You are listening to the HumAIn Podcast. HumAIn is your first look at the startups and industry titans that are leading and disrupting artificial Intelligence, data science, future of work, and developer education. I am your host, David Yakobovitch, and you’re listening to HumAIn. If you like this episode, remember to subscribe and leave a review. Now onto the show.
Welcome back to the HumAIn Podcast listeners. Today on the show, we have the CTO of Play Magnus² and the CEO of Solv Oslo, Felipe Longe¹. I’m super excited to engage in this conversation today because many of you know, as a data scientist, chess is one of the big sports in the world that we love to learn about.
Not only with coding, but playing and staying mentally active. And back in 2016, I attended the Five World Chess tournament in New York City and got to see first and behind the glass Magnus Carlsen play against another world grandmaster and little did we know at the time, but Felipe was there, and we didn’t meet yet, but here we are today. Learning about everything, not only in the world of Chess but also an AI and ML and what you’re doing with startups. So Felipe, thanks for joining us on the show.
Thanks, David. Looking forward to this. We’ve been emailing for a long time, but now we’re finally here and it’s really fun to just kick off talking about all these things that you mentioned.
You had a lot there on the introduction, I am two roles, basically. I work as a CEO at Solve Oslo where we do product development and consulting. We have a focus on machine learning and then also a bit on med tech. But at the same time, I’m also a CTO at the Play Magnus. And it’s kind of funny.
Both offices are strategically located on the same floor in the same building. I have this double agent life running back and forth between the offices. But now with the Corona situation, the only positive thing about this whole spectacle list is that I can actually work from home and, then. It’s kind of easy to switch context, but of course, that doesn’t mean it’s worth it.
It’s, we’re seeing with coronavirus or COVID-19 that life’s become digital-first, and it’s, we almost have these digital twins. As you mentioned, you had your in-person office for the ventures that you’re growing, but everything’s gone #digital. And it’s amazing to see, like in the history you’ve been involved with, Play Magnus.
It’s gone from Magnus Carlsen world champion to this app where anyone like myself can learn and practice and learn from the great. To an app that’s even built-in data science and AI as well. I’d love to hear about your journey with the team and the app, how that’s evolved over the years.
This started back in, 2013 actually. So it was quite cutting edge. It was quite an ambition back then. So it was Magnus Carlsen himself and his father kind of. They wanted to enter the digital space and do something with this brand building basically. But their vision was to use #chess as a fun tool to make the world a smarter place.
And, the first ideas were not the best ones. It was having chess pieces with Magnus Carlsen’s face on it. Unfortunately, there were objections against that, and now you can see that we’ve made tons of apps. And the reason being is that that’s a good way to spread your brand, to create the awareness and also to spread joy.
I mean, it’s actually, I hope people find these apps fun. I would rather see my kids playing with Play Chess on their phone than others, maybe I shouldn’t mention specifics, but, other games, in general. I do believe that chess can make the world a smarter place. It’s not proven yet actually that chess has an effect on your intelligence, that you become more intelligent by playing chess, but there’s, at least a correlation between intelligent people and chess. So if you pick a random test player, there’s a higher probability that that chess player is more intelligent than a random person in general.
So there’s definitely a correlation, but we don’t know if, if one leads to the other directly, all the smart people want to play chess might be that’s as well. Anyway, it’s a practice for brain activity that’s for sure. I wear this, i-watch right on there. You have, if your heart rate is too high and you get the warning on, it’s never triggered apart from when I spent some time at the chess club, I was planning this, let’s game, then it’s actually figure.
It’s sports on, it’s a lot of brain activity going on. I find it fun to work with. Well, back to the story, this is, I’m branching off here. So back to sorry. So we found out that we’re going to do an app. That could be a good start at the moment. It was #Playmagnus, it was a company with one person I wasn’t hired there.
They went to a consulting company. I was working in a consulting company with the innovation stuff. When I heard about the project, I just left all the projects I was working on and just focused on Play Magnus. And we created the app and two months, actually it was, was a heck of a marathon, code monkey period where we didn’t know where we’re going.
It was just, we just did something on. But that’s mostly a nice experience. Of course the code now is it’s not readable, but we’re working on refactoring the whole thing. So that was the first Play Magnus App already there. We have the AI, which plays at Magnus. This AI is interesting in the company and Play Magnus, there were two developers that were hired early on after the CEO.
It was me full-time and then we also hired a tool homestar part-time and he has actually developed the first iteration of Stockfish. So he developed this, glowworm at, which is, like an early version of Stockfish. It was his code and then it’s brand stuff to Stockfish. So we have this sophisticated, early version Stockfish we can use, because it’s that open source and #opensource license, which we can’t put into the app, but we use this early iteration and we modified it to be able to emulate in Magnuses play. I’m really not letting you talk, David, so, I’m not sure if I should just take a break. So you can, you can say something.
This is super fun to learn. Everything that I’m hearing so far at your sharing Felipe is that to make the world a smarter place, it starts with seeing where are the challenges with humans.
And we like mentally challenging tasks. And by we, I mean like data scientists and people who like code and math and, in my early days of learning, I was playing with poker and blackjack and a lot of these card games and realizing that memory palaces are so essential for understanding patterns and that behavior that is repeating over time.
And that sounds like the work that you’ve done with your core engineers and the branching of Stockfish. We talk about and code all the time with get and version control, all products, branch in different directions based on the big inspiration. And it sounds like that big inspiration was building an #app that plays like Magnus, and it’s helped a lot of players, upskill themselves.
To get to a new level, but then also to be inspired that, should we be spending all our time playing words with friends and angry birds, but, maybe there’s something more, especially during times like COVID. What can we be spending our time to learn new skills and pick up that craft? It’s incredible to see that what you’ve built has evolved from 2013 to now this whole series of apps and this whole like empire, if you will, that has expanded with the Magnus brand.
You touched on something there that I want to elaborate a bit on. We were very thorough with it, making sure that the app feels personal. So you, when you open the app, it kind of feels like you’re going to actually play Magnus on it. It’s branded. Throughout the app, with this appearance, and also this kind of chess black and white feeling. And that helps, it makes the app a bit personal. You have Magnus in your pocket, and you can challenge him. I think that’s crucial to the experience. That’s why it’s not just another chess app.
It’s a totally different experience from displaying random chess against AI. It’s actually playing against the AI as it’s known to be quite boring in general. And also if you, if you play against AI or at the maximum performance, you don’t send the town’s not even Magnus does. So, as you probably know, Magnus uses #artificialintelligence to actually train, and find out some new combinations.
It’s interesting that we look at world-class players and that’s in any sport you’re sitting in like peak performance. We see like Formula One where you have Lewis Hamilton, who’s won so many races with Mercedes training with simulations, you even had basketball LeBron James, who has always, even using virtual reality and different tools to make sure he can get that experience to be perfect. So it sounds like, as you mentioned, Magnus is like going against the top chess AI engines to see, can he break his current mind limits to get and achieve new peak performance?
And also analyze these previous games. So you can go in, and look at what could have happened if I did another move or my opponents, did this move on? How strong was it, actually like one thing as far as they can see, but how strong was it? And, I could mention that, I actually talked to minor software, an important game. And he said that his opponent probably had analyzed the position where the Lyla, and he had analyzed Stockfish, and that was a deciding factor.
So it’s okay. It might be a far stretch here, but it’s actually AI against AI at some points because they try to memorize as much as possible within the branches that they play. That’s fascinating. And I don’t know, maybe it’s a merger between machine, man and machine. So, Maybe this is how it is, it starts by expanding what we can do with our abilities with software that kind of integrates, right now with an interface.
That’s not like totally great. It’s not the implants. It’s not intuitive in the same way as it will be in the future. And, when it comes to sports and using digital tools and #machinelearning in general, I’m also working on another project. Which I actually can mention. It’s an app where you can record your tennis swing and you get a lot of attributes and measurements on how that performance is compared to a pro player.
So we collect an annual will only become better with all the data that we collect, but already it’s, we showed it to the best coach in Norway, training with the best tennis player Casper Ruud. He was, he was like, impressed. He never saw something like that. And we managed to create this and just a few months, I’m wondering what the future holds for these kinds of applications and how you are here. We’re using a smartphone. So the potential here is fantastic. I think we’ll see a huge revolution in how sports is executed based on. On what machines can learn about it. And not only humans.
You’re spot on there Felipe, tennis for example, a sport that we both have passion for, I went to Tunisia a few years ago to live a pipe dream, to see if I could qualify in the ITF tournament. And of course I didn’t do so great because I’m a data scientist, not a tennis player, but I had a lot of fun and I think what’s so fascinating is everyone wants data. We want to understand whether we’re doing good or what can we improve on?
And usually it’s a human to human relationship. You’ll have a real in life tennis coach, who’s saying, Nope, lift the ball higher. Make sure you get it centered over, or you’re not going deep enough into the court or, whatever is failing and that’s the tennis sport. Everywhere across the board, we can see that we’ve seen that in the last 10 years, the movement has been to somewhat automated machines, but they weren’t very smart.
So in sports, like tennis that we both love. So dear, there were machines that you would program them, hit the ball to these three places on the court. And I just run around like a maniac and hit it back so I can at least get my cardio in, but now I’ve been seeing even on my Kickstarter and Indiegogo, there’s new tennis machines that are claiming to use #computervision, that they can find the right spot to help me hone in on the skill. And it sounds like what you’re doing is also quite unique because we should all be able to find that coaching, but that’s not always accessible to everyone.
Exactly. It’s not always accessible. And I should be careful here, but it becomes a bit subjective, like to actually find out the parameters. Which we could use to program this thing and to do their machine learning algorithms. We had to spend a lot of time with tennis coaches and try to kind of distill what they’re actually saying. So you listen to three different coaches and they say, they say the same thing, but with their own words, and I think computers can be better at that.
They can give you the kind of the facts. And we can pre-program patterns that you can see in this session. This will improve a lot of sports in general, but when it comes to chess, there are a lot of things that you just need to be good at right there. It’s like that in every sport, but then chess, there’s a lot of mental things like a memory, of course you can’t escape it. You need to have a good memory.
But I’ve been thinking about what is a good memory. And I, it’s not, I haven’t been thinking about it because I’m especially smart. I’ve been exposed to this last seven years and I’ve seen these chess players everywhere playing. And I noticed that in the beginning, when I wasn’t exposed too much to chess, I was looking at these chess boards and I was seeing like these superhuman, attributes and patterns. How can they remember like whole chess game? How is that possible? I can’t even remember my last three moves. And how is it possible to talk about previous games and even play them?
You don’t need to be a good chess player, far above average, but still it’s doable for all of us. And the thing is I figured out that it’s like everything else, like practice, repetition, understanding because. It could be a mask, but it could also be geometry, but also a linguist or like, it’s linguistics.
It’s also a, you put a name on an opening, you put a name on a strategy and it’s easier to connect the dots. All of this kind of reveals to me that there are people that are fantastic playing some, like Magnuses. The world champion is by far a legendary player. But now I can actually understand how that tends to be on and how that’s possible and that it’s actually not being superhuman, it’s doable. And it’s something that most people can become quite good chess players. So that’s more motivating.
I know. I love chess. I mean, I don’t play as much as I wish I did, but I recall my middle school and high school days, I think middle school was one of my favorite times prior to computers.
Chess was only an in-person experience. And I just know blitz chess. I can recall it right now. I mean, it’s as if there’s this muscle memory. You’re going in for these predefined moves because you’ve experienced the over and over your predicting this deterministic or this behavioral pattern for how to operate.
And that’s where we’re seeing technology go. Today in the world of COVID we see some fantastic shows out there, like devs on Hulu and #Westworld on HBO and Upload on Amazon. They’re all talking about how much free will is there. How much can we predict or get predetermined, and it sounds incredible.
When we look at peak performance poker, chess, and other sports. Magnus Carlsen, isn’t playing 10 games a week. Like, it could be hundreds or thousands. Depending on where they are and we even see that in tennis, I remember watching Raphael Nadal, my favorite tennis player.
Roger Federer might be my truly favorite one, but there’s always contention there. And that both of them, when you see them with their coaches, they are still practicing dozens of hours a month to stay at peak performance. I think one of the misnomers in peak performance sports at Magnus Carlsen is that he doesn’t have to practice anymore. He’s just there. But what happens if he doesn’t practice? Is he going to still get peak performance?
Muscle memory has to say about it, I have my own way of looking at it, we’re creating, when you repeat something a lot, you are creating shortcuts, you’re wiring your brain to do that specific task perfectly.
You do this over and over and over and, and it just becomes, embedded in your software. So to say, And that’s basically, the language of the brain. And you can do that. As I told you before with symmetry with logic, with numbers, with music, all of that. But you can do that with feelings.
And maybe there’s a connection there, like the intuition part where you kind of feel you’re in some flow. You feel that because you’re experiencing these shortcuts, you’re experiencing that book, but you’re not consciously thinking about it. You can actually take a much simpler example than, than being world champion.
The test like when you drive a car, you’re not thinking about driving, at least not when you’ve driven for a few years. So you’re thinking about everything else, but you’re still doing quite a complex task. So it’s subconscious versus conscious. And you train your subconscious, you create this really good software packages, which are hidden beyond our understanding.
And, you’re talking about, you also mentioned free well. And the way I see it, it’s like we, of course we don’t have free will, it’s all deterministic, but we experience it because I don’t know a better, we just, there’s an emerging thing going on and nobody knows exactly why, that’s going on.
And, some believe that it’s everywhere, like, panpsychism and stuff like that. And that’s up for debate. But obviously it’s deterministic, maybe probabilistic if we want to go into quantum mechanics and all of that. But, actually I lost the thread. What am I talking about?
This is so interesting though, because I got us on this tangent about devs, Westworld and Upload. What is that fluid behavior you’re talking about getting into flow and flow is so important for learning anything. It sounds like a lot of the world today, as we know, learning is not just about project-based learning, but, love it or hate it. There’s a lot of credentials out there and you’re having to test and be proctored and being monitored for results.
Although we’ve seen school systems like the University of California drop the need for the SAT and certain MBA programs dropped the need for GRE and GMAT. Proctoring is still very important. And I wanted to explore a little bit about this area as well, for what you’re seeing in the world of chess. It’s all moved online, Magnus is leading the charge, but one of the concerns we always see is, are people being honest? Are they using a simulator or just, going forward, whether you seen with these online chess tournaments and how we’re ensuring people are keeping honest.
In chess, you have a few mechanisms for anti-testing and they’ve been around for a long time. Obviously you can actually, if you have the game, you can do these algorithms, which we’ll find out if you’re playing too similarly to one of the popular chess sentences and that we detect tickers since I study for all the time and also even in Play Magnus.
Because we have these yearly competitions. Of course, some of them were teaching a lot to get help scores, but that’s kind of the digital aspect of it. Of course you can, in a tournament when you’re sitting at home and playing others, you can be teaching. You need to have your webcam on to play in one of the Magnus invitational, which was the most recent one, but also you have to, we ask them to install another camera which is behind them to monitor what they’re doing.
I think with newer technology, we also start trusting people more. I could be wrong here, but if you’re in the chess tournament, they, you need to go to the bathroom. You can’t take your camera with you. So at some point we worry less about these things. It’s a good thing. And it’s happening also with employment in general.
Like everyone works from home now. Like we knew that was going to work fine, but a lot of people were surprised like, Whoa, this is actually working and my employees are performing maybe better. Like how can that be? They can do whatever they want. They were working at home. And of course they are, I think freedom for most people gives, like responsibility, which is good.
And also people are different. Some people have teens staying awake at night. People are absolutely different than creating an environment, which makes them thrive in this, in their hobbies or their work in the best possible way. That’s fantastic. If you’re a manager listening to this one and you’re thinking, but I can’t trust them, there’s always a few people that will take advantage of the situation.
Then I would say, but do you actually want to have those people around, like, do you want to have them in your company? Maybe the hiring process should be better than, and there will be less and less of those people because the culture will change. This will be the way to work moving forward. Not only because of COVID, it would be because it’s more effective and it gives this type of freedom.
It’s so interesting that, of course we’re spending a lot of the conversation seeing how chess is being reshaped in the world, being digital first with the Magnus invitational. But it’s more than that, it’s work. I mean, we are moving to a distributed society. Post COVID world will be one. We’re sure there will be some in-person. There will be some distributed. I think that will. Still be to be determined how that will shake out for different cities. But I think what’s not as unclear is that we are moving more into a device first world.
And we’ve been spending a lot of time on the show today talking about chess and the whole evolution of the sport, but you do a lot more than Play Magnus. You’re also involved, leading up new tech ventures. And I’d love to hear about whether some of these projects and ideas that you’re incubating at Solv Oslo.
There’s a lot of cool stuff happening. And, we see a lot of different projects that we can contribute to, but we’re also starting from scratch on many of these. So, as I said, we do both consulting and product development in house. One of our cool engagements has been to work with, it’s like a camera that can scan your eye and recognize.
And recognize patterns on diabetes to people with diabetes too, and therefore find out if you’re becoming blind. So a blindness prediction, basically. That’s interesting. This doesn’t have to be expensive, the quick ones, because it’s already, it’s a solved problem with quite expensive equipment with equipment that could cost like a hundred bucks.
It’s not, and we’ve managed to do that. We’ve helped the company develop that. And, it’s only a matter of time before that’s produced and then shipped off to all over the world where we’re, where it could be expensive to get your eyes scanned by hospitals and so forth. So you could basically do this at home.
It’s so interesting thinking about cameras, because I mean, there is camera technology that’s been out for a while to forensically confirm that a painting is authentic or that the fabric of a handbag that’s a Louis Vuitton is real as well. We featured on an earlier episode of HumAIn a founder, particularly who works in that space. But I wonder in that case, they said we’re actually able to use the cameras from the IPhones and the Samsung Galaxy devices to get that quality. Do you see a future where we can be, holding up a phone to the Iris and scanning and seeing that performance, or, what might those machines of the future look like?
Absolutely. I see it on and we experienced it like every year, these new phones come out. Then when we were doing this tennis project, where were you supposed to record a tennis player and then see how the swing is being executed on how well the performance is compared to a pro player? I actually didn’t know that you could record a slow motion, 240 frames per second, on a 10 ADP on most IPhones today.
And that’s insane. Like the opportunities we were thinking like, can we actually execute this stuff?, 30, 60 FPS, but, it does not feel that way reminded to pull it off anyway, on 60 Fs. But, the opportunities are huge. Wearables will be like, it’s, we’ve seen it, but it will be a huge thing in moving forward on.
There are so many health benefits you can get. I also think people will be more, exactly right now with COVID-19 was maybe not the best location to say so, but in the future, we still will have a situation where, where people are less ill and things are going quite well, health wise. So you start thinking about longevity instead of just survival. All these #wearables will contribute to that. So you’ll be much more occupied with health in general and not only physical health, but also mental health. At some point we’ll figure out how to come closer to the connection between a scan and psychological states of the mind.
Mental health is going to be just as important as we’re emerging in a post COVID world. I recently shared my top 25 trends for Q2 2020. Whereas looking at tech investments as an investor in this space and mental health tech is one that I’m very bullish on. We even see devices in the United States like Muse, which helps you meditate. We have the Aura Ring where you wear this ring that can track the pulse for you, for your heart as well.
So there’s so much going on in the medical space. What do you think Felipe is helping accelerate this? Is it more the AI and this breakthrough in computers?. Is it the capability to get sensors so small? I mean, what are some of the things you’re seeing, that’s getting interest and traction in this space.
That’s a hard question to answer. There’s so many factors, but of course, it’s hardware gets, quite improved every, every year. But it’s not, in the same way as before where you, for instance, the graphics on games this insane already. Of course everyone was hyped. When they saw the PlayStation 5, tech demo, unreal demo but really, there’s so much more now than just graphics. We’ve come to a point where we’re wearing the technology that we previously had in a living room, and it just, or, in our offices and we’re huge. So we’re just wearing it.
There most people will have smartwatches and bluetooth that devices all over their body. And of course the smartphone, but that will shrink, or at least become thinner and thinner up to a point where it’s maybe a bracelet or something and you can bend it. You can, for all, I know you will swallow your smartphone at some point. #Nanotechnology will at some point become cheap to use and the manipulation of genes, it’s all there, but there are so many sciences just expanding exponentially.
We were used to it now. So, we don’t think so much about it. We’re used to seeing new stuff. That’s why we can sometimes have a feeling that though. it’s not going that fast, but it really is if you pay attention and if you, if you look up all different technologies being expanded, it’s a lot. I don’t have a better answer to the thunder.
No, I think that is so perfect because at the onset of our show today, we were talking about Github branching and how software branches into different versions. Looking at how Stockfish and the core engineers at Play Magnus have developed new technology to see the direction of AI.
That’s exactly what we’re seeing with all these hardware and software products today. You mentioned Felipe the PlayStation Five, which is coming out. We know the Samsung Galaxy S 20 Ultra, the iPhone 12, even the unreal engine five. All this technology is accelerating at such a fascinating clip.
And part of those trends is collaborative science publications, no longer are we at the level where it takes 12 months. Two years to get published and see research. COVID alone we’ve seen what it’s like thousands. I saw the numbers somewhere 80,000 papers around COVID who knows if they’re all valid, but the point is there’s so much research accelerating and that’s leading to us a world.
I think that’s going to be humans augmented by #machines. Whether it’s playing chess, whether it’s wearing a device to detect type two diabetes so that you don’t get COVID. I know at least in New York, we saw some statistics, some early indicators that about 25% of the people who have died from COVID are type two diabetic. It’s so shocking that this is a big problem in the world.
That is, and we can do something about it. Then I also think these problems are, it shouldn’t be a problem that one company is solving by themselves. Of course they could be developing a product, but they’re, but it’s, we need to become more connected on these things, especially on med tech.
And I think we’ve with COVID-19, we’ve actually seen that people are starting to understand that collaboration across a huge distance is not it’s the same as like, if you have an office in New York and in Oslo it doesn’t matter if you have 50 people in Oslo and 50 in New York, you’re a hundred people.
That’s the size of your company. So, collaboration of course is important. And throughout the evolution of human beings or like history, it’s always been there. There are more and more people collaborating. So it starts with tribes and, then small cities, doing this quite fast. And then you have countries like, continents, etcetera.
At some point it would just become global, like pure global collaborations, especially on science, no borders and full transparency with the especially weekend with the common goals of eating. All diseases, which I think will happen at some point, of course viruses are tough, but it will, it will happen.
It’s doable. It’s not an impossible task to execute. And I, as I said, when we do it, that it will be more about longevity, which I also think is important. A lot of people die of AIDS and it’s proven that it’s dying of AIDS is not something that needs to happen. Degrading cells and your whole body is not something that needs to die eventually sort of to get deprecated basically. That also will eventually come, not sure when, but there’s a lot of people working on that. So might be faster than we think.
Absolutely. It’s the simulation that we’re living in that many of us think is this world of COVID, but we’re moving into a post COVID world very fast, the economies are reopening, in Norway, it is business as usual in the new normal, the United States has been moving into its new normal, gradually as well. Felipe, what call to action would you like to share with our listeners today about the topics we’ve discussed and, what do you think is important for everyone to hear?
There’s just so much I could say, stay healthy. That’s one. And, I still remember to stay a bit more at home and keep distances and all of that. I think that’s important. Of course, we were getting to a stage where it’s not as dangerous as before, but, still there you have other people’s life on risk here, if you’re not careful, that’s a number one, but I also have general advice and that’s of course, to try out or Play Magnus app and also Magnus train around all our apps and have fun with them.
They’re free. So just play and have fun and see if chess is something for you. If you’d never played chess before you could try out the Magnus trainer. That actually teaches you chess from the very beginning. Like you don’t have to know how to move pieces to play that game. So stay healthy and play chess.
Chess on, whether it’s on play Magnus, chess.com, or maybe you participate in a tournament. With chess 24, you could be in the next Magnus invitational. It’s incredible to see how we’re bridging the gap of the digital and in real life world, Felipe Longe a CEO at Solve Oslo and CTO Play Magnus. Thank you for joining us on HumAIn.
Thank you for listening to this episode of the HumAin Podcast. What do you think? Did the show measure up to your thoughts on artificial intelligence, data science, future of work and developer education. Listeners, I want to hear from you so that I can offer you the most relevant trend setting and educational content on the market.
You can reach me directly by email at firstname.lastname@example.org. Remember to share this episode with a friend, subscribe and leave a review on your preferred podcasting app and tune in to more episodes of HumAIn.