Ep # 143: Building Climate-Resilient Cities with AI ft. Angelos Chronis (CEO, Infrared City)

Adil Saleh 0:33 Hey, greetings everybody. This is Adil from happening podcast, long time coming for a lot of conversation that that were still in the in the queue for different niches, different industries we are exploring with AI and how AI is making a huge impact in the adjacent industries. You know, we previously talked about manufacturing, oil and gas, Energy recently had some enterprise segment. We explored a lot like how AI is transforming the internal tooling, internal processes and all. And it is my pleasure today to have Angeles, who's the Chief Executive of Infrared City. That is basically a climate simulation. They're different. That's why we were talking today and and, you know, we're trying to explore that, how this industry has been impacted in a positive way, and for all the good and bad of AI that we we see everything every, every day on the internet. And you know, it was, it was a real opportunity for me to, you know, speak in depth with Angeles today. Thank you very much Angeles for taking the time. Angelos Chronis 1:33 Thank you for having me. It's great to be here. Adil Saleh 1:36 Likewise. So Angeles, I was also, like, getting notes from the team. And I was looking at some of your prior experiences. I know, like, you spent like, good, almost like, a decade in the in the Advanced Institute of, you know, architecture. You know that is like we would love to learn more about, like, what entails your your part of a senior faculty and, of course, a PhD candidate. So how did this all come together, even from before, like back in 2019, when you were head of city Intelligence Lab, how did it all like shape you to build a platform like Infrared City? What kind of problems have you seen throughout this journey? I know that it's half of our lives, you know, almost 15, 16, years, you know. So, you know, we all would love to learn like, how from someone like with that background can make an impact and then come at an age in the late 40s, or, you know, early 40s to build a platform like in Infrared City and make this huge of an impact. Angelos Chronis 2:35 Yeah, thank you so much. I think I'm gonna have to show my age at this one, but it's true that it's been about 16 years now of development. I'm an architect. I studied architecture, and I'm from Greece, and that's where I started. I started with architecture, but I wasn't really the typical architect. I never really did architecture, per se. I moved out of Greece. Soon I was in London, and I started at UCL. I sort of like studied the intersection of computer science and architecture, and very quickly moved into programming. I worked for foster and partners, which is one of the largest architecture firms in the world. I was one of the founding members of the applied research and development team there, and I was teaching at UCL, and I was doing really cutting edge tech stuff for for really big projects, from Apple's headquarters to, I don't know, airports around the world. And I tried things like the first Oculus Kickstarter edition. And, you know, the first iPad, sort of like AR stuff. I built a lab there. And then, you know, I sometimes, at some point, I kind of, like, got sick of London, I think, and I moved to Barcelona. I did a PhD there, where I started, also teaching in the Institute of Advanced architecture of Catalonia, as you said. And worked with a lot of like advanced material, advanced things, sort of like advanced fabrication, let's say, and eventually ended up in Austria, where I took over research lab, and for five years I was leading that. Throughout these 15 years of work, I always worked on the intersection of data, computer science, programming, architecture, technology and climate resilience. So my goal has always been to bring whatever technology I need to bring to architects so that they can make buildings more sustainable, more sustainable, more sort of resilient, more climate friendly. And one of the key problems of that is essentially bringing the effect of, you know, a building on the climate in the hands of the decision maker, the architect. When an architect designs a building, when they're going to put, like a window here or there, or like, you know, somewhere else, they're actually making a huge decision on the life cycle of the building. You know, how much day later will be there, or what sort of, like, air flow patterns you will create, and they don't necessarily have this ability to understand the effect of their decisions. So simulations, climate simulations, or like, environmental design simulations, as we call them, they bring that information to the architects. Hands, right? And my goal has always been to bring these simulations closer to the architect. I've worked a lot like you know, my PhD, my masters, all of it was, how do we make them, you know, more accessible, more sort of like, you know, fast, more more more easy, easy to use. And at some point at the at the city Intelligence Lab, together with my co founder, theodos, and my other co founders, Juan. Nine and Christos, we we actually did a breakthrough. We had a model in our hands, and this was early days, AI, way before chatgpt and mid journey and the stuff, and we were able to do something that takes only a week for an architect to do, which is a typical wind simulation around the building done in a second. So we could literally, like, move a building from one place to the other and see the effect. And that was kind of like the groundbreaking thing. So that is what created Infrared. There was a lot of hype. There was, like, a lot of people wanted to use it. We tried to sell licenses as AIT, but eventually we figured out that we have to spin off Infrared as its own, sort of like company. And that's what created. I'm not a typical entrepreneur. I didn't decide in my early 40s, as you say, to sort of like, yeah, like, figure out a way to make, to make money. But it's most of like, came out as a, sort of, like a natural outcome of something really cool that we developed. And normally I call myself like an AI Boomer, like, I've been doing AI for quite some time. I'm like, I was teaching AI models, artificial neural networks built from scratch to architects 15 years ago. So for me, the AI boom right now is fantastic, but also a little bit sort of like a little bit like overhyped, but I love what's happening in the world, and has changed completely how we develop things. But yeah, my my background, my goal, my sort of like vision is more on becoming climate friendly. And other things that maybe I say sometimes when I'm talking about Infrared, is, we're not, you know, replacing a writing an email, which takes you like, you know, 20 seconds and very little power, you know, computing power into something that you know will will consume the water of a small village for a couple of hours to just write an email. We're actually taking something, we take a week to compute with 100% CPU of, you know, multi core machine, and doing it in a second, we are, I don't know, like, millions of times sort of more efficient than an actual simulation, and that's why I think it's also important to understand what you're using AI for. What is it really doing to you? I think with the power outage of Spain yesterday, that's a very good, very good point. Adil Saleh 7:45 Very interesting. And how did you like, I know that, like, how simulations work in terms of, like other industries, like health simulations, you know, familiar with, there's one sender, and they acquired a company named Cognito back in the years when I was, you know, they were one of our clients. And, you know, then I first, you know, back in the years, like it was seven, eight years back, and I first realized, like, there are some simulations that you can, you can, you know, you can make and really help people, you know, with mental readiness and coaching, and it goes into corporate as well. So, could you tell us our audience a little bit about like in, in the way that you know, that's more relatable to them, and in terms a lot of these folks, they were thinking a lot of lot of things when it comes to climate simulations, you know, could you explain a little bit more on that? Angelos Chronis 8:33 Absolutely, so, generally, physics simulations, because when we talk about the climate, is essentially like replicating the physical environment around the building, right? And you can do that in multiple ways. You can do that with Ray tracing, which is what rendering, for example, for you know, game engines is doing, and there's a, there's a lot of like, such examples of, like, you know, fast rendering engines today, or even zero physics involved, sort of like, you know, how can I get from from an sketch to an actual rendered image? That is, that is essentially tracing the light around surfaces in a environment that simulation is already being sort of like used a lot in architecture, for example, to measure the solar radiation that falls on a certain facade and how hot it becomes, or how much PV potential you have if you put like a PV on The top of of a building. This is a very straightforward simulation. It's essentially just tracing rays around a CFD computation of fluid dynamic simulations, which is my favorite one, is a fairly more complex one, which is essentially one that requires you to create a a sort of like domain, a virtual domain, like a virtual wind tunnel, where you have like, 1000s of millions of, like small cells that you have to calculate energy and, you know, and movement of fluids around, so that you can understand what happens to wind see if these simulations are like the most well known computationally intensive things. Whenever you get a new sort of like parallel computing algorithm. You're testing it against a CFD simulation. What is really astonishing about AI is that it's changing the way we and we, you know, we can actually approximate physics. Everything, every simulation is an approximation, right? Even when you send essentially a, sort of like a a. And the rocket to Mars, you're you're not exactly right. You're always simulating, but all of these are similar physics simulations. You actually need to be able to be accurate enough to make a decision of whether you're going to send the rocket to Mars or, you know, you're going to crash it somewhere else. So it's not exactly rocket science, but it's also rocket science. It is like, literally, rocket science, because this is what rockets are using, but it's like, less, let's say, let's less important to be as accurate as a rocket, to figure out how the wind will blow around a certain building. But that's exactly what we're doing, trying to figure out how the wind will move, how the solar radiation will affect the building, how the thermal comfort will be for the people around the building or inside the building. That's what we're trying to do, figuring out through physics what essentially becomes a climate, environment in and out of buildings. Adil Saleh 11:28 Oh, amazing, amazing thinking about, like, when you started, like, Infrared back in, like, not too long ago, it's about two years now, and how was the like, initial journey like, I know that you're not as big of a team, like less than 15 people, and I know that to build the technology. And you know, if you're smart enough, if you make the right decision on the top line as leadership. You don't need a lot of people. You know you today, need a lot less people, like people are trying to do more with less. So of course, like getting getting their heads around, getting people onto the table. What was that journey, building, building such, such platform, and how was, how was the technology challenge at that time? Like it was not too long ago, tell us a little bit about your founding funding journey. Angelos Chronis 12:12 Absolutely. So first, first of all, like, I think everything is people, right So, and it's lovely that we're talking today about AI and how AI can replace, you know, but like, at the same time, AI was made by people, and it will always be trained by people, and it will always be led by people, I find every discussion that removes the human from any loop, kind of like obsolete or irrelevant. You can make very elaborate agentic systems that can do crazy, difficult things, but you can never, at least until this point, there's no evidence of it replace the ingenuity, the creativity, the sort of like ability to do systemic thinking that the human can can do. So for me, it's always, you know, it's always people's sort of like relationships, like when I started with Theodore and then Christo San Johanna, you know, working the Infrared City concept. I think it's, it's, it's this sort of like excitement that we had between us, of like, solving something that was our issue for a very long time, that drove us. And I think the most difficult thing with a startup is, like, if you don't have, if you don't share a vision, if you don't actually have like excitement for what you're doing, if you can't really wake up in the morning and do something that, you know, really excites you. There is no way you can actually succeed. I think that's, you know, for us, it's always been the excitement and the sort of like, you know, the sort of like idea that we can actually address something that our field has been trying to solve for a long for a long time. And I think when we started hiring the first people, the first thing we were looking for was not necessarily, like, they have, like, excellent skills of like, you know, programming, or do they share that vision? Is it possible that we can wake up in the morning and, like, meet the team, and the team knows where we're going, understands what we're doing, like, shares that vision. And I think sounds cliche, and like a lot of quick people say things like that, I think that that that is, for me, the most important thing. The most valuable thing you have is definitely not your capital investment or your like, VC money, or your whatever like you know, what you think is essentially your, your your biggest asset. Your biggest asset is time you wake up in you wake up in the morning, you have half of your actual life to dedicate to work, right, to whatever you're going to do. And if that work doesn't give you satisfaction, there's very little chance that you're going to end up with something really, really cool. And I think that's what's driving us now. That's what's that is what was driving us year and half ago, two years ago, and I think that as long as I am in this position, and we're building in Infrared City together, that will be what, what will drive us in the future? And we see it also now. We're hiring four people right now, right? And we see it also in the interviews we we hear people telling us, we want to be part of your team because we share what, like we see, what you're saying and what you're sharing, you know, your vision, and we're sharing that vision. There's a lot of talk today about this Gen Z or whatever, like, you know, whether people want to work for for money or for sort of, like, other sort of, you know, material rewards people need to feel. That they're doing something, they're contributing to something that makes sense. And if you don't make sense, people don't come to you. Adil Saleh 15:38 Yes, you have to make sure that you actually, I know that it's not easy to, you know, get people bought into your vision as founders. Because, of course, a lot of these people, they are, of course, they're not there for like, they're not married too much like vision that you have, it's so hard to get people bought it unless you have, like, a co founder on equity, or maybe some founding engineer on equity, that is a different thing, but, but, of course, a lot of this is a longer term, like, it's not an overnight success, so you got to make sure you get it this translated to the team from day one and but at the same time you you can get people that are extremely passionate, even equally passionate about solving that problem, and, and, of course, building the culture, even that is like small, 1015, people, making culture well connected people teams, and that gives them everyday motivation, if that's not coming From within on days, you know, people don't get, you know, wake up motivated every single day. So there has to be, you know, a culture that can, you know, fulfill that, and people around you that you can fulfill that. And this makes it all the culture. And I can absolutely let the way like and this. I mean, when I see this, these teams, when, when people founders, come to my park, as an as he been saying, Hey, these are, like, 15, 20, 30, even, 70, 80, up to 100 people. But a lot of them are sticking for for more than two years. That means, like, you know, that they're doing something valuable, and people are trusting them. And this is one signal that, you know, I get. So, you know, it's so interesting to, you know, especially in this kind of platform that you're building, like, there's so much of research and development, there's so much of industry, like evolving with the industry. And then AI is another challenge to make sure you keep yourself updated with the new AI updates and, you know, as engineering as a platform. So how hard was it? You know, build a platform and call it an AI powered platform. I know it's not like everybody else, as you mentioned. Angelos Chronis 17:37 Yeah. So one of the things I didn't say before about, you know, hiring, like getting getting people sort of fulfilling the team. And by the way, we're like, less than 10 people now, we're going to get about 15 people in the next month or so. But it's not, sort of like, it's not the typical, sort of like software development company really, like, Okay, if you know how to do software, you know, it's very easy to take any idea and convert it today, especially where, like, 80% of the code is written by a copilot or some sort it's actually pretty straightforward. But when we're talking about coming up with new AI models for physics, this is not your typical, you know, AI sort of like software developer. It's not your typical engineer. We need someone that knows how buildings work, how physics around buildings work, and then how AI works for physics. So that's like a very niche and very difficult place to get, you know, people to commit. I think one of the things that has been always helpful, and it's actually pretty much like, you know, kind of like a cheat for us, is all of us. All the founders, have been active in academia, in in, sort of like teaching in in involved in, sort of like PhD programs and consortia. So the there's a lot of people that that kind of, like join our, let's say community through there. I think most, or if not all, of our LinkedIn followers are naturally coming from our spaces where we operate, and we will have to be part of the communities and give to the community as well. So we're not, you know, it's, I don't think we're the typical startup in that, in that sense, we've not tried to, sort of like, push, let's say, paid advertising, advertisement and stuff like that. We haven't done much of that, if any. There is a community we believe a lot in, and there's we all teaching, or we all sort of like part of programs, and we we try to involve that community at every step. So that means trying to find that person that could do physics, AI and architecture would probably come more from our community than, I don't know, sort of like a headhunter or, you know, a recruiter. And I'm talking all about hiring, because that's what I'm currently dealing with. But at the same time, you know, when we're talking about, like, you know, the technology leaps that we have to do, because, again, as you said, we're not the typical, you know, we're not one of these chat GPT rappers that I was talking about when we when we met, we actually develop our own AI models. We run hundreds of 1000s of simulations that we produce ourselves, and we train our own models for them, with them. And you know, to do that, we you need expertise. You need deep expertise and and I think that's what can be. Differentiate us from from a lot of other companies, especially in our field, but we have that expertise. We've been working on this field for a long time, and I think we have a little bit of a awareness of what's happening. Adil Saleh 20:42 Yeah, domain expertise actually makes you stand out. And this is huge number, more than 100,000 okay, how do you guys, like, take to, you know, build this kind of, like, this shoe jet, level of pain, the simulations? Angelos Chronis 20:57 Yeah, it takes. It's one of the biggest challenge is to actually compute a lot of simulations before you actually train a model. And we've, we have done different experiments. Sometimes you might have done like, you know, 1000s of simulations that you would not use, because you will figured out that, you know, the model that you were trying to train would not be trained with that without, with that data set. And then there's a lot of experiments. There's different models for different physics. There is combinatoriums. There's sort of like, you know, multi model solutions, also that we were trying to build. There is a lot of like, like natural language sort of as well involved when we're talking talking about, like giving, giving the recommendations to the user, or like consuming guidelines and being able to train and say a rag model to give, you know, advice to a user. There's a lot of there needs to be a lot of cutting edge knowledge, but also domain the domain expertise that you talked about is actually crucial. I don't think that without domain expertise you can do something about what I like, what we're doing. And you know, if anybody's out there thinking of like building something similar, I'd say, I'd say the biggest challenge is to combine the domain expertise and the AI cutting edge software knowledge. I teach AI to architects the past five years, like quite a quite a lot, and 15 years now, I've been trying, and I've been sort of like in the in the sort of like in different courses on data and AI, and at least in the past, last five years, you see a lot of sort of sort of, like examples from students, for example, that, let's say they want to jump into another domain and provide, like, essentially, a software solution for a domain that they don't know very well. And I always advise against that. Like, if you your biggest value, your biggest asset, is the stuff you know in your own domain and how it operates, which you can, you know, skyrocket, augment with, with AI knowledge. But that domain expertise would be the basic ground. Adil Saleh 23:01 How you can basically how you're, you can take help of your domain expertise to make this AI specialized for your use cases, for your industry, and how you're basically it, as you mentioned, it's, it's not possible without having a domain expertise, without living through that problem as as you've been, because a lot of these folks did look at your business page and say, Hey, these guys started only two years back, but they never know when you started planting the seed. And all the experience that you had, like teaching people, and, you know, you know, being, being a PhD student, teacher. You know this, there's a lot that that takes a lot of years, like, it's 14 years, I 15 years. So nobody like sees that. Angelos Chronis 23:40 Yeah, it's very nice that you're mentioning. It's like 2023. Is the year that Infrared City started. Right? The first model that we developed, the first AI model that predicted wind simulations was done in 2019 so now we're like in the 66 years ago. Six years ago, we developed our first CFD prediction model. From the first model to having a company, to making it usable, to figuring out like users want to use it, etc, is there's a lot of work, Adil Saleh 24:08 Like four years, yeah, four years to actually get it in the production. Angelos Chronis 24:13 And also there was another component, which is actually spinning it off from a research organization in Europe, which for a lot of people that hear this, understand what they what this means. It's not the easiest thing, and we're very grateful that we were actually able to do that. So I think it's a combination of things. It's not just sort of like, you know, but also, for anybody that's watching the AI space, if you look 2019 March or May, or whatever, we developed our first model, very little AI talk back then. Adil Saleh 24:45 There was barely chat GPT three at that time, I guess became late in 2020 or early 2021 chat gpt 3. Angelos Chronis 24:54 There was, there was early versions of LLM models, like after we developed. I think the first model, I don't remember exactly what, yeah, I think so. Adil Saleh 25:06 Amazing story. I love the way that you know, the journey has propelled covered here. Now that you know, you have, you know, trust of some of the some of the VCs, is that a private equity funding that you raise, or is that from a VC or angel, like, what was the mode of raising capital? Angelos Chronis 25:25 Yeah. So, as we said, we we're spin off, but we had a, a. Been of sort of like company founded in july 2023 and we raised a seed round. No, sorry, a pre seed round. I'm sorry, we're currently raising a seed round. Now. We raised the pre seed round in September 2023 which was 1,000,025k in euros, and that is four VCs and a business angel. The Business Angel is a 25k and we also got a few grants along the way, about 600k or so of grants from from a couple of different sort of like sources. And now we're raising again. We're in a seed round at this moment. Adil Saleh 26:08 Backed with all the growth metrics. You know, how does the growth sound to you two years down now. Angelos Chronis 26:13 Say again, I didn't, I didn't hear you. Adil Saleh 26:15 How does, like, how do you see the growth metrics year and year growth ever since, like, it's been two years now. Is that pleasant for for your previous investors? You guys, of course, you always want more. You always think of more. But considering your industry. Now, it's not like 10x or 20x multiple, but you know, how do you see the growth metrics? Angelos Chronis 26:36 So I'm going to go into two detailed numbers. But you know, we we started in 2023 we developed from scratch, again, the sort of like commercial version of the software, because we had an MVP that was working and everything. But, you know, we had to start again. And we actually just released our first, you know, commercial release of Infrared City in late November last year. So it's not, you know, first year and something we were, like, mainly developing, and we, you know, we already have some traction. We have some, you know, initial revenue, and some, some sort of, like, contracts. And it's not like, it's not terrible, like, it's not it's not too bad. It's actually pretty good, if you ask me. So looking at at users and, you know, and subscriptions, there's like, people, people showing interest in what we're doing, which I think is a positive sign. And there's a lot more developing, development coming. So. Adil Saleh 27:09 Okay, so there is so much on the roadmap. Angelos Chronis 27:31 Yeah, absolutely. So there's a lot more things that we're doing currently. And we were very early stage, for sure. Like, you know, we just had a pre seed round. And, you know, I think that the seed round would give us, essentially the put, put us in the in the map, like, quite heavy, Adil Saleh 27:49 better position to, you know, of course, fast track a lot of things because a lot of these technologies they I met the founder of Adobe and now founder of SaaS back in London, Jason Lemkin, and I asked him, Why do you think we should raise the funding? I'm also a B to B SAS company, by the way. So he said, if you're raising funds for to do marketing better, to you know, to do anything else better. You don't do it. You don't need it if you want to do if you want to raise funds to do product better, maybe collaborate with a co founder, a CTO that is fractional CTO that's that going to help you fast track that product and you do it right for the customers, you know, only then raise funds. So, I mean, in your industry, it's always going to help you, you know, fast track the development stream that you have that is sitting on the roadmap because a lot of this, because you're dealing with enterprises. But assume you know a lot of these, you know, customers that you see these are like, these are B to B, and they have like different use cases. So you gotta make sure you serve them all. You cannot make like a one stop shop sort of thing. So you have to make sure you work hands on with them to build features. A lot of the customers that are already paying you, they're, of course, they're still waiting on some of the features that you're building that are sitting in the pipeline. So how is the customer success journey look like for these enterprise segments? I know I look at your pricing, some of this is for SMBs or smaller customers. You know, if you, if you have something to share, like, What? What? What is your success criteria or process around measuring success between number one, making sure they retain even those are smaller enterprise. Number two, if those are small, what is, what is the process around expansion? Like, once they get bigger, you're up. You already know that, and you have the, you know, data and everything and visualization and all those data points, track traders, anything that people use for success, for expansion, what is? What is that? So could you walk us through your post sales? Little bit. Angelos Chronis 29:50 Sure. I mean, as I said, we're early stage, so it's like, you know, we're still discovering things, but, but for us, the important thing is, like, it's not just about, like, you know, someone subscribing to Infrared City and, like, buying a one year license or whatever it's, making sure they're using the product. Because we're also breaking into spaces that we were not able to break into I'll give you, I'll try to explain this a little bit. So we're building a back end that can serve anyone from a very small architecture office that does, like no one, new simulation, or has never even done simulations, to a large enterprise that does that bread and bread and butter. Every day and and that's sort of like also expanding into not just architects and planners, but also developers, real estate, municipalities, policy makers and other sort of like, let's say, professionals. And what we really believe is that we, if we, if we're developing a powerful physics simulations, sort of like analytics back end, we can actually serve all the different segments of the market, from small to large. And that's what we're trying to do. For sure, we have, like an Enterprise Client. We're trying to make sure that we we have a sort of like onboarding process with making sure that they're actually understanding how much they can use with the product, right? Also for a small, sort of, like, smaller, sort of like user, we were definitely getting in there to make sure that they can see what can be done. There's a lot of things that we're developing hearing our customers. So like, you know, automated reports, for example, which is a huge part of our sort of like offering, is like a smaller architecture office, for example, doesn't have the time to generate, create a report, read some essentially simulation results and create like, different reports with a click of a button. In inference, you can actually get a whole report done without you ever doing anything. There's a LLM model that, like, spits out all the information. So that's what, where, we're trying to, to address the different sort of like user needs, to make sure that you're, we're useful for us. The number one key metric is, how much are they using? How much, how many simulations are they running? How much are they using the product every, every week? Adil Saleh 32:04 Okay, so you're, you're all, yeah, every event, all of these events, I'm sure you're tracking, like, all these interactions. You're talking any anything drop, you haven't you have a team, or you have a system that can communicate and get it sorted. Because a lot of this gets lost between the onboarding to adoption the next chain, the biggest, you know, the the problem in these kind of platforms are making sure they're not only being onboarded the quickest time, the fastest, but also being adopted the platform, the shortest of possible, possible way, and that is, as you mentioned, is possible only by tracking all those interaction, how they're interacting with the platform, using the platform, number of reports that they have downloaded. There are some actions that you can crack, that you have within the system. Angelos Chronis 32:47 Yeah, we're actually tracking the user, weather dropping off. Then we actually ask them. We go on interviews and sort of like test, why are you dropping off? At some certain point, what can we fix there? And we're, we're following essentially the user to develop the components that the user like needs to make sure that they you know, the simulations are done in a useful way. Adil Saleh 33:16 Okay, perfect. And I was also curious. I know there's, there's a lot that I can learn, and we have only limited time. And of course, we'll catch back. I have some questions for my wife. She's like more in the process of automation space. Rpa, you guys are technical, so you know better about how RPA has been impacted by AI. So I just wanted to, you know her, to just jump a little bit on the AI side of things as well. I know that Microsoft, they have their own co pilot inside. They have, like, power Automate, you know, do all the process automation, like she works for, like enterprises, like dozen bots and, you know, these, like action driven, like repetitive processes that they've actually built bots within UiPath, blue prism and Microsoft. So I just wanted to, you know, seek your your guidance on how should she should go about her career, because I'm afraid that lot of this can go obsolete when it comes to, you know, building bots now the AI can do it, all of that. So how she as an individual, as a professional, should go about learning AI differently to be able to, you know, be as effective as as a career oriented woman. Angelos Chronis 34:22 It is definitely, like, you know, it is definitely an interesting time, right? Like, like, I'm also teaching, as I said, like, you know, seeing Infrared as well. There is a lot of a lot of, like, space moving, you know, there's a lot of jobs that will become more and more obsolete. And that's normal. Like, if we go back in time, when the internet came about, like, there was a lot of like, you know, jobs that were lost because, you know, you needed less sort of mail coming in. And there was a lot of jobs. And it's going to be like, a lot of like, it's going to be a lot more impact now. But I think you know, anybody who is intrigued by AI should at least jump on and become a user as quick as it's the same thing as the as the internet. I think it's a good example, because if you actually knew how to use the Internet back in 1995 you were, like, super valuable, right? And then I think that, you know, AI has this ability to actually empower people, rather than make them obsolete. And I think that eventually what you will, what you will end up being is a power user. Of AI, if not, you know, an AI developer, and I think that already today is going to be hugely, hugely relevant, you know, use AI against AI. I would say it's, you know, if you're, if you're worried about the future, get on and fight the fight with the right tools, and I think, Adil Saleh 35:51 Absolutely, have a defense system. Angelos Chronis 35:54 Yeah, exactly right. I mean, it's the whole, the whole, like every domain is affected, like anything from, you know, customer, let's say success customer, sort of like service to management, like management design with AI. I'm now doing AI with, you know, management with AI, because I, you know, I don't necessarily know all of this stuff from before. But also the most important thing is, like your your core education, like knowing, I don't know, psychology will always make you better than an AI, sort of like assistant that kind of like, has read about psychology but doesn't really know how to use it. Adil Saleh 36:39 Interesting, yeah, of course, I'll pass on this method. And of course she's gonna listen to this podcast when it's live. Anyway, for this sake, at least she will listen to one of my episodes. So one thing that I have some questions regarding, like, how you thinking about going Multi Product or multi industry, like multi geography? I know that you're more stronger in the and, and, and you, you guys are more comfortable in the Europe side. But what about North America, what about having multi licensed product, you know, and thinking of like, thinking long term as a chief executive, are you having such these conversations, you know, or these thoughts, because a lot of these, these platforms, technologies, not specifically in your industry, but other industries they are going about, you know, hey, this is something that we can Also do using AI, and this is pretty adjacent to what we are solving. So why don't we have, like, let's say Gong they have, like, their own specialization, more than 10 specialized agents that they have launched only last month. So everybody is trying to do everything that that that comes within the adjacent workflow of their customers. So how do you see this, as a chief executive? Angelos Chronis 37:51 Yeah. First of all, we're a global product. So, like, we are operating globally, like our models can conserve anywhere around the world, any project, and you have to do that when you're doing architecture, because an architect based in London does projects in Middle East or Asia. So it's always like, you know, a large architecture office, we always do, like, things like that. The other thing that's actually very, very interesting, and where AI can help, is regulatory framework. So, like, essentially, if you're talking about buildings, right, you have very different planning, you know, planning sort of like rules in Germany than in North America or in Middle East or wherever. And the ability to consume these regulations with AIS is fantastic, because previously you'd have to employ people in every in every corner like they know that sort of like the regulations, and they're able to convert them into rules and stuff like that. So being able to do that today is really a huge benefit, and that's why I can actually have, like, a, you know, a significant impact in a growing company. When we're talking about the future of Infrared City, we're only, we're not only talking about, like, expansion, like, you know, geographically global, becoming a global, sort of, like climate resilience index, but it's also about expanding into all the different, let's say, domains that relate to resilience in urban environments, whether you are a city trying to figure out where in the city you'll have to implement climate resilience measures so that you can essentially make The life in the city better. Or whether you're a developer trying to figure out where you're going to develop your next building and how that building will be, you know, resilient and sustainable. Or whether you're a user, an actual sort of like, you know, owner of a house or or someone who, or even a designer, like, you know, you're actually able to understand the impact of your design decision, or your essentially, your assets that you're going to buy or you're going to sell with relation to the climate. So that's what we're trying to do with interest, trying to touch the whole value chain of architecture, from development of land to to selling it. Adil Saleh 40:09 Interesting and you're, I mean, it's, it's important to have the visibility and clarity around your vision. And it may change as year year over year. You may pivot. But there has to be one vision that, you know, when I spoke to the to the C suite at Gong, when he first came in, rather than he said that when, when we were as small as six people, we had the same vision. Yeah. And now, almost like eight, nine years down, we still have the same vision. A lot of things that we have done differently as a product, or change changes or maybe, but the vision remains the same. And it seems like you are pretty much clear on on where, how you want to scale this technology, and what problem at scale you want to, want to solve. And then it may come up with around different use cases, other use cases that may change city to city, geography to geography. But engineering wise, you're pretty much sorted. So now last question, before I set you free, Angelus, what makes you excited going into I know that we are almost like bordered down in 2025 makes you excited that gets you closer to the north store. Like, whether it's like, let's say you might be thinking, hey, why not we just get some administration and talks in Riyadh, and we, like, we start doing in big city, as big as Riyadh, as big as cities in, like, maybe the bay UA, you talked about Middle East. So I'm more turning towards that. So what is that makes you excited. Angelos Chronis 41:32 Yeah, so that's actually a really good point, like, you know, and connected to what you said before about the vision. So, like, having a clear vision, whether you're going to do an AI model for physics of wind, or you're going to do an LLM that consumes a regulation and helps an architect, you know, make a decision, or you're going to do a visualization of data, sort of like component for your for your app. The thing that I care about is when I, when I think of an architect, of someone like, you know, that starts, I don't know, like designing that that city or that building, or, like, Neo, right, like making a decision about, like a huge mass, massive, sort of like building a construct in the middle of the desert, if they're able to understand even a tiny bit more than they do today, the impact that they will have on the climate, and they're able to change their mind and Change their decision and make it more climate friendly, more sustainable, more livable, more sort of like, more in sync with, you know, with what we actually need more than everything else, which is essentially climate adaptation. I am happy and like I believe that we've actually moved the needle in what I think is the most important thing, because we can make as much money as we want, climate crisis will keep coming back. You know, we can vote whoever we want. We can choose whatever we want to do with our EVs and our sort of like brands. But climate change will be there and will come back. So, you know, unless you don't believe in it, which then that case, I'd say you have to do some research. Eventually you will be affected. And you know, any thing that moves the needle will actually make this situation better. Adil Saleh 43:24 And make make you get closer to your North Star. Angelos Chronis 43:29 Yeah, in the North Star is a climate is a climate friendly built environment for us? Adil Saleh 43:34 Yes, absolutely. And yes, it was really, really nice, inspirational. Having a conversation with you, I've learned a lot, and of course, I of course, listen back once it's live in two weeks and see maybe what more I can learn. Because I cannot, like I cannot consume a person who's sitting right in front of me with 20 years of experience with this deep knowledge around AI and about infrastructure and climate and all of that. And, you know, I can probably say, hey, 40 minutes or 30 minutes, I've learned this. I've gotten this. What he was speaking, I've gotten this. But, you know, learning is is from cradle to grave, as you know. So I'm a young guy, so I will always ping you anything that if I want to learn and I'm feeling so special to you, know, having conversation with you and sitting right next to you. Angelos Chronis 44:20 It's lovely to be here, and I feel the same anyone who wants to reach out. I'm really always open to to talk about these things, and you, of course, included. Adil Saleh 44:30 Yeah. Thank you very much. Have a good rest of your day. Of the day, brother. Angelos Chronis 44:33 Have a good day.