Kostas has tons of experience as a product manager in companies such as Microsoft, Skype or BBC and he’s now building the world’s most advanced Artificial Intelligence Fashion Stylist for retail.
Ciprian Borodescu: I’m here with Kostas Koukoravas, Founder and CEO at Intelistyle. I’m super excited, and it’s an honor to have you on this podcast. Thank you so much for being here.
Kostas Koukoravas: Thanks, Ciprian. It’s an honor to be here with you.
Ciprian Borodescu: Over the years, you’ve gained tons of experience as a product manager in companies such as Microsoft, Skype, or BBC, and you’re now building the world’s most advanced artificial intelligence fashion stylist for retail. I want to start by inviting you to briefly talk about your journey, from those early days to now being the founder and CEO of Intelistyle.
Kostas Koukoravas: Alright, thank you. Well, I guess I started writing code when I was little, still in primary school, I can’t quite remember the age. You know, all my friends played video games on Amstrad if you remember them. But, you know, I preferred to learn the basics of rudimentary programs. Change the color of the screen. Yay! Yeah, during high school, I then sort of started moving on to Pascal, I wrote a bit of assembly, a bunch of different things. But I ended up studying business, actually. My real passion was coding, so I managed to find an internship through my university abroad and I went to my first coding job, and did everything – from databases to front and backend, object-relational mapping systems, which were new at the time; I know they’re pretty standard these days. The whole thing.
Ciprian Borodescu: Yeah.
Kostas Koukoravas: So then I moved on to product management, doing a couple of different jobs, ended up at the BBC. And, at the time, what we were trying to do is take a traditional experience – radio – and make it relevant to today’s age of YouTube and short-form content. And for me, working with music was like a dream come true. We made the first version of I Play radio, bringing together content from 57 different radio stations, who had websites of the time, but not a real product. And then I moved on to Skype and worked on various products. I launched Skype on Windows 8. And before you say anything, I know everyone hated that. But, you know, we then moved on to a few different other products, and one of the projects that was quite interesting was Surface Hub, which is a fantastic device for video conferencing. It’s got dual cameras, directional microphones, and they can find out using machine learning where the action is in the room, and focus there. You know, the action might not be at the place that you think it is or the place where there is sound because somebody might be tapping their pencil or something or might be scratching a piece of paper. So, it helps you understand where the action is and effectively films the whole meeting like a cinematographer. So while I used machine learning in my Master’s to model agent-based behavior in artificial economies, that was really my first commercial experience. So eventually, I kind of decided to solve the lifelong problem that I have – I am absolutely terrible at putting an outfit together. So, you know, at the time, I had my girlfriend who would help every now and then, but she wasn’t that patient, you know. So I thought, you know what? I can just use AI and replace all that and make it happen a lot easier.
Ciprian Borodescu: That’s exactly how an engineer would think about solving this problem.
Kostas Koukoravas: Exactly. Having a girlfriend… I just need some AI. So yeah, that’s how Intelistyle got born. So initially, it was an app. It allowed you to star your own clothes or pulling new clothes that matched what you owned. And yeah, eventually we saw a lot of interest from retailers who wanted to use the technology for their own websites and apps. So, we decided to pivot and focus on helping them personalize their product offering by helping customers find the right clothes and outfits for them.
Ciprian Borodescu: Excellent. And you’ve been involved with a lot of product teams over the years. What are some of the most important roles you believe such a team should consist of, especially if it’s an AI company at the core? And maybe talk a bit about these teams as relative to the stage of the company from startup to scale-up company, to an enterprise the size of maybe Microsoft or BBC?
Kostas Koukoravas: Okay. Look, I mean, you can have a million machine learning engineers, but in every AI project, you need a subject matter expert. So, when we first started Intelistyle, we were just a bunch of guys – initially, one guy, me, then more of us – and we were all trying to solve the question of style, which, you know, I mean, you can imagine three guys, engineers, trying to solve the question of style – absolutely hopeless! So soon, we brought in clothing designers, stylists in the team and that really helped us shape the product and bring the right expertise in. And eventually, it got to that level that we managed to build an AI that beat the human stylists at London Fashion Week. Now, these subject matter experts really ended up taking the form of product managers in the traditional kind of agile sense, in a Scrum team. What I saw in the past is a lot of teams making the mistake of putting someone with previous product management experience in that role, maybe someone technical; and I think, they’re certainly needed roles, and they can help a lot, but you really need the subject matter expert working directly with the engineers. And that’s what we did at Intelistyle, but similarly at Microsoft, when we were trying to solve the problem of, you know, how do you create a cinematographer for a meeting, we had UX designers work directly with a machine learning team to decide exactly what was the problem that we were trying to solve? What were the different scenarios that we had? You know, the machine learning engineers would feedback on, hey, these are the limitations, this is what we can do. And then we’d have this iterative process where we’d try different scenarios, we’d see what the model can do, and change things again and again, and again, either in the model or on the user experience, trying to solve the problem. And, you know, that iterative cycle got really impressive results.
Kostas Koukoravas: So, you know, while I do think that the subject matter expert is a key person in a team like that, I think it’s really important to have a strong Scrum Master as well, in the agile Scrum sense – understand the principle of Scrum, can guide the team towards best practices, it can help them do the retrospective and plan it properly, and discuss requirements, probably write them and measure and plan and set KPIs – all of these good things that you need in a strong product. So, you know, in a smaller team, I guess, you know, initially, you might just start with a machine learning engineer who deploys the modules as well. But I think, eventually, you want to get machine learning engineers on what they do best, which is research. These people have a really mathematical mind and the mathematical mind and algorithmic mind that you want from a backend engineer is not always the same thing. So I think, eventually, as the team grows bigger, you definitely want backend engineers to handle the deployment of those models and making them more robust for production, all the security considerations that you need to have, all of these things.
Ciprian Borodescu: Yeah. That’s such a good point on having a diverse team with subject matter experts and so on, not only machine learning engineers or not only engineers. That’s supercritical and I’ve seen that with us and with other startups and businesses. That’s critical.
Kostas Koukoravas: For sure. I mean, if you see the difference from when we didn’t have a subject matter expert, where we had to figure it out ourselves to when we actually got one, we managed to make huge leaps.
Ciprian Borodescu: Yeah, yeah. And, however, sometimes I wonder how can AI startups succeed with limited access to funding and talent, especially in the AI space? I mean, what are the levers that we can pull to kind of push through and reach a point where AI is no longer considered elitist? Because even today, I still feel AI is a little bit elitist. Because right now it seems there’s a huge gap between AI tech giants, the FAANG or GAFA – whatever you want to call it – and the rest of us.
Kostas Koukoravas: That’s a very good point and one that we struggled with a lot when we started. I guess there were less frameworks that made machine learning models readily available to people, and we did spend a lot of time building our own model from scratch. I actually think the barrier we were trying to solve was quite challenging – I don’t know how easy it would have been to get something that was a little bit more packaged, a little bit more off the shelf. But yeah, when you’re trying to raise funding as a startup, you’re faced with a problem of, you know, investors will expect significant traction from you. And, you know, the amount of investment that you have to do in technology if you’re an AI startup is absolutely huge. So, it’s very difficult sometimes to meet those requirements. And a lot of investors might not realize that. But I think these days, technology has moved on and there are a lot more platforms available that do allow you to quickly iterate and they will make some models available off the shelf to you, you’ll be able to train them very easily, the process of training and building data pipelines. It’s a model for everything.
Ciprian Borodescu: Yeah, at least you have a baseline.
Kostas Koukoravas: Exactly. And if you can try 10 different models really easily and find which one is right for you, or gain some insights into what you’re trying to do, build a very basic prototype that you can put in front of customers and see how they react, even if it’s not perfect. And, you know, what you get if you build something completely from scratch, if I can help you get 80% of your way there – or even less – I think the value to cost ratio definitely helps and it helps you accelerate that process. And that’s mega important when you’re trying to prove a concept and you want to spend money on marketing, you want to spend money on a number of different areas and not just technology.
Ciprian Borodescu: Yeah, yeah. It’s such a good point. And speaking of proof of concept, what are some of the challenges you’ve seen when taking an AI project from the proof of concept to production?
Kostas Koukoravas: Okay, I mean, even if you do every effort to do it by the book, use exactly the same data to train your model that you’re expecting the same production, at least, we found in our case, that the data will be slightly different in ways that maybe you didn’t imagine, in very subtle ways that could potentially have a huge impact. So, the very first version of the model that we trained was all about helping you sell your own product and your own clothes. And we didn’t use our own photos, we used user-generated photography. But what we didn’t realize was that the user-generated photography that we took – and, you know, we used it to an extent to train the model, or the one that we found online – was not quite the same as what a user would do, the type of photo that the user would take. So I think what end users do always surprises you and it’s good to solve some of these UX considerations upfront, even by putting smoke and mirror prototypes in front of users just to see what they’re going to do. So, you know, data can catch you out many times. I think when you’re trying to move from proof of concept to production, one of the things that might fall through the cracks is responsiveness. People, when they press a button, they expect instant results. And yeah, sometimes ML models don’t run instantly. They need a little bit of time before they calculate the results. You definitely have to think about that. And what’s the hardware that you’re going to run these models on? What’s the cost of the hardware? GPUs on the cloud tend to cost. Can you run your models trained on GPUs rather than CPUs? How can you scale them? What happens if you need multiple servers to run the models on to scale? So there’s a lot of considerations like that. Sometimes you might be tempted to just jump ahead and say, “Hey, I’ve got the model. All good. Let’s go live.” But no. You’ve got to think about these.
Kostas Koukoravas: I think, for me, the biggest challenge was the one that we saw, was that even if your results are cutting edge in terms of research, and you’ve implemented overlays papers and your model gives better results, and you’re celebrating, those results don’t necessarily mean that they’re good enough for production or they’re good enough for the business. You know, in the example of styling, if you tell someone that you’re going to style them and you fail, even a little bit, make a little mistake, and you lose their trust, that’s it. It’s very hard to regain that trust because it’s all about how you make me feel and my confidence, and when you play with my confidence, obviously, I don’t like that.
Ciprian Borodescu: That’s such an important point. I mean, it’s really emotional, right?
Kostas Koukoravas: Exactly. And when you look at the numbers, you’re like, “Yeah, like, I’m better than all the research.” But is that good enough for customers? And sometimes you can forget that and get carried away by the fact that your results are good. But sometimes AI doesn’t really have the maturity to go to production and you’ve got to think about the level of investment that you’re prepared to make. I mean, we see the companies that are involved in self-driving cars are making huge investments and having the luxury to wait all these years to get the technology to that stage, as a startup – as a smaller startup because we don’t all have three exits behind us and can raise millions without having a proof of concept – as a smaller startup, you need to drive results now. Your investors are expecting results now. So you need to focus on a lot more solvable problems, as opposed to big shots that need significantly more.
Ciprian Borodescu: And I think we already touched on the do’s and don’ts of building an AI product. It doesn’t matter, necessarily, the performance of the model. The focus on the tech is not that important if the business is not validated.
Kostas Koukoravas: I totally, totally agree. I mean, my motto is simple. Do you really, really need to use AI? That is the first question that I ask them. If somebody comes to me with an idea and they’re like, “I’ve got this amazing idea. We’re gonna use AI to do it.” I’m like, “Okay, just wait a minute.” Have you yet figured out how you’re going to do this without using AI? Once you figure that out, and then you manage to scale it so much – because actually, you’ve got so much demand that you can scale it enough – then you can start thinking about how can I chip different bits of that process and automate them. And I’m probably going to start with the easiest bits and not the bits that need a lot of human brainpower. As opposed to saying, “Right now I’m gonna sit down and have this idea with machine learning all the way from the beginning.” You don’t even know what features your machine learning model should have and what should you be looking for if you don’t know how to do it manually?
Ciprian Borodescu: Absolutely. But this opens up another topic. And it seems that there are two schools of thought when it comes to communicating about your company’s AI capabilities. Either package it as a solution to a problem and don’t communicate at all that there is an AI involved behind the scenes, or, on the other side of the spectrum, yell from the rooftops that everything the company does is AI or machine learning. Where do you stand on this spectrum? And why?
Kostas Koukoravas: Yes, everyone likes to shout it out these days.
Ciprian Borodescu: Yeah, it’s our new shiny thing, right?
Kostas Koukoravas: Exactly. And I think just because everyone’s trying to kind of shout about it, there is a lot of noise in the market and it’s very difficult, we found, for clients to understand who’s truthful, right? And, you know, loads of clients might say, “Hey, you know, everyone tells us they’ve got the best AI, but how do I really know that?” So I don’t think is that an effective communication strategy. I think any product that you make really needs to be solving a problem. And first and foremost, you need to know what that problem is and communicate how your solution solves that problem. Number one. Now, you know, you don’t really use AI for the sake of an academic exercise, right? Now, I think where you might want to talk a little bit about AI is about the particular benefits that AI might bring into the future. So while before you might have done this thing manually, hey, now we can automate it, right? Or while before it cost you a lot of money, now, we can make that a lot cheaper, or you can scale it a lot more, or you can analyze a lot more data, and it’s a lot more flexible to adapt to different situations. So I think you need to talk about the benefits that the AI brings into the mix in relation to a real customer problem you are solving.
Ciprian Borodescu: Yeah, that’s a good point. Speaking of which, tell us a bit about the AI Styling Showdown you posted some time ago on LinkedIn. Basically, you put your AI to the test against Zalando to see if your AI can create better outfit recommendations than Zalando. Where did that come from? And how did it go? What were the results?
Kostas Koukoravas: So yeah, we noticed quite a while ago that Zalando developed similar tech to ours – it might have been at least six months that we’ve kind of noticed that, and they quite openly advertised it in conferences and events and talked about it. And they’re not direct competitors, but we were always very curious to see what is the quality? How does that compare against our tech? And then Coronavirus struck, we were all locked up at home, so we did end up having a little bit more time in our hands and thought, “Hey, why don’t we try doing something that we haven’t done before? One of the things that we always wanted to do.” So we decided to put the results side by side, just to see what we think, what the customers think, what prospects, what different people in the industry think, from execs to stylists. And, of course, we’d hope that our model would do better. But we didn’t really expect anything of the likes. We found that three out of four users prefer the results of our model, which was huge. So we had three times as many users. Yeah, I know, you’ll say “Of course you will say that”, right? Because we rerun the research. So, of course, we say that. But we wanted people to see the results for themselves, so we actually opened up the data, and we did an analysis and we saw, okay, we did better in different categories and Zalando did better in other categories. What was going on there? So we created a lot richer outfits, we went up to six items while Zalando was doing about two to four. And the reason why we ended up being better was because a lot of customers, subject matter experts, and a number of luxury brands have said, “You know what? There’s not many items. Not good.” So we actively put a lot of effort into that, in the richness of the recommendation. And then we looked at different patterns, right? So we found that we did better at floral prints and striped patterns. But in order to sort of show people and open it up even more, we changed the game. So, while before you couldn’t tell what you were choosing because we wanted to have a blind test, we then opened it up completely, and now you can go in and see the results for the same item from Zalando and the result from us so that people can judge for themselves. So, it started for us as a little game, I guess, and then we opened it up to everyone.
Ciprian Borodescu: Okay. And speaking of this crazy period and the Coronavirus, I think from a digitalization point of view it’s almost as if we jumped from 2020 to 2030 in a matter of months, and I’m still trying to figure out the implications of all of that. On one hand, if we take the retail, for example. I mean, clearly more and more businesses have today an online strategy versus a year ago – let’s just compare it with 2019. At the same time, more and more consumers are buying online. So, long term these trends will hold. However, I can’t help but wonder who are the real winners here. And what can we learn from this forced pivot that took all of the businesses by surprise? And it kind of pisses me off that it took a pandemic of this magnitude to force companies, let’s say, in the retail space to prioritize digital transformation. And, you know, AI is part of the digital transformation, but it’s the next step. And I wonder, do we need to go through a similar painful experience for companies and people to seriously adopt and embrace AI? I mean, did we really learn our lesson?
Kostas Koukoravas: Yeah, you’re right. I actually think that for any big change, you need a painful experience or a shock to the system because, by default, if people are comfortable with where they are, I think the majority has little incentive to do anything. And I think that’s a lot of what we saw in the retail space. And we have a number of prospects that completely froze up with Coronavirus, and they said, “Look, we’re holding our investment, and we are getting into preservation mode.” And, you know, you can’t hold it against them because it’s been a huge problem for retail. You plan your season, you’ve made all this investment in stock and suddenly, you have your warehouses full of stuff that you can’t sell, you’ve got all that capital tied up in stock, where you can’t buy new stock, you can’t use it for anything else. And, at the same time, you’re paying to store that stock. So it’s a huge, huge, huge, huge, paralyzing problem for many, and your only option is to heavily discount it and either sell it at a loss or a significantly smaller margin. On the other hand, we did see companies that maybe hadn’t moved for a while, you know, maybe for a year, just completely accelerating and things happening in a month or two, which is lightning speed for retail. And I think these are the winners, right? The companies that do understand that you need to move fast and see what’s happening out there as an opportunity to do things faster, change the way to do things, and respond to shifting needs. And, you know, some might say that’s an opportunity for a land grab, right? Because some companies are not going to be that active, they’re not going to take advantage of the opportunity, and therefore, now is a good time for us to move quickly. A lot of companies are thinking that way and I think that’s fantastic. These are going to be the winners. And, of course, on the other side, we definitely saw companies that were traditionally online, and they had a strong online presence just benefiting further. Of course, the obvious example here is Amazon that has done really, really well. And, you know, all the other tech giants, really, were by default digital, and they’ve all done really well during this pandemic.
Ciprian Borodescu: Even Shopify. And a few weeks ago, BigCommerce went IPO.
Ciprian Borodescu: Exactly. Yeah. Truth be told, we’ve also been approached by companies that were already online and they had a strong foothold, and once they saw that offline retailers also started having an online strategy, they were like, “Okay, now we need something else. We need a competitive advantage. Perhaps this AI thing might be something interesting for us.” So yeah, we definitely had these kinds of conversations. One of the things that I wonder about myself as a founder and as a CEO, and I just wanted to also pick your brain on that; how did you internalize – you and your team – these first six months of this year? Was it peacetime? Was it wartime? Maybe hybrid. You know the dichotomy, peacetime CEO or wartime CEO? Where do you see yourself?
Kostas Koukoravas: I think it was a time for soul searching and definitely exploring all the options out there. Because I think for the first time, we have that luxury to take a step back from the business and say, “What are we doing right? What are we doing wrong? You know, should we be in this market? Should we be in a different market? Are we targeting the right customers? You know, what will the future look like?” Because, you know, at the time you’re focusing on operations and getting things moving. But now is a good time to completely reset things, and reevaluate what you’re doing. And we did that, we spent a lot of time doing that. And I think that led to two things. One, it really reinforced our belief that this is the right market for us. And that really helped us give a new energy to the business and a new determination. So, in a way, that was really energizing, despite everything that was happening. And on the other hand, it helped us look at different aspects of what we’re doing and to build a new product that adapts to these new times, because, you know, standing still these days is moving backwards, effectively. So we’re now building a new product that it’s all about remote selling, which is something that we hear a lot from our customers, right? You know, with shops closed or with customers reluctant to go into clothes shops, how do you utilize the resource that you have there to provide excellent customer service? And how do you make interactions with customers a lot more unique and a lot more memorable? And how do you drive them across different channels? So we’ve used this time to explore this area and I think that’s going to be key for the business in the next few months and longer-term.
Ciprian Borodescu: Yeah, it makes a lot of sense. I wonder about your hiring process when it comes to looking for AI and machine learning engineers. Is there anything special in your process when hiring for that?
Kostas Koukoravas: It’s a fantastic question. We started the journey by hiring machine learning engineers with very little experience, but we learned very quickly. And one of the things that we noticed is that a lot of people in the market, you know, maybe with a lot of experience, they might not have the latest skills, the skills that you need – and literally the skills that we needed. Because, you know, machine learning is a very new field. And, you don’t look at the news in the field for a week, and you feel like you’re now years behind; things move so fast. And we found that a lot of people in pretty good positions, working for great brands, having great brands on their CV, really didn’t have the skills and didn’t really know those latest models that we needed. So, we found that there are some fantastic brands out there that do follow the latest developments and have a real passion for it, and that’s who we prefer to hire, especially people that have practically shown that they spend time to learn things. And so, you don’t need to have a great CV in terms of work experience, but if you worked on some cool projects, and you can talk about these cool projects, you didn’t just copy the code from whatever repository and you show that active engagement and passion, I think that’s what really sets out an exceptional candidate from an average one.
Ciprian Borodescu: Exactly. It’s not only the technical skills or abilities, right? It’s also the attitude and the soft skills, basically, and the willingness to commit to working in a startup because it’s a different environment, isn’t it?
Kostas Koukoravas: Oh, yeah, for sure. And I think you’re touching on a very interesting point there. It’s about employing people that are willing to learn because, in machine learning, you will have to be learning every single day. So it doesn’t really matter what you know; what matters is how quickly you can learn things. And, yeah, I think, again, you hit it right on the head there – working in a startup is very different from working in a corporate, and you need that passion.
Ciprian Borodescu: And what about the diversity topic? Where do you see things heading? Personally, I feel like there are more and more women involved in STEM, which is absolutely amazing. However, it still feels that we need to be intentional about it and be on the lookout for women that are interested in a career. At least, this was my experience at Morphl. And I’m just wondering, based on your experience, how the ecosystem feels, you know, in the UK, or Asia versus maybe Eastern Europe?
Kostas Koukoravas: I do think that it’s still a very male-dominated space but we increasingly see more women sort of getting into it. And, you know, we’re lucky to work in the area of fashion that traditionally draws more women. We’ve brought some very, very capable engineers to the team, that I’ve had the pleasure to work with. But, I think one of the things that we’re seeing is that these things naturally take some time, right? Because there is that sort of education piece, right? That this space is more accepting and unique people can go through university, you need to start early talking to people and educating them about these things. And, you know, there’s now a multitude of games, programming games for teenagers, which hopefully are getting into schools, parents’ perceptions are changing. So, you know, more girls are getting into them. And, you know, I think that naturally, we’ll see more and more and more people, more and more girls, females getting into the space, which is something that you need. You know, I feel like if you have a team just with guys, it’s not as interesting. You need that variety, you need the different perspectives.
Ciprian Borodescu: Exactly.
Kostas Koukoravas: You need that emotional intelligence. And that’s not just with girls and guys, it’s with everyone – you know, people from different cultural backgrounds, societal backgrounds, ethnic backgrounds.
Ciprian Borodescu: Absolutely. Yeah, yeah. All right. Well, for the final special section of the podcast, lightning questions and answers – a series of fun, short questions that you have to answer really, really fast. Are you ready?
Kostas Koukoravas: All right, okay. Go.
Ciprian Borodescu: Game of Thrones or Lord of the Rings?
Kostas Koukoravas: Game of Thrones.
Ciprian Borodescu: Star Wars or Star Trek?
Kostas Koukoravas: Star Wars.
Ciprian Borodescu: Okay, favorite movie?
Kostas Koukoravas: I don’t watch movies. I’m a terrible person, I know. I love Peaky Blinders, though – the series.
Ciprian Borodescu: Okay, okay. Is that on Netflix?
Kostas Koukoravas: I think it’s on Netflix, yeah. I think it’s a BBC production on Netflix. Very good. Very good.
Ciprian Borodescu: Yeah. Okay. Cats or dogs?
Kostas Koukoravas: Hard to choose, right? If you asked me a while ago, I would’ve said dogs, but now I say cats because they’re low maintenance and I haven’t got time.
Ciprian Borodescu: Exactly. What was the last book you read?
Kostas Koukoravas: I actually just finished one – Perfume: The Story of a Murderer by Patrick Suskind.
Ciprian Borodescu: Okay. Was it good?
Kostas Koukoravas: Very good. Classic. I mean, it’s exploring so many different aspects of love, and it is beautifully written, and such a complex book. I thoroughly recommend it.
Ciprian Borodescu: Did you read that on your vacation?
Kostas Koukoravas: I did read it on my vacation. I actually heard it on Audible because I find it hard to read books now.
Ciprian Borodescu: Yeah. Nice. Well, you just mentioned something during the episode and I just want to introduce a new question here. Microsoft Teams or Zoom?
Kostas Koukoravas: I know my old team is gonna hate me for this, but Zoom right now is doing a pretty good job. But Teams has a big vision.
Ciprian Borodescu: Awesome. Okay. The bonus question. We’re both TechStars alumni, we have been through TechStars Montreal AI – different cohorts. Who was your favorite? Bruno or Justine? And I promise I won’t tell.
Kostas Koukoravas: That’s a hard one. Look, I think they were both special in their own ways. You know, they’re like Ying and Yang. You’ve got to take both.
Ciprian Borodescu: Absolutely.
Kostas Koukoravas: They’re connected in this weird way and it’s impossible to separate them.
Ciprian Borodescu: And you know they actually have a name for that, right?
Kostas Koukoravas: No, I didn’t.
Ciprian Borodescu: It’s Brustine.
Kostas Koukoravas: This is amazing. I’m gonna tell them next time I see them.
Ciprian Borodescu: Awesome, man. Awesome. Kostas, this was so much fun. It was a pleasure to have you. Thank you so much for sharing your wisdom with me and with us. How can people reach out to you for ideas and comments?
Kostas Koukoravas: Thanks for having me, man. That was amazing. I had some good fun. So yeah, if people want to reach out, they can reach out at firstname.lastname@example.org.
Ciprian Borodescu: Awesome. Thank you so much, man!
Kostas Koukoravas: Thanks, man.