganizing the importance of being data literate. So a lot of executives know a lot of the stuff that I’m learning now. But my experience involves a lot of implementation and making things happen in environments that are not really used to those kinds of processes.
Ayobami Ayodeji: So learning about different ways we can incorporate data science within an organization is really important for me. Maybe not as much deep technical stuff because there’s other ways of learning technical stuff. But learning that high level management kind of things that you need in order to actually build a data culture and build a data different organization, is really important to me. So that’s why… This has been a really good conference for sure.
Kirill Eremenko: Nice. You also mentioned before the podcast that you enjoyed the, there was a motivational talk at the start by Antonio Neves. Tell us a bit about that. Like very different, right? I think he actually, he doesn’t… So he was the keynote speaker for today and he doesn’t have a data science background at all. He’s more like of a career background. What did you learn from him?
Ayobami Ayodeji: So I learned a loved about the importance of what do you do when no one’s looking? What do you do? When you think about that, the first thing you think is what does it matter what you do on the one’s looking? You know what I mean? But the truth is a lot of the time there’s people actually watching you. I think the real question is what do you do when you think no one’s looking at you? Because you never know who’s watching you. You never know what opportunities are out there.
Ayobami Ayodeji: A lot of people, we all try to think that we don’t judge like we said, “Oh, I don’t judge people based on their parents or whatever.” But the truth is that’s what happens, right? That’s just a reality. We can try to fight it and be like, I don’t judge people or people shouldn’t judge people based on their cover or whatever. But that’s just the reality. So the way you present yourself, the way you look-
Kirill Eremenko: The way you talk.
Ayobami Ayodeji: The way you talk-
Kirill Eremenko: The way you dress.
Ayobami Ayodeji: The way you dress, exactly.
Kirill Eremenko: How you approach people.
Ayobami Ayodeji: Approach people, exactly. So those things are really important. So the question is what do you do when you think on one is looking? So are you that guy that showed up on time or are you that guy that didn’t show up? You paid for this conference, right? You probably went out last night, probably didn’t wake up on time to get to attend that keynote speaker, for example. So no one’s looking, some people got their organizations to pay for this, but probably the initial up for the morning, the very first discussion kind of thing. You think no one’s looking, but like you never know what you can learn.
Ayobami Ayodeji: So I think the main key was, okay, who’s, even when there’s no one looking at you, right? Even if it does happen and no one’s actually watching you, it’s really important to develop yourself. So some of the things I do like for example, attend those lectures or actually go out of your way to come to conferences like this or some of the stuff that you do to actually do learn machine learning, no one really sees when you’re doing that. But sometimes there’s other things that really matter because, it’s one thing to be in an organization and do all this fancy stuff, those are the things that people are looking at, right?
Ayobami Ayodeji: But how do you even learn or get to be able to do that stuff? It’s the stuff that you do when no one’s looking that allows you to do that. So, for example, for me, I know I don’t come from a computer science background, for example, but the amount of work that I had to do in the background and no one knows is pretty much what helped me be here, for example. So just knowing that the opponents of, you never know who’s watching, you never know what you could possibly do with that could possibly change your life. You never know what lecture could possibly change your life.
Ayobami Ayodeji: So those are some of these things that he talked about. He also talked about the importance of resilience and grit. So say, for example, if you fail that something, what’s your immediate reaction?
Kirill Eremenko: If you fail.
Ayobami Ayodeji: Exactly. If you fail at something, right? So it’s really important in data science and pretty much every other profession that whenever you fail, learn how to pick yourself up because those are really important skills because if you’re not failing, if you’re not doing stuff that makes you uncomfortable, you’re probably not growing. So if you push yourself to the limit you do something different, you try to just push yourself, stretch yourself to that next point to the point that you’re not comfortable, you actually learn a lot from that. Sometimes you’ll fail, but you’ll be surprised at how often you actually learn.
Ayobami Ayodeji: So, one of the key notes was find your edge. Find your edge. If you’re on the edge of the cliff, you’re super uncomfortable, but if you’re not on your edge, you’re probably taking too much space. So that was one of the things that he said. If you’re not on your edge, you’re probably taking too much space. So find that edge, find that next thing that can actually take you to the next level, and you will be surprised at some of the amazing things you can do. I personally, I’ve done a lot of that.
Ayobami Ayodeji: I’m finding that I’m learning a lot more about things like this recently. This is actually the third conference I’m attending the last two weeks. Yeah. I know. It’s not so much about the technical skills to be honest. Like they were all tech, well actually the two of them were tech conferences, but it’s not just about the technical skills that you learned from these, but some of the soft skills also. A lot of this stuff that I’m learning actually really resonate with me because those are a lot of the things that I’m already doing.
Ayobami Ayodeji: But being able to come through a place like this and actually structure, have a structure for that, is really going a long way. It’s really good to know that some of the things that I’ve been doing are the right things. But knowing the right things to do now I know that maybe I need to keep doing this, keep finding that edge, and hopefully we’re all successful.
Kirill Eremenko: Yeah. Wow. Fantastic. I love both of those and doing things you’re uncomfortable with failing and the first one where you, what do you do when no one’s looking? It’s like, for example, if you are a person who, I don’t know, leaves your clothes lying around your apartment when no one’s looking, you know you’re going to be by yourself for like a week or so, then that just becomes part of your character. Then when people are looking, you’re going to have to put an extra effort to clean up and make sure, you just got to know that about yourself, that you’re messy.
Kirill Eremenko: In terms of more a data science related example. For instance, when I’m coding something, I make sure to follow a certain convention that I’ve developed for myself of how I code, of how the code is structured, that it’s clean, that I put comments, that I… It might, may take me extra effort to do it properly and not take shortcuts, but I forced myself as much as I can to do that because I know that builds my character and even though no one’s looking, that’s just how I’m going to code in the future.
Ayobami Ayodeji: Yeah, exactly. That’s really good. That’s a good point actually.
Kirill Eremenko: Yeah. So it’s, there’s two parts to it I guess. Personality and character. Personality is how others perceive you, which is important, especially in this day and age where we live mostly in big cities and we meet lots of people at a time and there’s just no way around it. You have to have a good personality in order to, for people to want to engage with you. But what comes out to the top after you start building a deeper connection is your character.
Kirill Eremenko: Something that was really valued back in the 19th century when we lived in small cities and everybody knew each other and character was the most important thing. Now it’s less valued, but it’s still there. It will come out eventually.
Ayobami Ayodeji: It will come out eventually. Yeah, for sure. Honestly, you can’t get to a certain level without a lot of hard work. I’m a strong believer that there’s no substitute for hard work. It’s not all about talent or whatever. People that are not classified as talented have done amazing things from hard work. Right. Just like you said, the character stuff, having that hardworking character really goes a long way.
Kirill Eremenko: Yeah. Like somebody, I forgot who said that a genius is 1% inspiration, 99% perspiration. That’s good. Okay. Ayo thank you for the quick rundown. I think that will be very helpful for those who decide to go to recap on some things and those who want to get some takeaways even though they didn’t make it to the event. But now I wanted to switch a little bit. Tell us about yourself. So you said you got an engineering and product management background.
Ayobami Ayodeji: Project management.
Kirill Eremenko: Sorry. Project management. How and why did you get into data sentence?
Ayobami Ayodeji: Okay. So-
Kirill Eremenko: This was not long ago, right?
Ayobami Ayodeji: This was not long ago.
Kirill Eremenko: How long ago?
Ayobami Ayodeji: Okay. As a professional, two and a half months ago.
Kirill Eremenko: Two and a half months ago. No. When you said two and a half I thought you were going to say years. Wow! That’s so cool.
Ayobami Ayodeji: Yeah. Pretty vanilla here.
Kirill Eremenko: Yeah. But, and as an amateur, like you were learning before that?
Ayobami Ayodeji: Yeah. Okay. So just a bit of my background. So I studied industrial engineering. That was my undergrad.
Kirill Eremenko: Sorry, where? You’re from Toronto, right?
Ayobami Ayodeji: I’m from Toronto.
Kirill Eremenko: You flew here from Toronto?
Ayobami Ayodeji: I flew here from Toronto for this, yeah.
Kirill Eremenko: Wow! That’s so cool, man. That is awesome. Really cool. Thank you. I’m really like humbled that you came all the way from Toronto for this.
Ayobami Ayodeji: Yeah. So, actually, I moved to Canada about 12 years ago for my undergrad. I come from a family that owns manufacturing. So I grew up in a plant. I’ve always been a manufacturing person.
Kirill Eremenko: Where did you grow up?
Ayobami Ayodeji: Nigeria. So I wanted to learn more about manufacturing and making processes, manufacturing processes better. So I decided to study industrial engineering, which pretty much focuses on optimization and making things better, processes, machines, whatever it is. So that was why I studied in industrial engineering. I went to University of Windsor because I wanted to work in the automotive industry. I wanted to work for one of the big three.
Kirill Eremenko: What are the big three?
Ayobami Ayodeji: So GM, Ford and Chrysler. Automotive organizations. So I went to Windsor studying industrial engineering. I got to work at Blackberry for one of my internships. I’d like to mention one of the biggest life changing experiences for me was my internship because I was working at Blackberry, which is, it was always known as an innovative organization. I started there in 2010 as an intern. I remember that when I started, I think it was about a couple of weeks after I started, the Blackberry brand had just gone to, I think it was the first time they made the top 10 most valuable brands in the world or something like that.
Ayobami Ayodeji: That was pretty much a couple of, like a month after I started, and by the time I was done Blackberry was down. So I got to experience Blackberry at the top and Blackberry at the… Just saw the whole organization go from the top to just permitting down.
Kirill Eremenko: How long were you there for?
Ayobami Ayodeji: I was there for a year and a half.
Kirill Eremenko: In a year and a half? Wow!
Ayobami Ayodeji: In a year and a half. That was like-
Kirill Eremenko: On top of the iPhone.
Ayobami Ayodeji: Yeah, pretty much. Yeah. That experience you cannot find anywhere else. It was just so life changing. Obviously the reason why was because, I think one of the big reasons, my opinion, was that they weren’t focusing as much on the customers as they should have and they weren’t innovating like the other organizations. Right. So I got to see what happens when you don’t innovate, when you don’t push yourself, when you don’t live on that edge as much as you should. So that was a big life changing experience for me.
Kirill Eremenko: They were taking up too much space.
Ayobami Ayodeji: They were taking… That was pretty much what was happening. They were taking up too much space and this phase just got sucked away from them. It was amazing seeing that happen. Yeah. So pretty much that’s why I think that’s probably the biggest reason why I’m so driven right now. I’m always trying to find the best, how to improve myself every day because I don’t want to be Blackberry. So I finished my internship, finished my undergrad, and then I got my dream. I got to work for Chrysler.
Kirill Eremenko: Wow! There you go.
Ayobami Ayodeji: So I worked there for a year and a half. Well, two years actually. One thing, one of the big things that I learned, I learned a lot about being a professional and all that kind of stuff, engineering, optimizing process. But one of the big ones that I think were more transferrable was the people skills. Because my job was to make processes more efficient. In an organization that was unionized, I didn’t like, the people on the floor didn’t really like that because it meant that they had to do more work.
Ayobami Ayodeji: So learning how to understand people’s personalities, their motivations, and using that to get them to do what you need them to do, even when you’re not in a position of power. I was a new grad. This is the people that have been there for years and years, like decades. So I had to learn how to deal with different kinds of people and get them to do what I needed them to do. So that was my experience at Chrysler. Then I started working at Magna, which is one of the largest auto-
Kirill Eremenko: Before we move on to that, what’s your one biggest tip on how to… Or like I love how you put it. Getting people to do what you want them to do while you’re not in a position of power? What’s your one biggest tip for someone? Because in data science that’s often the case we face when dealing with executives or managers or other line managers or team managers in other divisions that don’t understand data science, don’t understand statistical significance, maybe don’t understand the implications of the insights that we convey. Sometimes we need to convince them to do certain things, but we’re not in a position of power. So what’s your one tip on how to do that?
Ayobami Ayodeji: Okay, so I think one of the big things to start is the way you present yourself. Right? I think the way you present yourself goes a really long way. So present yourself in a way that number one, people feel comfortable with and also so that… That’s the one thing at the beginning. Then the next one is try to really empathize with that person. I try to understand what they’re feeling and what their motivations are and try to figure out which one of those are actually valid and which ones are not.
Ayobami Ayodeji: Learning how to explain in a respectful way why some of their concerns are not exactly valid. Then using that to get them to understand that, okay, maybe we can just try this. Let’s try this, let’s see what happens. Right. It’s also important to have that concept of having a good reputation. Because if, for example, you’ve done some optimization or some work in other places, it’s easier to get people to believe you and actually tune into some of your ideas because having a good reputation really helps. So I think those are the main things that I think can work in any situation.
Kirill Eremenko: Okay. I really liked the one you said dispel their fears and concerns. The ones that are not valid, help them get rid of them. That’s good. That’s powerful. Okay. So what happened next?
Ayobami Ayodeji: So, then I went to Magna. I wanted a role that allowed me to have bigger responsibilities. So I was a process engineer. I was responsible for optimizing a bunch of stuff. For example, how do we get materials in order to make a car, you need to make the parts, but in order to make the parts that Magna makes, you need to get the components from somewhere else. So how do we package those components the best way considering a bunch of factors? Then how do you get them from the supplier to a storage location within the plant and from the certification to the production floor?
Ayobami Ayodeji: So pretty much streamlining that process, considering a whole bunch of constraints. Now it’s been watched my job. So there again, I got to apply the knowledge that I got from Chrysler because some of the products I’ve worked on were quite transformative. There were completely different way of storing materials and moving materials through the plant. So I learned a lot more about implementation, like how do you implement a large scale change management kind of project. Right? So I did that and then while I was doing that, I decided that I wanted to go into product management.
Ayobami Ayodeji: So I did a one year program while I was working full time at the University of Toronto in the project management. Then I got my PMP certification, Product Management Professional.
Kirill Eremenko: Oh, congrats. That’s awesome.
Ayobami Ayodeji: So as an engineer, I got my PMP and then I got bored. So, okay, so now I have my PMP. I finish-
Kirill Eremenko: You got through this, what’s it called? The astronautics. [inaudible 00:41:46].
Ayobami Ayodeji: Yeah, I go on PMP. And I was like, “Okay, so what else do I want to do? I have all this free time. I need to do something productive.” I’m like, “Okay, so what’s the next thing I can do? How about a master’s degree?” Then I said to myself, that’s crazy. How are you going to do that? Then I decided to myself that I was going to do it. So the next question was master’s degree in what? So I have an industrial engineering background. It made sense to do a master’s in industrial engineering. But I wanted to do something that would allow me to be useful in any kind of organization.
Ayobami Ayodeji: My undergrad was more manufacturing related. So I wanted to be able to do something that applied to a broader kind of sets of organizations. So I have a few friends in data science and one of my friends actually guilt trip me into the introduction to data science course.
Kirill Eremenko: How’d he do it?
Ayobami Ayodeji: She pretty much guilt trip me into doing it, because she was-
Kirill Eremenko: How do you guilt trip somebody to do a course?
Ayobami Ayodeji: Like, “I can’t believe you’re going to make me take this course by myself.” So I took the introduction to data science and I also took the business process management course, which had to do with taking data and converting it into a business process. So automating the process of taking log data and converting it into actual business process. We don’t actually explicitly defining the process. So that was a pretty good course.
Kirill Eremenko: Where did you take these courses?
Ayobami Ayodeji: University of Toronto. So my master’s degree was in university of Toronto.
Kirill Eremenko: So, while you were doing your masters in industrial engineering, you took a few courses in data science?
Ayobami Ayodeji: No. So, okay. So my master’s was in industrial engineering, but at the University of Toronto, there’s a lot of different options within the masters-
Kirill Eremenko: So like the electives?
Ayobami Ayodeji: No actually there’s a lot of emphasis. So you have an emphasis in analytics and have an emphasis in advanced manufacturing and healthcare engineering, financial engineering. So I pretty much took a couple of courses just to see what I liked to figure out what to emphasize on. So I took the introduction to data science and I decided that that was what I wanted to do. So I focused on analytics. So I did this while I was working full time. So usually when people are working full time, they usually go with the part-time option to do their masters.
Kirill Eremenko: The longer version.
Ayobami Ayodeji: The longer version, exactly. Takes about three years. But I’m a little bit crazy, so I decided to do it a little faster because I just wanted to get it done a lot faster so I could move on to the next, whatever it was for my career. I saw the value of data science in all organizations. I didn’t know exactly where I was going to land, if I was going to stay in my current organization or not. So I pushed myself. I pretty much set that goal and I pushed myself to make it happen. So my original plan was to finish my master’s degree in two years and then I got a job in data science.
Ayobami Ayodeji: It was a campus recruitment job. So usually when you’re in school you have these large organizations come and recruit students that they want. So I got a data science job really early. So I had to cut my timeline from two years to a year and a half. So I had to figure out how to make that work. So I set this goal and I focused. I think what really helped me was my focus. I decided that I had this goal. I was already working in taking a lot of data science courses. This was really important to me. I felt like this was something I was passionate about.
Ayobami Ayodeji: So I set that focus for the deadline for myself and decided that I was going to do it no matter how difficult it was, I had to figure out how to do it. I wasn’t sure, I didn’t know what was going to happen. Things change. But I decided that, you know what, I have this opportunity. I do not want to lose this opportunity just in case that was actually what I wanted to do. In order to do that, I needed to finish it in less time than I thought. So just like a science, any kind of science project, you never know if it’s possible, but you put yourself on the edge, find your edge, make it happen. So I decided to make it happen.
Ayobami Ayodeji: So I focused on that, finished it in a year and seven months. So that was a really good experience for me because I got to learn a lot about myself, how to prioritize, how to not listen to naysayers, because a lot of people that thought I was crazy but also had-
Kirill Eremenko: This is while working full time?
Ayobami Ayodeji: This was while I was working full time. It actually gets better because while I was working full time and doing a lot of work when no one was looking, I got promoted. So not only was I working and doing my classes, but actually I was able to be effective enough to get promoted to product manager. So that was how I got officially into product management. So it was a lot of prioritization and working really, really hard, doing a lot of things that people can’t see. People didn’t really understand why I was doing them, but it was an amazing experience because not only did I learn the technical skills for data science that I needed to launch my career, I also learned about time management.
Ayobami Ayodeji: People being able to prioritize things. Being able to gain that credibility that allows the leadership team that I worked for to give me the flexibility that I needed to be successful. So like these are some of the things that they probably wouldn’t let just anyone do, but they allowed me to take courses, for example, during the working hours because they knew I had that integrity to actually make it happen. So I think the key takeaway for people listening is that have that integrity, have the right character and have the right mindset.
Ayobami Ayodeji: If you have those, there’s nothing you can’t do, because people will actually believe in what you’re trying to do and actually support you for that. So if I could do it, you can do it too. So yeah, I finished my master’s degree and I decided to take the job going into-
Kirill Eremenko: What was the job? Who offered you the job?
Ayobami Ayodeji: So I work as a data science associates at CD, TD Bank. So, TD Bank is one of the-
Kirill Eremenko: Toronto Bank?
Ayobami Ayodeji: Toronto-Dominion Bank. Yeah. So it’s one of the largest banks in Canada. So I work as a data science associates. It’s one year quote unquote “rotational program”. It’s supposed to be a leadership program. So you do that for a year and then you move on to hopefully a more senior role within your organization or wherever you end up going. So I felt like it was a perfect start into data science for me because number one, it allowed me to do it, not as a regular employee but as an associate.
Ayobami Ayodeji: So I wasn’t actually tied to the department, but actually got to learn, try different things within the organization. So that was a really good… that’s a really good experience for me. So I started there about two and a half months ago and I’m loving it.
Kirill Eremenko: Congrats man. Such an inspiring story. Wow! So would you say that you gave up all your PMP and project management for data science or is there some continuation in that story?
Ayobami Ayodeji: So, that’s honestly was a very difficult decision for me, because I was actually on the leadership track in my previous organization. I had a really good mentor. He was actually the general manager on my plant, which is like the top guy at that time. He believed in me and he saw a lot of potential, so he actually pretty much put me in this leadership track. So, for example, I got to be a product manager and then when I was done with that I had the next level.
Ayobami Ayodeji: So it was a big sacrifice for me. Also, I didn’t know how good I would be in data science because I was already comfortable and good at my previous job. But I pretty much found that edge. I found something I wasn’t very comfortable about. Like I said, I’m not a computer scientist. I took one intro to computer science course when I was in first year and that was it. So I had no programming background when I started my master’s degree, but I pushed myself while I was working full time and the fact that I was able to do this, I think anyone can do it.
Ayobami Ayodeji: Yeah, so it was a very big decision for me, but I had this long-term goal of being able to do as much, have as much impact as I can and being able to be in the field that was as flexible as possible, where I could apply my knowledge in various domains. I didn’t want to be tied to one industry, for example. So that was the driving decision. That was the main thing that I used to make that decision. I believe that data science has a lot of opportunities for growth.
Ayobami Ayodeji: So even though I was stepping a little bit backwards in the non-leadership position, just the thrill of being able to learn something new in a really good organization was what pushed me forward to take that step.
Kirill Eremenko: Got you. But there has to be also reason, right? As you said, for you it was that data science is more applicable to many industries. Because otherwise you’re just going to end up like next to go to become a chef. Next you’re going to become a professional water skier and you’re just going to change like every three years. But I can see now the reason, like in your project management, you would have been focused in one industry or as data science has the promise of, hey, you can apply this wherever you want.
Ayobami Ayodeji: Exactly. Like I mentioned previously, I’m trying to create a career that has to do with taking the best methodologies, management and different team styles in data science and apply it in an organization. So just having that people skills, that chain management skills from my previous background with my product management skills, I think a lot of that is actually applicable in what I’m trying to do in data science. So, yeah.
Kirill Eremenko: Okay. Very cool. Would you say… So basically you’re saying those are skills that are transferable, right?
Ayobami Ayodeji: Yeah, definitely.
Kirill Eremenko: The project management skills, you’re leveraging them?
Ayobami Ayodeji: Yeah.
Kirill Eremenko: Okay. Very cool. Tell us, can you tell us a bit more how for a project manager that’s listening to this and wants to get into data science or somebody with experience or maybe let’s just put it other way, somebody in data science who hasn’t done the education project management that you’ve done, what is your one biggest, I don’t know, actionable piece of advice on the how to better… What’s a skill from project management that you can share with us now that is really useful in data science?
Ayobami Ayodeji: So, I think data science is largely a team sport and whenever you have multiple people working on a project or whatever it is you’re working on it really helps to have some product management skill kind of thing. Maybe not specifically product management, but some kind of management skill. So being a product manager, I learned a lot about how to take different things in a system and actually build it together to make one goal kind of thing. Right. So, as a project manager, I learned a lot about understanding the business value of what you were trying to do.
Ayobami Ayodeji: Being able to quantify that business value, being able to create a business case for that project, because a lot of the time I was responsible for finding ways to make things better and a lot of that stuff was independent. I had to figure it out myself. So I had to come up with an idea. I wasn’t always the expert in the room for the idea, but I had to figure out what kind of resources I needed, what the cost would be and actually pitch it to the management team and let them know that this is what’s going to cost us and this is what value we’re going to get.
Ayobami Ayodeji: And a lot of that applies to data science as well, right? Because when you’re doing a data science project, a lot of the time you don’t know what the output is, but you can kind of estimate a lot of that. So, my project management background allows me to be able to do that in a structured way to understand what the management team is looking for, what their goals are and figure out exactly how to use chain management, people management and communication management to make those things actually happen, implement them.
Ayobami Ayodeji: I think a lot of that is also applicable to data science because a lot of the time they don’t understand some of the things I need to do to make this happen and the fact that there’s a risk that this won’t work. So those kinds of things are the things that are transferrable. So there’s a lot of ways to learn project management. I’d say, there’s a PMP organization certification you can work on. Now to become a PMP, you need to have some experience with product management. I think most data scientists have some product management experience, but you also have some learning the methodology.
Ayobami Ayodeji: But also, I personally, I’m trying to learn more Agile product management because I think that’s more applicable today, especially in data science. So that’s something else you can probably focus on. Yeah.
Kirill Eremenko: Okay. Yeah, we’ve been doing Agile more now at SuperDataScience. It works really well.
Ayobami Ayodeji: It works really well. Yeah.
Kirill Eremenko: Yeah. I love it. I love it. What does Agile in a nutshell? For those who don’t know.
Ayobami Ayodeji: Okay. So what I’ve learnt about Agile so far, I haven’t really done a lot of Agile, but it has to do with being able to iteratively make whatever you’re trying to do better. Right? So the idea is you have a minimum viable product or you have, you say you have a goal in mind, but you need to understand that the people that are investing in whatever you’re doing sometimes need to be able to have something to see some kind of progress, right? So the idea behind Agile is a series of methodologies to be able to do that.
Ayobami Ayodeji: So, for example, if you are trying to build like a recommendation engine, for example, you can probably start with some sort of minimum viable product that you can actually use the data now you have. Maybe you don’t have all the data at the beginning, but whatever data you have, you can build something really simple. Then, you can take that and test it. Maybe it doesn’t work the way you expect it to be. Learn from that and reiterate, make it a little bit better. Figure out…
Ayobami Ayodeji: I was also working closely with the customers or the product managers, for example, because they know what the final goal is. So they can actually, based on the feedback that you get from the initial test, you can actually make it better. So the advantage of Agile is you actually have something concrete pretty early on and you’re also learning from some of the mistakes, some of the failures from your previous situations. Then that way you can actually go do better. Whatever you have at the end is most likely closer to what the customer actually wants versus what you think the customer wants.
Kirill Eremenko: Yeah. So a good way to think about it to understand Agile is to… Like in engineering, I think engineering came from software, website development or software development, basically development. If you think about it in terms of like, let’s say you’re building a website or an app, then one way, the traditional way called the waterfall approach of product development is you put out the product scope, you just find the product cope, you do the designs, you put the database together, then you put a website or the app together to work with the database.
Kirill Eremenko: You do all the testing. You do like, as in that everything, the functionality is working and then finally you roll it out to your customers and then your customers say this is not what I wanted.
Ayobami Ayodeji: Exactly. Yeah. That’s like the worst thing. I feel like the worst thing you can do is to build right that which is not what you should be building in the first place.
Kirill Eremenko: Yeah. The other thing is that, along the way, in this waterfall, maybe on step two when you’re designing those mistake and then if you pick it up by the end you have to redo you the whole thing.
Ayobami Ayodeji: Exactly.
Kirill Eremenko: Whereas Agile, there’s this thing called the Agile manifesto that was designed, like just, it’s basically states rules that for instance, Agile, when you work under Agile you really value customer duration, you’re much more flexible. You don’t care about rigorous rules and things like that. So in Agile, what you do is you would, for instance, if you’re using Agile under the framework of… What is it?
Ayobami Ayodeji: Scrum?
Kirill Eremenko: Scrum. Yeah. Then you’d have sprints like for instance, two weeks sprints or a month or a week sprints. So basically you need to do that whole iteration from start to finish within two weeks. So you put together a very basic design, not the full design, a very basic design. You put together a very simple database. That’s all you need. Then you put, that’s a website or app, on a database. Then you do basic testing and then you roll out your customers. By the end of two weeks instead of six months, you’re already showing your customers something very trivial with lots of holes in it. But it has some functionality that they can give you feedback on.
Kirill Eremenko: Then you do it again, you improve next two weeks and again and again. So, in the case of waterfall, in six months, you’ve only done one iteration. In the case of Agile, in six months, you’ve done what, 12 iterations minimum. Is there any cool, there’s a competition, like you know those XPRIZES to make… I don’t know. Like a rocket to go to space or something. You get $1 million. Lots of teams are competing. They’re competing for something to like make a rocket in your basement that’ll go to space or I don’t know, like create an iPad where people can just learn English before knowing how to use the iPad.
Kirill Eremenko: Like just challenges like that, they’re called XPRIZES and I think they’re run by Peter Diamandis. I think this was either an XPRIZE or something similar. Basically there’s a big price, $100,000 or $1 million for the team that creates the first plane, solar powered plane that can fly a kilometer. That was a while ago. Now we have, I think the solar powered plane is done around the world, but back when this was happening, there’s like a solar power plane that you can fly a kilometer, five kilometers, whatever the threshold was.
Kirill Eremenko: So what’s the, all these teams are competing and they’re like, all right, so they building these planes that goes like 10 meters, 20 meters, falls, breaks and that’s it and then they do it again. So the team that won wasn’t the team with the best engineers on board or the team of the best scientists that could put everything together, it was the team that used the best approach and they used something similar to Agile. They made the plain intentionally out of foam material. So every time it fell, it took them like a day to put it back together and iterate. So while everybody else was iterating like once every month or two months, they were iterating like 20 times per month.
Kirill Eremenko: So they were learning so fast from their mistakes and they won because that’s the power of iteration.
Ayobami Ayodeji: Exactly. This is really important for sure.
Kirill Eremenko: Yeah. Yeah. So that’s a cool… There was a cool article I was reading about Agile in data science. We’ll link to it in the show notes.
Ayobami Ayodeji: Oh, that’s really good.
Kirill Eremenko: I’ll send it to you.
Ayobami Ayodeji: I’ll try to check it out. Yeah.
Kirill Eremenko: Well, thank you very much. This slowly brings us to the end of the show. Really enjoyed having you here at DataScienceGO and talking to you about your career and takeaways and also about project management. Before I let you go, what’s the best way for people to contact you, get in touch with you?
Ayobami Ayodeji: I think the best way is LinkedIn. I’m on LinkedIn. My name is Ayobami Ayodeji. I’m sure you’ll have the link on the show notes, so feel free to connect with me. If you have any questions about product managements or data science, whatever, or if you just want to follow my career, just-
Kirill Eremenko: Or if someone is in Toronto, catch up.
Ayobami Ayodeji: Yeah, exactly.
Kirill Eremenko: Yeah. It’s always good to meet in person. All right. One final thing before I let you go, what’s a book that you can recommend to our listeners that’s helped you in your career or life?
Ayobami Ayodeji: Okay. So I think there’s a lot of data science books you learn about data science. What I like to actually recommend is a little not exact data science, but… So it’s a book, it’s actually a book that I like to recommend. It’s not a data science book. But I think it’s super helpful for every data science. It’s a book called Mindset by Carol Dweck. I think-
Kirill Eremenko: Carol Dweck?
Ayobami Ayodeji: Yeah.
Kirill Eremenko: Mindset?
Ayobami Ayodeji: Mindset by Carol Dweck. I think it’s extremely important for data scientists because a lot of the stuff that we do is experimental and having the right mindset is important. That’s the only way you can be successful in an experimental profession like data science. The book actually talks about two different kinds of people. You have people with a growth mindset and then you have people with a fixed mindset.
Ayobami Ayodeji: So people with growth mindset, they’re thrilled about learning new things, right? But people with a fixed mindset, they’re all about, they believe subconsciously that their abilities are limited, it’s fixed and it can’t grow. So their motivation is to look good, to prove to other people that they’re good. While people with a growth mindset, what makes them happy is learning new things. So ask yourself the question, how do, what makes you feel smart? Do you feel smart when you’re able to do, finish something, complete tasks really easily, really quickly maybe quicker than other people?
Ayobami Ayodeji: Or is it when you’re actually able to complete a very difficult tasks they’ve never done before, you had to struggle and actually learn something from that? So what actually gets you going? Those are some of the things that she goes over in the book. So you learn a lot about, number one, what kind of person you are, what are the different things that determine what motivates different kinds of people, but also how to have the right mindset for growth. I think this is extremely important. If you’re new in data science, I’d say this is a really good place to start. Not a data science book, but it’ll change your life for sure.
Kirill Eremenko: Thanks. I haven’t read it. I’ll look into it. Mindset, right?
Ayobami Ayodeji: Yeah. Mindset.
Kirill Eremenko: Okay. All right. Well thanks Ayo, once again, for coming. I really enjoyed having you on the show.
Ayobami Ayodeji: Thank you.
Kirill Eremenko: So there you go ladies and gentlemen. That was Ayobami Ayodeji. I hope you enjoyed this podcast as much as I did. My personal favorite takeaway from today’s episode was the way that Ayo decided to shift his career into data science and what amount of sacrifice that required and what amount of courage inherently that he had to have in order to make that move. It was a very interesting decision in the sense that it’s expands new possibilities for him. As he said, data science is more applicable across different domains and he wants to have that variety and also it allows him to grow further.
Kirill Eremenko: So it’s very interesting that when we stop growing, it’s time to move on to something new. In this case, not only did Ayo follow that principal, he actually took it a step further. He changes completely his career, his industry, their required level of expertise and also the things that he had to know and things that he had to study. So hats off to Ayo for that. For me, that was very inspiring. Also you heard the takeaways from DataScienceGO 2019. It was a very exciting conference and there were lots of people who had a great fun time. We had 600 people actually attend, actually take their registration badge. Not just registered, but actually check into the conference.
Kirill Eremenko: This week we have the super early bird promotion happening, so it’ll end on Friday. So there’s still two more days to go to get your super early bird tickets for 2020. They come with a whopping 80% discount. So if you want to save 80% on your 2020 DataScienceGO tickets and learn a ton, meet people like Ayo and other inspiring data scientists and speakers, then head on over to datasciencego.com and grab your tickets today. Once again, that’s www.datasciencego.com where you can get your tickets today only until Friday with an 80% of discount.
Kirill Eremenko: As usual, you can get the show notes for this episode at www.superdatascience.com/301. That’s www.superdatascience.com/301, where you can get the show notes for this episode. Plus of course Ayo’s, LinkedIn, and any other materials that we mentioned on this podcast. So make sure to connect with Ayo. One more time. If you haven’t gotten your tickets for DataScienceGO yet, you can get an 80% of only until Friday at datasciencego.com. So head on over there and grab your tickets today and I’ll see you next time. Until then, happy analyzing.