Kirill: This is episode number 143, with senior consultant at b.telligent, Dr. Andreas Hopfgartner.
Kirill: Welcome to the Super Data Science podcast. My name is Kirill Eremenko, data science coach and lifestyle entrepreneur. And each week, we bring you inspiring people and ideas to help you build your successful career in data science. Thanks for being here today. And now, let’s make the complex simple.
Kirill: Welcome, ladies and gentlemen, back to the Super Data Science podcast. Super excited to have you on board. And I literally just finished a very captivating conversation, which I had with Dr. Andreas Hopfgartner, who is a senior consultant in the space of data science at b.telligent, which is a data science consulting firm, and we talked about a lot of really interesting things.
Kirill: So, in this episode, predominantly … Probably the most exciting part was when we chatted about the Internet of Things, and different applications. What the Internet of Things means, how different cities are becoming smart cities by using the Internet of Things. This was a completely … Like, we went on a complete tangent with this conversation, but it was so exciting to talk about, and we both gave a couple of examples in that space.
Kirill: Also, in this episode, you will find out how Andreas built his career in data science, what a fundamental approach he took to building his career. And when I mean fundamental, I really mean fundamental. Andreas had a Trello board set up just for his own career to understand how to attack the question of his career. So, I think a lot of us can learn from his experience, his ideas about how to approach building your career in the space of data science.
Kirill: And, of course, we’ll talk about data science and consulting, and why Andreas made the move from academia to consulting, and what that meant for him, and what tips and tricks he has for anybody else looking at the space of data science consulting as a career.
Kirill: All in all, a very exciting podcast, saturated with lots of content about data science, about the Internet of Things, about careers, and I can’t wait for you to check it out. Let’s dive straight into it. Without further ado, I bring to you Dr. Andreas Hopfgartner, a senior consultant in the space of data science at b.telligent.
Kirill: Welcome, ladies and gentlemen, to the Super Data Science podcast. Today, I’ve got an exciting guest on the show, Dr. Andreas Hopfgartner.
Kirill: How are you going, Andreas? Welcome to the show.
Dr. Andreas: Hi there. Thanks. Thanks for a warm welcome. Great to be here. I’m happy to be on the show. So, yeah.
Kirill: Fantastic. You’ve been listening to our podcast for a while, you mentioned, and now, you’re on it. Like, how long have you been listening to our podcast?
Dr. Andreas: I think it’s now … It’s like two years listening to the podcast while driving, while commuting, like, everywhere, and it’s quite exciting to be on the podcast now myself.
Kirill: Oh, that’s really cool. That’s really cool.
Dr. Andreas: Yeah.
Kirill: And where are you right now?
Dr. Andreas: I’m in Munich. I am in my apartment right now, and it’s freezing cold outside.
Kirill: Yeah.
Dr. Andreas op: Like a week ago, we were … Spring was on its way, and now it’s snowing again. So, yeah.
Kirill: Yeah. Yeah. I can feel your pain, because I’m actually, as we discussed before the podcast, like, I’m not far away. I’m in France, on the border between France and Switzerland, near Geneva, and yeah, it’s so cold. It’s like unbelievably cold. Like, why? It’s March, right? It’s like-
Dr. Andreas: Yeah.
Kirill: The middle of … End of March, and it’s like, still freezing. Where did this come from? Is this normal for Europe?
Dr. Andreas: The last years … The winter is getting later and later. So, I remember when I was a child, minus 20 degrees celsius, or even 30 degrees, was quite normal, like, in December, but now, it’s like a bit warmer, and snow is coming later to us. So, it seems like normal for the last years, but yeah, everybody is like … Has the flu-
Kirill: Yeah.
Dr. Andreas: And is waiting for the Spring, and the barbecue season here. So …
Kirill: Yeah, and this flu is very dangerous this year.
Dr. Andreas: Yeah.
Kirill: It’s like, it’s very different to what was happening previously. So, everybody should be careful out there.
Dr. Andreas: Yeah. Yeah. Yeah.
Kirill: Anyway. I got a cool story for our listeners. I already mentioned it to Andreas, that I’m like, sitting like, near Switzerland, and I can see Switzerland outside the window. I’m in a hotel, and so like, my Internet connection at the hotel is bad, but I have a Swiss SIM card right now, and so like, I’m close enough to Switzerland for the signal to be picked up, and I’m using that now, and so I actually put the phone near to the window, so I can [inaudible 00:05:39] to Switzerland.
Dr. Andreas: To the mountains.
Kirill: Yeah. Europe is so amazing in that sense. Like, in Australia, you have no borders. There’s no other countries there-
Dr. Andreas: Yeah.
Kirill: Which is still … Cool. Alright. Let’s jump into your story. So, we’re here to discuss your background, how you got into data science, what you’re doing right now, and things like that. So, let’s go through … Let’s kick off with your background. So, let’s start with … Because like … Just like a quick teaser for our listeners. There’s an exciting career that Andreas is pursuing right now. It’s a mix of background in physics, and now he’s in data science, and consulting, and things like that. So, I really want to dig in, and understand how you structured this path for yourself, and what you’ve been through. So, let’s kick off with the background.
Kirill: So, what did you study? What education did you undertake?
Dr. Andreas: Yeah. In the end, it all feels like things come one after another, and suddenly, I’m here. Yeah. I started with physics. I was very lazy, and bad at school, even at math, and-
Kirill: No way.
Dr. Andreas: Yeah. I was.
Kirill: You have a Ph.D. in physics, and you were lazy at school?
Dr. Andreas: Yeah.
Kirill: No way.
Dr. Andreas: I remember the first lecture, like, in math, and they started with complex numbers, and I was sitting there like, “What is a complex number?” And it was really hard, but it was … That was what I always wanted to do. Like, physics and astronomy, that’s … Yeah … Fascinated me, and I wanted to understand.
Kirill: Why did it fascinate you? I’m sorry to interrupt, but like, why? Is it like, because what you could see outside … The stars, you could see … Or is there any other reason?
Dr. Andreas: It’s like … Can I describe it? It’s like modeling the world, and yeah, describing it in terms of formulas, in terms of data. Can I derive something from the data? And that was what fascinated me in the end to … I came, like … In the middle of the studies, I did a lot in computer science, in computing physics, algorithms, and like, in Chaos theory, nonlinear dynamics, and [inaudible 00:08:08], and yeah, working with a computer fascinated me, as well.
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And that was a part … So …
Kirill: Why I ask, because I can totally relate to that, like, for me, as well, physics is so amazing. It’s because it’s like, it’s this world we live in, but it is all governed by formulas. It’s like it’s as if it’s-
Dr. Andreas: Yeah.
Kirill: As if it’s like we’re living in a computer simulation, you know? If you do this, then that. If this, then that. It’s like-
Dr. Andreas: Yeah. Yeah. Yeah. Yeah.
Kirill: It’s fascinating when you look at it from that perspective.
Dr. Andreas: Yeah. I remember one prof said … What was it again? “So, every physical problem has a beautiful but completely wrong solution.” And that’s what it is though.
Kirill: But like, how is that? What does that mean? Like, I don’t quite … I don’t quite understand. Like, why? Why wrong solution?
Dr. Andreas: So, well, because like … Everything is … It’s a model. It’s just a model. It doesn’t describe your problem perfectly. And in terms like, when you want to make predictions with formula, which is kind of regression too, so then you have to know your initial conditions, and if your initial conditions have errors, so you propagate the errors over time, and then you’re … Yeah. Everything gets more and more errors, and that is meant by … This. So, no model is perfect. It’s just a model.
Kirill: Yeah. Yeah.
Dr. Andreas: It’s not the truth. So …
Kirill: Gotcha. Gotcha. Like, even when … But if you propagate errors over time … I like that. It reminded me of … One time, I heard that in astronomy, sometimes when they predict like, the speed of a distant star or something like that, because it’s so far away-
Dr. Andreas: Yeah.
Kirill: And in this case, you’re propagating the error not by time, but by distance-
Dr. Andreas: Yeah.
Kirill: Some … Like in some … There are specific applications … You can correct me if I’m wrong here. They are happy, even if they get like, a … You know, plus/minus 2000%. You know? Like, if they predict the speed-
Dr. Andreas: Yeah.
Kirill: They’re like, “Plus/minus 2000%-”
Dr. Andreas: Yeah.
Kirill: Or like the location, you know? Like, the accuracy is ridiculously low-
Dr. Andreas: Yeah.
Kirill: It’s like, completely out of proportion … The error.
Dr. Andreas: Yeah.
Kirill: The error like, of 2000%. That’s crazy, but-
Dr. Andreas: Yeah.
Kirill: That’s the best they can get, sometimes.
Dr. Andreas: Yeah. Signal and the noise.
Kirill: Yeah. Yeah. Okay. So, sorry. Completely got off track there. So, you were … You mentioned you were lazy at school, but like, how did that translate into your career, but like, into your paths, through physics and math?
Kirill: Like, you were fascinated by physics. That’s where we stopped. You were fascinated by physics. Okay. Take us from there.
Dr. Andreas: Yeah. That’s … When physics started fascinating me, and I got more and more into algorithms, and I found myself doing a PhD in Magnetic Resonance Imaging, which there was a department for biomedical physics in … At the university, and I started there, basically doing motion correction, and taking the [inaudible 00:11:18] in an MRI is quite complicated. You are taking frequencies, so you have to translate from frequency domain to like, a spatial domain, using a [inaudible 00:11:30], and that makes all quite complicated, because you’re dealing with frequencies, and then with spatial, and so motion correction’s dealing a lot with algorithms, optimization algorithms, and yeah … Signal and noise problem is common in MRI. So, you’re always dealing between signal and noise, and try to optimize [inaudible 00:12:00]. So …
Kirill: Mm-hmm (affirmative). Okay.
Dr. Andreas: And I worked a lot with like, with principal component analysts to separate signal and noise, and uncorrelated signal from correlated signal. So, I had a few touch points to data science then, as well.
Kirill: Mm-hmm (affirmative). Alright. So … Is that like, what you did for your thesis for your PhD?
Dr. Andreas: Yes. Yep.
Kirill: Okay. And so, basically, when we are doing MRIs, maybe some of your work is involved in those MRIs that they use in medicine right now.
Dr. Andreas: Probably. Hope so. So, we did a lot of research in dental MRI, which I think just one or two people tried before-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: Even high precision impressions from the teeth to build … To re-build teeth, and so on and so forth, and it took a few years to … Yeah, before the publications were written, but I think now, it’s getting more and more in high precision MRIs. So, people are reading the publications right now. I hope, at least.
Kirill: Yeah. Good. Gotcha.
Dr. Andreas: Yeah. Yeah.
Kirill: What is a normal price for an MRI? Because like, it varies crazily between countries. In Australia, it’s like $300.00. In the US, when I was there, it was like, somebody told me like, over $1000.00 for one MRI scan. What would you say is a good price for an MRI?
Dr. Andreas: The MRI system, I think, was in former times … It was a million for tesla in magnetic field strength-
Kirill: Yeah.
Dr. Andreas: And the examination … I think it depends on what you’re doing, but roughly speaking, in Germany it was like $800.00 euros per hour.
Kirill: Wow. Okay.
Dr. Andreas: So …
Kirill: Interesting. Interesting. Okay. Sorry. That was just kind of like my personal [inaudible 00:14:07]. Alright. So, MRI … And then you did a PhD in that space-
Dr. Andreas: Yeah.
Kirill: But then you didn’t stay in academia. What made you make the move away from physics into data science?
Dr. Andreas: I wanted to do work with like, cutting edge technologies, with something that has real impact, and that is … Yeah. Like, I did a PhD in four and a half years, and this is usually like, a hard time-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: You do everything all over again, and again, and again, and again, and then you have to write it all together, and then you think through again-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: Think like project cycles in industry are like, shorter. You’re working with new technologies and yeah … That was basically the point … To do shorter projects, and see many companies, many different sectors, and I wanted to work in consultancy. So … That was my-
Kirill: Okay. Gotcha. So, are you satisfied with the project life cycle in data science? Is it faster than doing a PhD?
Dr. Andreas: Actually, my projects are not fast. Not really. They are a bit, but yeah. They’re a bit faster. What was funny was, when I was at the University, I worked a lot with MATLAB-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And now I’m working more and more, or basically, I’m only working with open source tools, like Python, and Pandas, and Spark, and so on.
Kirill: Yeah.
Dr. Andreas: I find it quite funny, because I thought like, I want to see other things like MATLAB, and more like, not professional, but more cutting edge, more … And not the academic open source, and like, now, I’m-
Kirill: The other way around. Back to the open source.
Dr. Andreas: Yeah.
Kirill: Yeah, but it’s interesting, because your LinkedIn does say you’re a Python addict.
Dr. Andreas: Yeah.
Kirill: Interesting. Was it … How long did it take you to become a Python addict from, you know, moving on away from MATLAB?
Dr. Andreas: Like … A Python addict … It took me a year. I think a year.
Kirill: A year. A year.
Dr. Andreas: So, yeah.
Kirill: So, from that phase, where you’re like, “No,” you know, “This is open source. I want to do MATLAB-”
Dr. Andreas: Yeah.
Kirill: And then a year later you’re like, “Oh, I love Python. So good.”
Dr. Andreas: Yeah. My first project where I was right off the university, I was at [inaudible 00:16:43], and I was working in optical imaging for semi-contractor processing, so I was at that size, and I was using MATLAB too, and I was discussing with people that are using Python, and they’d say, “Oh, you can basically do the same … It’s open source. It’s like, it has advantages here.” And now I’m seeing all of these advantages, and I would say I would never go back, because I can use so many packages, even in private projects-
Kirill: Yeah.
Dr. Andreas: Like, if it is web scraping, or something, it’s just easy to work with, and that’s what made me become an addict, because so many people are contributing.
Kirill: Yeah. Yeah. And that’s where the world’s going, kind of, like, where you can see, you know, companies like, what is it? SPSS, right?
Dr. Andreas: Yeah.
Kirill: And then, you know, MATLAB, and so on, being slowly edged out by these open source tools. Simply, because it’s … The power of the community, right?
Dr. Andreas: Yeah.
Kirill: Like, people are contributing. People are … You can ask questions. There’s always somebody to help you out, and things like that. That’s really cool … What I like about those tools.
Dr. Andreas: Yeah. They’re just awesome. Even if you’re … If you’re using some minor [inaudible 00:17:58], you can raise an issue, get help, and people are happy at getting issues, and improving … It’s … Yeah.
Kirill: Yeah. Yeah. Yeah.
Dr. Andreas: It’s just fun to work with.
Kirill: For sure. And I had a question for you. So, like, this is a careers podcast. A lot of people listening to this are starting out data scientists, or continuing, or expert data scientists, thinking about how to structure their career. Like, the question I had for you is, why did you move from the academia world into consulting? Like, a lot of people make the move from academia into data science-
Dr. Andreas: Yeah.
Kirill: But they go into industry. You know, they pick an industry, they pick a sector, and they work there. Whereas a move into consulting … I’ve seen that before. You know, we’ve had guests like that on the podcast, but I’m always interested to understand the thinking behind that. Like, what made you choose consulting over industry?
Dr. Andreas: Yeah. So, for me, it’s like the most exciting or interesting is to have like, a holistic picture of something, of an idea, a method, a technique, and then apply it to different sectors, to different problems, and I want to focus and learn the technique, the idea, but not being in a job where I’m constantly doing like, this and that. As a consultant, I see many of our customers … And I was remembering a former project. I had to train one of the colleagues from my customer in a specific technique, and he was like way better than me before I started-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And so basically with many meetings, with all of that … That in the end, it’s like, a quarter of a year later, I like, had many things that I could tell him, or like, we could discuss, and he could improve from that. So, I can concentrate more on … Yeah … The techniques, and more on the business as usual, or … Yeah … Don’t know how to say it. I hope you get the point.
Kirill: Yeah. Yeah. No, I get it. So, like, it gives you … I guess, a way of putting it is, it gives you that extra variety of projects, like-
Dr. Andreas: Yeah.
Kirill: You don’t have to be stuck working … Not necessarily stuck, but you don’t have to commit to working on this-
Dr. Andreas: Yeah.
Kirill: Same type of project, or same type of business as usual task, all of the time-
Dr. Andreas: Yeah.
Kirill: You can actually diversify your interests-
Dr. Andreas: Yeah.
Kirill: Or specialization between different things, and try out different things.
Dr. Andreas: Yeah. Yeah. Yeah.
Kirill: Okay. Cool. And so, you’ve worked at a couple of companies, after moving away from you’re academic role, and now you’re working for b.telligent. So, b.telligent, for those listening … Can you tell us a bit more about the company? What does b.telligent do? I’m assuming it’s a consulting firm. Is that correct?
Dr. Andreas: Yeah. Yeah. It’s a consulting firm, and it’s highly specialized in business intelligence, data warehousing, and [inaudible 00:21:20] data-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And I think the cool part is it’s like, owner run. We are not big. We are, I think, 140 people right now, including the front office and a back office. So, it’s a good size. It’s not too small, and not too big. You can … We’re really flexible, from size and from projects-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And our idea is we have many partners, like AWS, like [inaudible 00:21:58], which I’m a bit proud of. We’re, I think, at the moment, the only Databricks partner in Germany. So, that’s quite exciting. But we try to be independent, and help our customers in the way they need, and we challenge the customer, and ask to find the real business case, and the optimal business case, to help them. So, that’s our idea, and …
Kirill: Mm-hmm (affirmative).
Dr. Andreas: I don’t like to sound like an ad, but I love the company, and the people behind it, because they’re warm, and friendly, and helpful people, with a lot of knowledge, and specific knowledge in certain sectors. Yeah. We are working cross sectors. So, we are in each sector, but not in defense and armaments. So, yeah. That’s basically …
Kirill: Okay. Alright. So, thanks. Thanks for that. That’s a good description. And I think it’s important when a company creates the right culture, nurtures the right people, so that everybody’s friendly, and warm-
Dr. Andreas: Yeah.
Kirill: It’s really exciting to hear that, you know, you found a company like that-
Dr. Andreas: Yeah.
Kirill: And what about your role? Can you, like … Again, of course, so, without disclosing any sensitive information or trade secrets, can you tell us a bit about your role? What do you do? Like, your LinkedIn says you’re a senior consultant in data science at b.telligent. What is that? What does that involve? What is your day-to-day role there?
Dr. Andreas: Yeah. So, my role is basically bringing a bit more sensor knowledge to b.telligent. So, I think that was my chance to get a position in data science, from like … Like, from scratch, and I dealt a lot with signal processing, and so on. And my daily role is I work at …
Dr. Andreas: A long term customer, right now, is a public supplier in Munich, and I’m doing a lot of IoT analytics, use cases there. We try to find use cases in mobility and transportation, like from all transport vessels you can imagine in Munich. Starting from the Tram, to vehicles, rental vehicles, and so on … And yeah, I want to discuss IoT applications, and so …
Dr. Andreas: Speaking of that, I can like, give you an idea … We have … So, our company’s trying to share a lot of knowledge. We give many workshops together with our partners, and we have a conference, which is in May, in Munich. It’s basically a networking conference, but we have an IoT showcase there, and this is kind of … I think it’s a remote control circuit game, like, with a remote control car-
Kirill: That’s so cool.
Dr. Andreas: People can play soccer, and there’s a laser barrier, which is like, if goals are scored … And yeah, we want to combine that with predictive analytics, with maybe with the noise. Are people like, are they screaming when somebody scores a goal? Doing statistics every goal, and every movement is like, there’s a blockchain in the background-
Kirill: Hmm.
Dr. Andreas: Where everything is registered, and we can do blockchain analysis after that, and even if people like … Are they moving? We have a video camera, which is, like, are people moving closer to the winning team? And so on, and so forth. It’s like … Yeah. Finding use cases for IoT applications, and bringing more industrial IoT data to b.telligent. Because b.telligent grew up in business intelligence and customer intelligence, the focus is on customers, on people, and now we’re trying to get more industrial data, where the final customer is human, but … Yeah.
Kirill: Mm-hmm (affirmative). Gotcha.
Dr. Andreas: So …
Kirill: Andreas, it sounded very exciting, and I can feel like, a lot of passion, but there was so many things you just mentioned, from the conference, to the Munich transport, to the IoT, to the cameras, and so on. Let’s go through this like, step by step.
Dr. Andreas: Yeah.
Kirill: I really want to dig into more.
Dr. Andreas: Yeah. Okay.
Kirill: So, to start off with … Can you tell us what is IoT? What is Internet of Things? What is implied by that term?
Dr. Andreas: Okay. So IoT is, like, for me, it’s the connection of everything, which is connected to the internet-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: Or connected to the network. This … Are basically sensory information, like, it can be a heart rate sensor, like the watch … The smart watch you’re wearing-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: It can be a car with different sensors, it can be your washing machine, like, everything. Everything that is-
Kirill: Your mobile phone, your laptop-
Dr. Andreas: Your mobile phone, and so on. Yeah.
Kirill: Like a sensor in the street. Something like that. Like, anything that’s connected to the internet, basically.
Dr. Andreas: Yes. Yes.
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And the interesting part is, now, can we combine the information from the sensors together-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And get some new insights, or maybe some decisions that are made … Like in the smart home, can the blinds be, like … Can we open them when the sun is shining? And so on.
Kirill: Mm-hmm (affirmative). Okay. Gotcha. And so, this kind of leads us into that project for the transport in Munich, where you are … Is my understanding correct? That you’re also taking data from the Internet of Things, from different sensors, whether it’s on the street, whether it’s other sensors … I don’t know, like, through sensors that might be on a bus, sensors that might be on a different type of vehicle that’s participating in the transport, and you’re combining that to understand like, traffic congestion. Is that right? Can you talk a little bit more about that specific project?
Dr. Andreas: Yeah. So, at the moment, the main goal is to learn, like, for later, and I think the main goal of every city, and every transportation system, in terms of, like … Let’s take a bus. I think the most important [inaudible 00:29:15] are becoming electrified, because of the pollution of the cities, and so on … And a regular bus, you can fill the gas, like, when it’s coming back at night, you fill it, and then it’s going the next day. Then it comes back, and you fill it again. So, if you’re doing this with batteries, then you might run into breakdowns of the systems, or you have to have a good infrastructure in your city to load all of the buses, because that’s a lot of energy. It would take a long time to load them.
Kirill: Mm-hmm (affirmative).
Dr. Andreas: So, like, this our [inaudible 00:29:59], where you have to learn from the movement of people, of buses, of transportation systems, right now, to learn for the future, and to … Yeah, get use cases for that.
Kirill: Oh, okay. Gotcha. So, you’re kind of like, preempting the problem before it happens-
Dr. Andreas: Yeah.
Kirill: Like, people observe the current small fraction or small fleet of electric buses to understand what will happen when it’s a bigger fleet-
Dr. Andreas: Yeah.
Kirill: Like, just one of those use cases. Okay. Alright. Makes sense.
Dr. Andreas: Yeah.
Kirill: Yeah.
Dr. Andreas: And then you additionally deal with demands from people like, where do they want to go today? Where do they usually travel? It’s a working day, it’s a weekend, is the weather good? Is there special events? And these are things you want to combine, and each city has it’s own kind of infrastructure, and problems, and problematic areas. So, you can like, develop ideas to overcome them.
Kirill: Mm-hmm (affirmative). Mm-hmm (affirmative). Okay. Gotcha. That’s really cool. I actually heard of a project they did … I really apologize to those of Spain. I don’t remember … It’s either Madrid or Barcelona. I think it was Madrid. Where they applied a similar problem, but they did that for ambulances. So, like-
Dr. Andreas: Mm-hmm (affirmative).
Kirill: They used the Internet of Things-
Dr. Andreas: Yeah.
Kirill: They had from the city … Have you heard of that one?
Dr. Andreas: No. No. Not yet.
Kirill: Okay. So they basically use the internet … Data from the Internet of Things in the city. They had like, sensors, traffic lights, video cameras, everything they could like, connect together, and they would predict, you know, which is the best route for an ambulance to take at any point in time-
Dr. Andreas: Oh, okay.
Kirill: And that way they like, really increased the number of, you know, like, people they were able to save, because, you know, heart attacks, and things like that … It’s a big city, right? Like, the ambulance can get stuck in traffic-
Dr. Andreas: Yeah.
Kirill: And there’s nothing you can do, but they really optimized it through the Internet of Things, like that.
Dr. Andreas: That’s really interesting. Yeah. That’s a good application.
Kirill: Mm-hmm (affirmative). Yeah.
Dr. Andreas: An important application. Yeah.
Kirill: And another one is like, somewhere in the Netherlands, or in Denmark, it’s like, they built a smart city as well, like … Or they’re building a smart city. They’re putting a lot of sensors into these lamp posts, the ones you see that like, they already switch on automatically, most of the time, when it just gets dark, but that’s just a basic like, what’s it called? Basic reaction, or device doing that-
Dr. Andreas: Yeah.
Kirill: But they’re actually adding to those lamp posts. They’re adding sensors, you know, like, that sends like, humidity, pollution, traffic, pedestrians walking by, everything-
Dr. Andreas: Oh.
Kirill: Like, because these lamp posts, especially in first world countries, and Europe, like, they’re all over the place. If every lamp post has a sensor, then they can monitor like, what’s going on in the city. Where are more people walking? Less people walking? Where was like … There are louder noises, because, you know, more traffic is moving, and things like that, and that is really helping them understand how to plan the city for the future. So …
Dr. Andreas: Yeah. Yeah.
Kirill: It’s a very popular application these days.
Dr. Andreas: And it’s so interesting which information is sufficient-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: To … Yeah, to solve your problem. Like, I heard of the Deutsche Bahn, the rail company in Germany, they analyzed their rolling stair ways doing sound analysis, when there are tiny stones in between the stairs, so that’s a sign for …
Kirill: Maintenance.
Dr. Andreas: Yeah. For failure.
Kirill: Oh, okay.
Dr. Andreas: Yes. So, quite interesting. What is sufficient for …
Kirill: Yeah.
Dr. Andreas: Yeah, like, when you mentioned the lamps.
Kirill: Yeah, and another one in the US somewhere, I think it was … I’m not sure. Maybe Chicago, or some like, busy city in the US, they put sensors that can register the sound of a gunshot.
Dr. Andreas: Oh, okay.
Kirill: So, whenever somebody like, shoots from a gun, that sensor can understand. It’s not just like, a tire blew up-
Dr. Andreas: Yeah.
Kirill: Or some books fell on the ground. It can realize that is a gunshot somehow through the way … Through the acoustics of all gunshots, and right away like, police goes to that place. So, it helps with crime, as well.
Dr. Andreas: Crazy.
Kirill: Yeah. It’s interesting how-
Dr. Andreas: Yeah. Crazy.
Kirill: How Internet of Things and data are-
Dr. Andreas: Yeah.
Kirill: Affecting our lives, and most of the time we don’t even realize these things.
Dr. Andreas: Yeah.
Kirill: We don’t even think about them.
Dr. Andreas: Yeah. Have you heard of this sensor … I think it was under development from Google. It’s a radar sensor, and combined with other sensors, and you just place it in your home-
Kirill: Uh-huh.
Dr. Andreas: And then you start doing things like, you open your fridge, and you tell the sensor, “This is the electric impression from opening my fridge.” Wherever it is-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And it learns like, what are you doing, and so, one day, you go out of the house, and like, your system is telling you, “You forgot to close the fridge,” because that signal is missing. So, like, it learns from electric impressions, or something … Yeah-
Kirill: That’s so cool. I want to get one of those.
Dr. Andreas: To get the information.
Kirill: That’s awesome. That’s such a good idea. Like, and that’s what kind of like, I think people need to look into, especially people who have creative minds, entrepreneurial spirit-
Dr. Andreas: Yeah.
Kirill: They should think about ways to use data, or capture … Because Internet of Things is ultimately about capturing data. It’s like, how do you capture … Like, the way I … Like, my view on this is data exists, no matter what-
Dr. Andreas: Yeah.
Kirill: Whether we capture it or not. Like, a plane flying back, that’s data. A gust of wind, that’s data. A leaf falling from a tree, that’s data. The problem is that we’re … In many cases, we’re just not capturing it. So, Internet of Things addresses the problem, “Okay, how do we capture this data?” And that idea that you just described from Google, that’s an ingenious idea-
Dr. Andreas: Yeah.
Kirill: Of capturing data from lots of different aspects of a home, without actually interfering with the devices, such as the fridge, or the washing machine, and so on. You just capture it from electric impulses, because again-
Dr. Andreas: Yeah.
Kirill: Those electric impulses, the waves, they’re there. The data is there. Just find a way to capture it. And it’s a whole new industry. Like, I believe it’s like trillions of dollars in the Internet of Things, right now.
Dr. Andreas: Yeah.
Kirill: Okay. Well, exciting, exciting field to be in. And that’s a great overview of Internet of Things, and now it makes a bit more sense. So, like, what you mentioned about that conference you have where people can play with … Like, remote controls cars, they can play soccer, right? Is that … So, that’s kind of like, if I understand correctly, that’s like, an experiment where you just invite people to have fun, but then at the same time, it’s a challenge for you guys to see what … How can you capture that data that is present in that whole experience itself, with people coming closer with the cameras, further … You know, making more noise, making less noise, and how can you use that in another way? Am I understanding that right? Or that’s completely out of line?
Dr. Andreas: Yes. That’s definitely it. So, it’s great for … So, that’s a conference where customers are speaking to customers. So, we have a lot of people from … Yeah, the engineering departments-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And the [inaudible 00:37:35] departments from our customers. So, they can play around, they can see the ideas, maybe they can adopt something for themselves, can talk about it, generate new ideas, and it’s just like … It’s amazing to observe all of this, and like … But there are so many ideas, and when I tell people of this, they bring their own ideas, and say, “Oh, you can do this.” And, “What is the part of blockchain here?” And, “Why do you need this?” And-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: So on, and yeah. It’s quite amazing to see this. And the amazing part for me was this all runs in Azure environments. So, even the blockchain, it’s like, I don’t know completely … I think it’s like 20 or 30 lines of code, so-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: It’s just like, the beauty of an idea, and the beauty of code in the end, it’s like, if you have the idea, it’s like, one, or two, or three hours to see it working-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: So, that’s the exciting part. It’s easy to do.
Kirill: That’s really cool. So, is that Azure? Is that the environment you mentioned? That-
Dr. Andreas: Yeah.
Kirill: The blockchains, and different things are running on? Okay. Alright. Well, that’s really cool to hear. You mentioned blockchains. So, like, I’ve been studying about blockchain for the past couple of months, and it’s really exciting technology. Like, what can you say about how blockchain is impacting the world right now, from your perspective, from what you’ve seen?
Dr. Andreas: From my perspective, it’s like the Wild West. Like, blockchain, everyone has a lot of ideas. Most people think of cryptocoins-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And I try to, like, get good business cases for dealing with data that are worthwhile for our customers. So, probably it’s possible to do permission management with like, customer data and so on, and so forth, so that’s occupying me, and us, a lot-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: I’m not deep … I understand how the blockchain works, but I’m not deeply into it right now. I try to develop that, as well as the deep learning aspects, so-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: I’m trying to get more … To get deeper into deep learning right now. So, blockchain is postponed to later on, but I try to work on ideas.
Kirill: Yeah. Yeah. Gotcha.
Dr. Andreas: Yeah. So, I’m pretty much excited what will be real business cases for the blockchain-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And which is nice to have, or which is a good idea, but not suitable for the application for our customers. So …
Kirill: Mm-hmm (affirmative). Gotcha. So, but like, in general, you think that b.telligent, your company, will be going into the space of blockchain, at some point? Or if you haven’t yet started doing it, like, it is a perspective area for you guys to be looking into?
Dr. Andreas: It is. Definitely. Surely. Many, many colleagues are working right now, and try to get deeper into it, to develop ideas, and working with this. Yeah.
Kirill: Okay. Alright. Yeah. I totally agree. I think it’s definitely a … At the very least, a very exciting technology that’s come up, and it has so much potential-
Dr. Andreas: Yeah.
Kirill: That you can really disrupt many industries, and people who get their hands on blockchain, or get the skills of blockchain right now, it’s like, I don’t know. It’s like … How to explain … It’s like companies that didn’t have websites versus companies that first started to have websites-
Dr. Andreas: Yeah.
Kirill: That type of advantage.
Dr. Andreas: Yeah.
Kirill: That magnitude of disruption.
Dr. Andreas: Yeah. Yeah. And-
Kirill: Okay. Yeah?
Dr. Andreas: And, I’m very much excited for your online course. I’m waiting for-
Kirill: Thank you.
Dr. Andreas: For it come.
Kirill: Yeah. The kick starter just finished-
Dr. Andreas: Yeah. Yeah.
Kirill: So that’s cool, and now the course is going to go live in the next couple of weeks. So, yeah. Hopefully, you know, like, we’re slowly getting things there.
Kirill: Yeah. Okay. So, we chatted about that. So, in terms of careers, what would you say are important skills for a data scientist to have if they’d like to be a consultant?
Dr. Andreas: Hmm. So, for us, I think, the most important part, is like, strong background in statistical knowledge, because we have to explain to the customer a lot what we are doing. So, statistics are very important for us, and even SQL. I know everyone who see programming like, in Python, or Spark, or something, is like laughing, “Ha ha. SQL.” And so on. If you’re sitting with the customer, and you’re sitting like, getting something out of a database, and then you have to Google, like, “What … How do I do this analytical da, da, da?”
Kirill: Not the best impression, right?
Dr. Andreas: Not the best impression. Yes.
Kirill: Yeah.
Dr. Andreas: So, it’s definitely … Okay … For other things … It’s like, in consultancy, it’s often like, learning while doing a project, but … Yeah. Statistics and SQL is like, very important for us.
Kirill: Mm-hmm (affirmative).
Dr. Andreas: We have the business. The business aspect is very important for us. So, having good business cases for the customer, and the process knowledge, understanding the company, how does the data flow into the data warehouse, and so on. This is very important for us, and I think in business intelligence data science, this is the most … These are the most important aspects. So, and then I think, having fun, wanting to learn and improve, is … Yeah, the most important, because when you start learning something, it’s old, so-
Kirill: Yeah.
Dr. Andreas: You can learn the next thing-
Kirill: Yeah.
Dr. Andreas: And I think Python and Spark is now quite [inaudible 00:44:20] tools, which we are working a lot. So, R’s get more and more into background, and Spark is like the new SQL-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: So these are like the skills I see for data scientists in BI consultancy.
Kirill: Mm-hmm (affirmative). Interesting, interesting. So, you’ve used R before, and you think that R is kind of like, moving to the background, right now?
Dr. Andreas: I didn’t … I didn’t really use R. So, I was starting right off with Python. I did … I think I understand like, the structure and the semantics of R, but I didn’t really work a lot with it, and I started with Python, and I started with Python right off when I was joining b.telligent, and Spark, too-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: So, yeah.
Kirill: Okay.
Dr. Andreas: These are from discussions with colleagues, and with colleagues from other companies. They see that R is going more and more to background. I think you can see it when you look at the Spark website, the website from Databricks. Like, the [inaudible 00:45:31] is what people are doing, which is most developed, and R is like, yeah … I think it is a development, but-
Kirill: Not as fast. Not as productive.
Dr. Andreas: Not as fast. Yeah.
Kirill: Yeah. No, I agree. I think, you know, I think R has a future, but, like, personally, I think in very specialized applications. I might be wrong. It might be different, but the advantage of Python, I think, is that first of all, syntax is simpler to understand for beginners to jump into-
Dr. Andreas: Yeah.
Kirill: And also it has that aspect that R doesn’t. That it is a language that you can and is used for development, not just for data science. So, R is specifically tailored to data science and statistics, whereas Python can-
Dr. Andreas: Yeah.
Kirill: And is used for application development, and when you combine the two, then you get super power. Like, if you’re a data scientist, why would you limit yourself? Maybe in the future you will want to develop applications one day. Or the other way around. If you’re developing applications, you might transition into using data science in your applications, and Python allows for that.
Dr. Andreas: Yeah.
Kirill: Alright. And so, we talked about tools a little bit. What about techniques? You know, like, what kind of data science techniques do you use? Again, if you can share these?
Dr. Andreas: For trying things out, I use [inaudible 00:46:49] methods-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: But it’s basically … At my project, I’m in the phases of, like … I have one small project, which is a proof of concept, with like, not the best data. I have to do a lot of data cleaning, and prep, and so on. So, it’s like trying things out, and so-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: I try to use like, simple and clean models with the customer. So, a lot of customers want to have the models explained. So, it’s easier to use simple models, even if they perform a bit worse, but you can tune-
Kirill: Yeah.
Dr. Andreas: And …
Kirill: Interpretability. Yeah. That’s an important component.
Dr. Andreas: Yeah. And right now, I’m doing a lot of data mining. Seeing like correlations of like, different attributes and something, and try to combine that with different use cases. So … This is what I’m working, in terms of techniques, right now.
Kirill: Mm-hmm (affirmative).
Dr. Andreas: I try to do … To develop a few predictive main tenants of things, and there are different approaches. So, either you can learn from your data, is something going to break? Yes or no. Like, you can predict a binary label, which is, basically, not working well, because of all of the false positives you have, and that’s, I think, the biggest problem there. And there are other approaches, like predicting some remaining useful lifetimes. So, at first you have to break things, and then try to learn if you can tell how … Or if you can train a model on that … When it will break. There are a few interesting … If you look for that on YouTube, I think there is an interesting video from Microsoft. I can’t remember the name right now, but if you are interested, I can give you the-
Kirill: Yes.
Dr. Andreas: The link afterwards.
Kirill: Yeah, please.
Dr. Andreas: I find it quite an interesting method for predicting [inaudible 00:49:06] lifetime of a product, and so, these are like, techniques I’m using right now.
Kirill: Okay. Okay. That’s really cool. So, some interesting methods, some interesting techniques, especially remaining useful lifetime. We talked to someone on the podcast about remaining useful lifetime. I can’t remember the technique they used for that, but … Yeah. Okay.
Kirill: Alright. So, then the other thing I wanted to ask you, what’s the biggest challenge you’ve had as a data scientist? Like, what’s the biggest thing that you’ve had to face?
Dr. Andreas: The biggest challenge I had to face was I was one month at b.telligent, and beginning to … I was at the customer … Yeah, and trying to get into it, learning to know all of the colleagues, and … Yeah … It was a lot of impressions. And then there was this one day where my boss asked me … There is a training of Microsoft in Amsterdam, and like, three of my colleagues, they have children, and there were public holidays, and they said … Two them said they want to stay with their family, and the other one said … So, it’s like too basic for him. It’s a basic training, and he’s like … Has advanced skills in Databricks … That it was like, pre-private announcement of Azure Databricks-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: And we were invited as the only German partner to train with a Microsoft global [inaudible 00:50:44], and that was the hard part. So, being a data [inaudible 00:50:48] for like, a month, and together, in one room with twenty global [inaudible 00:50:56] of Microsoft people, asking me like, “Is this your customer … ” [inaudible 00:51:03], with a guy from Shell, and like-
Kirill: Yeah.
Dr. Andreas: It was tough times, because it was really, really good people, and … Yeah.
Kirill: Yeah. So, just threw you into the fire, right away. Like, into-
Dr. Andreas: Yeah.
Kirill: Like, into the war. Yeah. Like, what’s it called? Training by fire, yeah? Sounds like that.
Dr. Andreas: Training by fire. Yeah.
Kirill: Trial by fire. There we go.
Dr. Andreas: Yeah. But it was a lot of fun. A lot of fun people. So …
Kirill: That’s crazy. Okay. And I like those situations, because you see something to strive towards very quickly. So, you see like, where you want to be like, in the next year, or two, or three, or five, or whatever. Like-
Dr. Andreas: Yeah.
Kirill: It’s like, even though you’re completely out of your comfort zone, you want that to be a comfort zone one day.
Dr. Andreas: Yeah. Yeah. Yeah.
Kirill: Okay. Cool. And so … Alright, and now, on the flip side, like, that’s the biggest challenge. What about a big … A recent win? Something that you can share with us? Something that you’re proud of, that you accomplished in your role.
Dr. Andreas: I think the biggest win was finally to get a position in data science. It took me two years after exhausting talks, and talking to … It was [inaudible 00:52:19], where I worked before, they didn’t really have a data science department, or people doing a lot of data science. So, only like, machine learning engineers, like here and there, and I wanted to do data science and consulting, and so, finally … I remember I had a Trello board, where I was like, figuring out what can I do in data science, how to get into data science, how to get a job-
Kirill: You set up a Trello board for your self to understand how to get into data science?
Dr. Andreas: Yeah. Like, I love planning.
Kirill: That’s such a fundamental approach to a career. Oh, my. Wow. You should give lessons on how to do that. I love that. That’s so cool.
Dr. Andreas: I love working with tools, and I love Trello, and I didn’t have a project to work with, so I said, “Alright. Let’s do a business case,” and like, I did it, and it was like, every quarter where I looked into it, and then forget it, and there was this one day where I wanted to like, delete it, or go back, and it’s like, I reached almost exactly what I wrote into this board. It was amazing. So …
Kirill: That’s so cool.
Dr. Andreas: It was quite a win. So …
Kirill: I hope those … Sorry to interrupt, but I hope those listening to this are like, take note of that, because you’re right. It’s like, once you write something down, it’s so different than just having it in your head. Once you write something down, like, “I want this.” “This is my aspiration.”
Dr. Andreas: Yeah.
Kirill: “This is my goal.” “This is my dream.” Once you write something down, it’s like magic how these things come true. It’s magic how the opportunities like, align and you … And you look back, and you’re like, “Whoa.” You know, “I wrote it down, and now I’m there. Now, I got what I wanted.” It’s always fascinating.
Dr. Andreas: Yeah. It is. And I remember, I was buying this white board for my apartment, just like, trying out, because I love like, writing by hand-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: It gives me like, more … More [inaudible 00:54:18]. So, she was looking at that, and … “What is this?”
Kirill: Yeah, yeah.
Dr. Andreas: “What is this doing in our apartment?”
Kirill: Yeah.
Dr. Andreas: But it’s exactly like you say. It’s like … You have to like, visualize, or like, when you see it …
Kirill: Yeah. Okay. And so, when you found … Like, that’s interesting. Like, I’d like to go into that a little bit more. So, you found it difficult … Can you clarify this for us. Like, you finished your PhD, and then you found it difficult … Like, for two years, you were looking for a job? Or you already had a job, but you were looking for a different job? How did that work?
Dr. Andreas: I had a job. I was working for [inaudible 00:54:59], which is like, a technology consultancy, or engineering services-
Kirill: Uh-huh.
Dr. Andreas: As you might want to call-
Kirill: Yeah.
Dr. Andreas: And I did a lot of system integration, engineering, and working with algorithms. They always did like, simulations, and optimization problems-
Kirill: Yeah.
Dr. Andreas: But … Yeah. I was trying to do like, real data science. Like, data science you see people doing in podcasts. Like, you know what I mean? And it was hard like, getting there, because we didn’t have many data science projects, and if … Then it was hard to get into them, because right now, you had a project somewhere else, or like, there were better people, or … Yeah. So, I tried to speak with colleagues from [inaudible 00:56:01] with different business units. So, we had strong business units for the sectors, like automotive, which has more projects like, in machine learning-
Kirill: Yep.
Dr. Andreas: And image recognition, and so on, and so forth. And there were other sectors, like rail, and-
Kirill: Yeah.
Dr. Andreas: Like air crafts, and so on, and it’s hard to move from one sector to another. So-
Kirill: Gotcha.
Dr. Andreas: And so, it’s always a process, and it was harder to move like, in this big company with 4000 people in Germany, I think-
Kirill: Yeah.
Dr. Andreas: So, I tried to speak a lot to like, head hunters and … Yeah … To find a good position where a company trusts me, that I can do data science, that I [inaudible 00:56:51], and that I can bring something within the company. So, I think it’s a hard part to see what you can bring to something. So, I spoke to a lot of [inaudible 00:57:07] consultancies, and all of that, and I think that was not the right thing. So-
Kirill: Okay. Go ahead-
Dr. Andreas: I said, “Yeah. Okay. Maybe it’s a bit [inaudible 00:57:19].” So …
Kirill: Gotcha. So, basically, it’s not that you didn’t have a job, or you were struggling to find job. It’s just that you were looking for the right job. The one that you would enjoy, the one that you actually wanted-
Dr. Andreas: Yeah.
Kirill: So, yeah. Okay. Gotcha. Because that part I found a little surprising. There’s like, so much demand for people with your type of skills-
Dr. Andreas: Yeah.
Kirill: Like, it’s probably quite easy. And that’s what people should also understand. It’s quite easy to get any type of job, but I really respect when people set out and create a Trello board to find that perfect job for themselves. That’s very important, right? To find the opportunity that you’re going to be enjoying, and happy with, and growing in, and it seems like you found that in b.telligent.
Dr. Andreas: Yeah. Yeah. Yeah.
Kirill: Okay. Okay. Cool. Where’s my next question I had?
Kirill: Okay. So, we talked about the challenge. Now, again, opposite side. What is your most favorite thing about being a data scientist? Something that excites you. Why do you wake up with a smile on your face in the morning? Hopefully … Most days.
Dr. Andreas: It’s like, when you get insights … Insights from data.
Kirill: Hmm.
Dr. Andreas: That is one part. And you can tell these insights to another person, to a customer, and he’s like, “Oh, wow. That’s interesting.” So, when you can see that, and give some other people that insight, and they are like, surprised from that, or something. That’s quite funny-
Kirill: Hmm.
Dr. Andreas: So that is one part, and it’s like, I can do the programming stuff, I can try things out, even in … If it’s data cleansing, or validation, that’s huge fun, and it’s a community where people are sharing a lot. So, you can … I’m quite overwhelmed with information. I read some posts at Reddit, there are communities here, there’s a [inaudible 00:59:17] community. I really love to listen to the short videos of [inaudible 00:59:24], and so, [inaudible 00:59:28] like, sharing some code, or I love this concept of sharing, and working together on like, a bigger team, bigger project. So …
Kirill: Hmm.
Dr. Andreas: That’s what … Yeah.
Kirill: Totally, totally, totally get you there.
Dr. Andreas: Yeah.
Kirill: The whole community side of things.
Dr. Andreas: Yeah.
Kirill: The sharing. The … Yeah. Gotcha.
Dr. Andreas: I think-
Kirill: It’s like a little club, right?
Dr. Andreas: Yeah. I think there’s no other … Or [inaudible 00:59:58] is there another community in another sector or something, which are so like, closely bound by [inaudible 01:00:08], all the tools, and all of the forums, which we have.
Kirill: Oh, good question. I would say like, fans of some board game, or like, card game. Like-
Dr. Andreas: Yeah.
Kirill: [inaudible 01:00:19], you know? Like, that’s what it feels like.
Dr. Andreas: Yeah. Yeah.
Kirill: People who like something … More hobbies, you know? Like-
Dr. Andreas: Yeah.
Kirill: [inaudible 01:00:28] people putting together cardboard planes, or what are they called? Like-
Dr. Andreas: Yeah.
Kirill: Models. Plane models, and things like that. That’s what it really feels like. And so, I guess, that’s because a lot of people do do it as a hobby, as well. Like, they have a job, their full time job, whether it’s data science or not, but then in the evening they like, learn more stuff, more skills. You know, if you’re … I don’t know. If you’re in a more traditional type of role, you usually like, you work hard at your job, and maybe you love your job, but when you get home, you know, you’re not going to like, I don’t know … If you make coffees all day, right, you’re probably not going to come back home, and learn how to make more coffees, and share that. Maybe people do that-
Dr. Andreas: Yeah.
Kirill: I don’t know, but like, in my understanding this is more kind of like a hobby, as well, in addition to just a profession.
Dr. Andreas: Yeah.
Kirill: Okay. So, that’s your favorite thing. And a philosophical question to wrap up our podcast … Slowly coming to the end. I can’t imagine … I can’t believe how quickly this hour flew by. I’ve been having so much fun chatting to you. This is crazy.
Kirill: Where do you think the field of data science is going? And what do you think our listeners should look into to prepare for the future, and to have … To be able to build careers for themselves in the future of data science?
Dr. Andreas: I’m seeing so many things developing, and I think the most important, for one data scientist, is to strongly specialize, and to pick out things which are the most fun for you. So, there are all of this sensor information, which are coming. You’re getting more and more data … Streaming data … Neural networks are a trend … So, audio image recognition comes more and more. So, I think it’s important to focus on what is fun for you, and yeah …
Kirill: Mm-hmm (affirmative).
Dr. Andreas: So …
Kirill: Find out what you enjoy. Like, there’s lots of … There’s so many sub-fields in the field of data science. Like, just don’t settle for doing something. Like, right? If I understand you correctly.
Dr. Andreas: Yeah.
Kirill: Like, do what you actually love doing.
Dr. Andreas: Yeah. Definitely. Projects get more and more … We see it. From the point of view of consultancy, we’re getting more and more requests and projects, and data is growing. So, it’s hard to say where it will go, but there are trends, which I see. There are trends going to the blockchain, de-centralized apps, and so on, and so forth, but I can hardly say like, how it will be in two years, or five years, so …
Kirill: Yeah.
Dr. Andreas: Too early for me.
Kirill: Yeah. Yeah. Nobody knows at this stage.
Dr. Andreas: Yeah.
Kirill: Okay. Gotcha. Well, thank you so much for coming on the show. I think like, a lot of people will get some valuable insights from here.
Kirill: If our listeners would like to contact you, follow you, find you, follow your career, where are the best places to get in touch?
Dr. Andreas: The best place is directly via Twitter-
Kirill: Mm-hmm (affirmative).
Dr. Andreas: Or my LinkedIn. I can share you the URL’s-
Kirill: Yeah.
Dr. Andreas: If that’s convenient for you.
Kirill: Yeah. We’ll include those in the show notes.
Dr. Andreas: Yeah.
Kirill: If you don’t mind saying your Twitter handle like … Now … Mention it. So, just in case people want to look you up right away.
Dr. Andreas: The Twitter handle is like, hopfi80, H-O-P-F-I-80. So, you tried to pronounce my name in the beginning-
Kirill: Yeah.
Dr. Andreas: You did very, very good.
Kirill: Gotcha.
Dr. Andreas: So, this is the abbreviation of my name.
Kirill: Gotcha. Okay. Alright. We’ll definitely include those links in the show notes.
Kirill: And one final question for you today. What’s a book that you can recommend to our listeners to help them become better at their careers or to develop their own lives in a better way?
Dr. Andreas: Uh-huh. For me, it was always The Elements of Statistical Learning … What I always come back to, because I think it’s very strong in mathematics. It’s definitely not a book which you begin to read, in two days, you read it from the beginning to the end, but it’s a book I always come back to, reading something. So, that would be my favorite book.
Kirill: Awesome. Elements of Statistical Learning. For our listeners there.
Kirill: Okay. Well, once again, thank you so much, Andreas, for coming on the show. It’s been a pleasure chatting to you. Like I mentioned, it’s kind of … The hour flew by extremely quickly, and I hope our listeners picked up a few ideas. I’m sure there’s tons of valuable insights that people can [inaudible 01:05:25] from this podcast. Thank you so much for coming on today.
Dr. Andreas: Thank you so much, Kirill, as well. It was really a pleasure for me. It was very exciting to be on your podcast, and … Yeah. Thank you for the chance to tell my story.
Kirill: Awesome. Glad. It was a pleasure. Alright. Take care. Bye.
Dr. Andreas: Bye.
Kirill: So, there you have it. That was Dr. Andreas Hopfgartner, from b.telligent. I really hope you enjoyed today’s podcast. And for me, personally, the most exciting part, I would say, was like … Of course, it’s so much to choose from. We talked about so many things. But I would say, the most exciting part for me was learning more about the Internet of Things, and those different examples, and understanding how not just companies, but even cities are starting to use the Internet of Things, and data to better themselves, to provide better services to their citizens, and how the Internet of Things is actually affecting our lives without us often even realizing it.
Kirill: So, that was our podcast for today. I would really love to know what your favorite part was. I would be interested to see what take-aways people had from here. If you liked this episode, and if you think there’s somebody you know who could benefit from it, then make sure to forward it on to them, so they can also find out a bit about the Internet of Things.
Kirill: And I highly encourage getting in touch or connecting with Dr. Andreas Hopfgartner. You can find the link to his LinkedIn, and to his Twitter, at the show notes, which are available at www.www.superdatascience.com/143. There you will also find the transcript for this episode, if you’d like to read our conversation in text.
Kirill: And on that note, thank you so much for being here today. I can’t wait to see you back here next time, and until then, happy analyzing.
