#389 AI and IoT for a Greener Future: Colm Nee’s Vision for Smart Buildings

[00:00:00]

Mehmet: Hello and welcome back to a new episode of the CTO Show with Mehmet. Today I'm very pleased joining me Colm Nee. Colm, thank you very much for being with me here on the show today. The way I love to do it usually is I keep it to my guests to introduce themselves, to tell us a little bit about their background, [00:01:00] their experience, and what you are currently up to.

Mehmet: So the floor is yours.

Colm: Great. Well, thanks for having me, Mehmet. Really appreciate it. Uh, so I'm, I'm Colm. Uh, I'm the CTO of Enlighted. Uh, we're an IoT and sustainability company, uh, headquartered in the Bay Area in California. Um, we're actually a Siemens company. Uh, so we're part of Siemens Smart Infrastructure Division.

Colm: Um, I'm passionate, so my background is, is really about, you know, using machine learning to solve, solve, uh, hard problems that exist in the world. Um, and so I, I try and use this also in my role at Enlighted. You know, we're very much focused on, on making the buildings that, where our system is installed as energy efficient as possible, and really giving.

Colm: Building occupants the power to to really understand, you know How the building is being used so that you can make operational improvements [00:02:00] So sustainability is everything for us. Um, and i'm very very excited to be on the show as well

Mehmet: fantastic same here also Colm because again, this is two topics that are close to my heart, you know, um, mainly like anything related to sustainability, of course and Uh iot ai and all this and Starting from there, you know, uh, we are seeing a lot of innovations in, in, in that domain.

Mehmet: So how are, you know, AI and iot innovations in, in European transforming the way, you know, let's start with the, the commercial buildings operate in terms of sustainability and occupant experience. And, you know, that would be great if you can shed some light on some examples of some successful, uh, implementations.

Colm: Yeah. Um, so I think it's difficult to talk about buildings and AI without first talking about the [00:03:00] challenges. Um, so commercial buildings do have challenges with data. And so I, I kind of like to use the snowflake analogy because buildings are in many ways like snowflakes. Um, and what I mean by that is, You know, each building is different.

Colm: So there's really no standardization of, of how buildings exchange data. There are some standards on interoperability, um, but it's not common, you know, for your average building to really see these implemented well. And, you know, sometimes there's even a lack of in, uh, alignment on incentives. So, Simple question, you know who pays to create a digital twin for the building so that various applications which are Uh working to operate the building can even speak a common language Um, the good the good news is that there's there's a lot of innovation going into this space There's a lot of startups in this space at the moment.

Colm: So my hope [00:04:00] is You know five years from now we're looking at this as really being a A fork in the road moment for buildings in ai. Um, and when I think about this problem Um, I really think about potentially using ai to really understand, you know to really make a building smart So a very simple example, you know, if you're If you're a lighting system or a HVAC system working on a building, how do you know what a conference room is?

Colm: How do you know what that conference room is used for? How do you know about the other assets in the building? All of this is relevant for solving data problems with buildings. And so I really think about AI as potentially helping us to create the digital twin. Um, I think another, another big use case is really on, uh, you know, pattern recognition.

Colm: So, you know, a lot of, a lot of things that happen in buildings can [00:05:00] be identified by machines, including problems. So using AI for, for smart maintenance and predictive maintenance identification is one that, that, that we see a lot of, um, you know, in terms of successful implementations. Enlighted has an app called the Connect app.

Colm: And, you know, originally back when Enlighted wasn't owned by Siemens, the initial version of this app, uh, was really designed to democratize building controls for the occupants. So one of the biggest problems that facility managers have is, you know, Is hot cold calls. So a hot cold call is basically what it sounds like someone will ring up and say hey This room is too hot.

Colm: This room is too cold. I'm having to wear a sweater. Can you please? Adjust the temperature and so I'm not going to assume you know much about how heating happens [00:06:00] in a building, but basically You know, buildings have set points for each zone where there's, there's a temperature control. Um, so there's a cooling set point and a heating set point.

Colm: And the cooling set point says, hey, if the temperature gets too hot, let's cool the building because it's not going to be comfortable. And by the same extension, if a building gets too cold, So if it goes below the heating set point, then we need to heat the building. And so the dead band is, is the difference between these two things, the heating, uh, set point and the cooling set point.

Colm: And so if you think about it from, from an energy perspective, uh, a wider dead band is actually better for energy consumption because you're being more conservative on when you choose to heat and cool the building. And so it's also better for the environment. But that comes with a trade off and that trade off is comfort.

Colm: So, you know, what we had done is we developed a product that basically allowed building [00:07:00] occupants to vote. Um, so essentially it democratized building comfort. And what that meant was, you know, If you, if you put an application in the hands of the building occupant and let them say, instead of calling up a facility manager, let them actually vote in the app on whether they're too hot or too cold, we could then use machine learning, uh, to decide not only when to, when to heat and call, but also to adjust the set points.

Colm: And actually what we found was pretty interesting. So, um, what we found was by giving people control, um, they could You could widen the dead band. So essentially, you could, um, increase the scope of, of acceptable temperatures for the building occupant because they felt empowered and because they had control.

Colm: And so, as a result, we were not only able to increase the perception of comfort in the building, but we're also able to save energy. for [00:08:00] the, for the building owner. And of course, when we're saving energy in a building, we're also talking about, um, it being better for the environment and, and having less, less carbon emissions,

Mehmet: you know, like this is really fascinating Colm.

Mehmet: And, you know, I'm, I'm thinking about it, uh, you know, from different perspective, and I think all this is possible because also of the advancements that happened with, uh, of course, AI working in the back end. And you know the iot being in in the front of you know in form of sensors and you know But the idea of digital twin and this is something that just came out to my mind Now is it something to simulate like different scenarios just for me to understand, you know, and maybe the audiences So is it like something to to to be able to simulate also different scenarios?

Mehmet: Uh, and even maybe we can take Kind of a extreme scenario and a, let's say, medium, medium scenario. And then, [00:09:00] you know, compare how things would looks like. Is it like, if I, did I understand it? Right?

Colm: Yeah. I mean, you can use a digital twin for this. It's, it's always best to do this in the design phase. So when you're designing a building, a lot of the decisions that are taken, you know, are fixed because once a building exists, it's very hard to change something.

Colm: And so, you know, a lot of effort goes in. And this is something actually that Siemens is, is pretty big into is, is really designing a building and a system so that it's very efficient and so that it's comfortable. And so, but that's a little bit different to the problem I'm describing. So the problem I'm describing is, and this is going to sound really basic.

Colm: So, you know, let's assume you have a lighting system that has sensors in it. And these sensors can detect when there's people in the room or there's people in a certain space. How can these sensors, which collect that data, share their information on [00:10:00] state, so on whether a room is occupied, with another system, and, as an example, the system that heats or cools that space?

Colm: So you would think this should be a pretty easy problem to solve. And unfortunately, due to just how the construction industry works, the alignment of incentives, the lack of interoperability standards, this is often very different from building to building. So it does happen. Um, but if you look at a, Portfolio of buildings, sometimes even owned and operated by the same organization, you will see different systems, different standards, and you won't see this kind of data layer concept in all of those buildings.

Colm: So I'm, I'm talking real basics here. We're not about, we're not talking about simulation. We're just, we're just talking about sharing state on what is happening inside the building. And of course it's difficult to talk about energy efficiency without talking about building occupants So [00:11:00] energy and energy efficiency of a building will be determined by how the building is occupied So it's you know, we're solving some pretty basic problems here.

Mehmet: I got you got you. Um, so talking about you know, um AI and again, there's a debate I think and you know There's a lot of thoughts about it between balancing AI growth and sustainability, because everything we're talking about, it, it touches sustainability, right? So the rise of AI, you know, driving an increased demand for, you know, compute power and of course, like data center space and all this, how do you think, you know, from a CTO standpoint of view, you know, we can balance the need of, This computational power with the sustainability goals.

Colm: Yeah, that's a really good question. Um, you know, first of all, I think we just need to acknowledge, you know, the point of the question [00:12:00] and allow that to soak for a little bit. Because as a CTO or as a technology leader, um, your decisions result directly in operational carbon emissions. And thus they also contribute to climate change.

Colm: So it's actually in many ways, not, not so different from the decisions we take in our private lives around how we get to work, how, and when we travel to far away places or, or how we consume goods, um, and as a CTO or technology leader, you know, you have this added burden. Um, and so, you know, I read recently that it was in the news that some big tech companies consume as much, uh, electricity as some fairly large countries.

Colm: So, uh, I just remember this one example that, you know, two of the big cloud providers consume more [00:13:00] electricity than Iceland, for example. But when you think about it, it's actually not that big tech is consuming electricity. It's organizations like the ones that we work in. And by extension, you know, the products of B2B and B2C technology companies.

Colm: And so, you know, I think the question is, it goes beyond what a CTO is. typically got in their scope, but the question is, you know, really, what do we do about this? And, and for me, you know, as society, it really starts with transparency. And so the good news is, you know, there are laws coming into force in, in many places that will force companies to report on their operational carbon emissions.

Colm: And so, you know, in case you're not aware, you know, these are broken down into scope one, scope two and scope three emissions. And so these are categories defined by the greenhouse, greenhouse gas protocol and [00:14:00] to help companies measure and manage their, their greenhouse gas emissions. And so, you know, scope one emissions are direct emissions from, from sources that are owned or, or controlled by the company.

Colm: And so this is stuff like. You know, emissions from combustion and boilers in in furnaces in vehicles that the company owns. Um, and then you have scope to emissions that are indirect emissions from from the generation of of electricity. And so these emissions typically occur at the facility where the energy is produced.

Colm: So, you know, by the grid. Uh, and not where it's consumed. And so this includes stuff like, you know, um, steam heating, cooling. Um, and so when you think about data centers and cloud compute more generally, um, you know, some certain amount of scope to emissions could come from. From data centers. So if you're, if you're a, an organization that is, is [00:15:00] renting your own data center space, you will need to report that energy consumption as part of your scope to emissions.

Colm: Um, and then scope three emissions is, is basically all other indirect emissions that occur in the value chain of the company. So. And it doesn't include the stuff in scope too. So this is stuff like, you know, business travel and employee commuting, uh, it's emissions from the production and transportation of goods that you've purchased.

Colm: Um, it includes, you know, use of sold products, uh, and end of life treatment of sold products. So it's, you know, it's a really comprehensive framework. And so now if you think about cloud compute, even if you're using, you know, one of the big cloud providers. Um, you will be forced as a company to account for this in your scope through emissions.

Colm: Um, and so again, this is going to be law. And by the way, it's a good thing. [00:16:00] Um, because you, you know, you can only improve what you can measure. A wise person once said, um, and then as a society, you know, we need to figure out, you know, who pays for these emissions. Does a company need to offset, uh, whatever they can't, uh, get to with, with the use of renewables?

Colm: Uh, you know, many companies have carbon neutrality goals, uh, and want to be effectively, uh, carbon net neutral by some of them 2030, by most 2050. And, you know, there are some pretty big questions in this, like, you know, how can we decarbonize the grid? But, you know, getting back to your question on the supply side or the demand side, I guess, in this case is where CTOs and tech leaders, um, will have more control.

Colm: And so let's think about that briefly. Um, you know, when it comes to sustainability, I really equate it using cloud compute and storage to something like flying on a commercial plane. You know, it's bad for the [00:17:00] environment, uh, but you have little control on, on how it's powered and it's probably going to exist even if you don't use it.

Colm: Um, so I try and take the similar approach to, to what I take when I'm deciding whether or not to fly anywhere, uh, which is, you know, only consume what you need. Um, and then I think a unique thing with, with computation is, is design the computational workloads to fit the problem. Um, you know, there are certain.

Colm: When you're thinking about ai certain scaling laws, and so if you don't have much data, you know Don't try and develop this this huge complex model. Don't overfit your models to the data basically Uh, and that means you won't be spending more on compute than you need Um, and then I think you you know, you need to track what you spend you need to exhibit cost control But also, you know by extension carbon footprint control So I think there's there's a certain amount leaders can do But [00:18:00] undoubtedly this is a massive problem for society and a very complex one.

Mehmet: Very, very enlightening, uh, call, but just, I'm thinking about like this whole, um, you know, concept that you mentioned, do you think all CTOs or all technology leaders are aware and they are ready for, for that, or, or like, it's, it's a work in progress?

Colm: Well, you know, I think when it comes to the reporting, probably not because You know, most realistically, for most of us, it's going to be about using a public cloud and about cloud compute, right?

Colm: And this is scope 3 emissions. Scope 3 emissions are notoriously hard to report on because what you actually need is You need one of the big cloud providers to tell you how the energy generation is happening, and then you need to report on this. So, I do anticipate, especially with Scope [00:19:00] 3 emissions and the laws around them, um, a little bit of slack to be cut by governments, and I'm sure there's some very intense lobbying happening in this direction, because, you know, it is complex.

Colm: But I think it will come, and as a result, you know, we will need to know about it as, as technology leaders, um, and of course we'll be asked, so if, you know, taking my organization as an example, if we did need to do this kind of reporting, I would fully expect to be asked, hey, Colm, what are you doing? You know, how much, how much, how many carbon emissions are we using by using cloud?

Colm: Um, you know, I think the easy thing to do really is, is just to question how your people are really using cloud. You know, what I see a lot of, and we've gotten a lot better on managing it, but a lot of the time a developer will, will spin up a resource. So like either a server or, you know, some [00:20:00] compute and then they'll forget about it.

Colm: Right, so we'll continue paying for it, uh, and it will continue being used, so it will continue drawing energy. And so what, you know, what we've tried to do is, you know, for selfish reasons, because it's about reducing cost, but, but it has a positive impact on the environment. You know, we try and now scale our operations so that even if someone does forget that they're not using a certain resource, uh, that it's not a problem for the company.

Colm: And we're not, you know, so we have like auto deletion policies, et cetera, on, on resources in our, in our research and development environments. So, you know, you can do small things as well, and, you know, we're not going to solve this problem as individuals, but, um, you know, every little helps.

Mehmet: Absolutely.

Mehmet: Like, so this is, I think, one of the challenges, Colm, when it comes to, uh, sustainability. Like, what do you think other challenges, um, you know, technology [00:21:00] leaders would be facing to implement, like, sustainable practices across the board? Like, not only necessarily to this cloud. You know, I think, but across the board.

Mehmet: So what, what do you think, you know, uh, they, they, they will be, you know, challenges will be facing,

Colm: you know, I think it depends a little bit, you know, what industry you're in, you know, if you're a SAS provider, you know, most of your footprint is going to come from cloud compute. I think if you, if you offer a physical product, so if you, if you've got hardware.

Colm: You have a different set of challenges because, you know, there's more than we took. We think a lot about climate change and about carbon emissions because of the change in weather. But there's there's a lot of issues around how we're consuming resources in our planet and what this is doing to the planet.

Colm: So even stuff like recycling of products at the end of life and helping your [00:22:00] customers recycle. Um, by accepting that material back and then reusing this when you, when you develop, uh, new generations of your product, um, there's a, there's just a whole world to be explored there. You know, it's referred to sometimes as the circular economy.

Colm: And so, especially in Europe, you know, there's, there's a lot of laws that will exist, uh, around, you know, re usage of materials. And, you know, we're lucky to be part of Siemens who, who is really forward thinking on this. Um, and so, you know, we recently announced at Siemens, you know, a policy on, on how we want to You know, develop products, uh, that then every business unit is, is going to adhere to.

Colm: So unfortunately with time, it feels like, you know, the world gets more complex and you have more things to think about. Um, but you know, when it comes to the environment, those are the, those are the big challenges. And I think. [00:23:00] You know, sometimes people also underestimate the societal aspect of sustainability.

Colm: So I, I spend a lot of time thinking about how to make the development in my team, uh, be more sustainable. So how do I, you know, have create an environment where the team gets up every day and thinks, Hey, I'm, I'm excited to get to work. I'm motivated. I know what I'm doing. I know why I'm doing it. And, you know, the company's working for me.

Colm: It's not working against me. And so, you know. I think if you, if you cast a broad enough net around sustainability, the there's something in there for everyone to think about.

Mehmet: Yeah, absolutely. And actually, you know, one of the questions I wanted to ask, you know, and this is maybe, uh, I gonna bring it before, you know, the, the, the, when I was planning to ask you, which is, you know, from, from leadership perspective. So, you know, how as a CTO, for example, uh, you can motivate, you know, your teams to.[00:24:00]

Mehmet: Focus on the sustainability and of course like because it's all about innovation also as well. So, um, What the strategies you found the most effective to do? So

Colm: You

Colm: know, so I I do think there's there's working sustainably And then there's working on sustainability and these are probably two different things, you know for me it's it's really about building products in a way that so, you know, let's think about working sustainably first because You can't really work on sustainability unless you're working sustainably as a team.

Colm: And for me, it's really about building products in a way that, that keeps the teams happy with their work and performing at a high level with, and taking pride in their work as well. And so, you know, I tried to build teams that, that are autonomous and empowered. Um, and then. By doing that, you know, they [00:25:00] tell me what to do not the other way around Um, so I spend a lot of time as a leader, you know focusing on on team composition You know, do we have the right mix of skills?

Colm: Do we have the right diversity in the team? Um, how's the morale of the team? You know, are we removing blockers? So there are Challenges that the team are having are we are we really working to remove these? We also like to understand, you know, what's working what's not working and so we always do Retrospectives and I try not to join these because I find that if I join you sometimes it influences what people talk about But I always like to read, you know, the reports from the retrospectives and see you know How can we how can we be better next time?

Colm: Um, and then when I think about you know working on sustainability Um, you know, I think it goes, it goes really to not extending beyond your, your means. So whether it's using, you know, compute that you don't need or [00:26:00] spending money that you don't have on, or on things that really aren't particularly useful.

Colm: Um, I think these things are all important. And then I think when it comes to, you know, if you, if you, if you're lucky enough to work on a product that has a real difference in, in sustainability, Um, then great. You know, keep doing that. And I think if you're not, and you know, maybe you're, you're at the point in your career that you're considering a change, I would encourage everyone to join the revolution.

Colm: Um, you know, it's still early, um, but there are really so many interesting areas now to sustainability. I mean, green tech is just booming. So whether it's, you know, decarbonization of the grid, whether it's, you know, Researching and productizing renewable energy alternatives, whether it's the energy, energy efficiency of the actual infrastructure, which is an area that we're involved in, or whether it's, you know, speaking about, [00:27:00] you know, carbon emissions, carbon accounting software is also has a really important role to play in sustainability.

Colm: And, you know, if saving the planet isn't a motivating problem, then I really don't know what is. So, uh, it's a really exciting time to be, to be working in sustainability.

Mehmet: Absolutely. Now, one thing, you know, I'm curious about it, Colm, which is, you know, for you, you did a transition from, you know, financial services before you joined Invited.

Mehmet: So what was the motivation, like, Can I say what you just mentioned at the end now? Is it like the the main motivation for you to change and how is the experience of changing the industry? You know for for a cto like how how easy how challenging is it?

Colm: Yeah, you know, so maybe let's [00:28:00] start with because I I used to work in financial services as you said um And, you know, I really think that financial services is tremendously important.

Colm: So, financial education is something that we simply don't get in schools, which is a real shame. I'm not, I'm not really sure why. Uh, but, you know, we tend to rely on our parents, uh, to give us our financial education. You who tend to set either a good or a bad example, you know, when it comes to managing money.

Colm: And so I had the opportunity to work for a company that was amongst other things. Advising pension funds on how to invest their money on behalf of those who might not be experts on this topic. And so, you know, I really enjoyed this time. The financial markets are very dynamic. They're very exciting. When you're working in them, they kind of feel alive.

Colm: You know, they're constantly changing. [00:29:00] And there are things that are true today that might not be true in the future. Tomorrow in financial markets. Um, but I think there are there are really two things that motivated my change one was I actually Wanted to get closer to the customer So, you know, I never really got to see the health care worker who was able to retire a couple of years earlier Because of the investments that our clients made on their behalf Um, and you know, I wanted to see a different industry Uh, I'm an explorer by nature.

Colm: I, I really truly believe that variety is the spice of life. I love to learn new things and, and what better way to do that than by changing industries. Um, and you know, at the time I was actually doing an MBA. Um, and I think one of the, you know, for me doing an MBA as, as more of an, having an engineering background was a really transformative experience because You know, I used to think about my development as [00:30:00] something like, you know, an engineer, an analytical thinker, I used to think, well, you know, I've created a certain expertise for myself and if I change industry, I'm going to be losing all of that.

Colm: And I think what the MBA taught me was that that's actually not true. Uh, and that I could do anything I put my mind to. So I think I use the MBA as, as a, as a way of changing industry as well. Um, And, you know, the reason I chose Enlighted and Siemens for this change was because I fell in love with the concept.

Colm: So, you know, you can invest in this, this smart IOT system, which not only, you know, saves you a tremendous amount of energy, but it can also tell you things about the building and what occupants are doing that are not otherwise possible. So, to be honest, I kind of fell in love with this idea, and I'm still in love with it today.

Colm: Um, you know, the, the challenges from [00:31:00] changing an industry, um, you just, you learn so much. So I think when you're working in a certain industry for a long time, you develop a pattern of thinking, um, that can become a bit fixed. So you just see things in a certain way. And so, you know, I basically had to relearn everything from scratch.

Colm: And, and so I, there were times where I felt, uh, you know, really out of my depth. But for me, it was an incredibly enriching experience to change industry. And I would, I would encourage everyone to do it. And I think there are ways of doing it that we don't necessarily change industry, but you can change roles.

Colm: So, you know, we have people that go from being engineers to product managers and they can make amazing product managers because they can challenge the engineering team. On one hand, they kind of have this mental map in their head of complexity, what's easily possible, what's less possible. And they can use that to their advantage.

Colm: So I think even within the industry, there [00:32:00] are moves you can make, uh, to really, really enrich your experience. Absolutely.

Mehmet: And I can, you know, uh, I can second that because I have like similar experience, you know, changing roles, going out of comfort zone, uh, you know, shifting from a role where I don't see, uh, you know, Anything outside of my cubicle, I would say to a role that I'm every day somewhere in someone else's office.

Mehmet: So, you know, like really these experiences, um, shape, shape, even, you know, our, our personalities and it opened and widened our, our vision and the way we want to do things. I can see Colm that, you know, the passion you have for solving a real problem, which is a, you know, Everyone's problem. And it's good.

Mehmet: Like recently we start to see, you know, the terms ESG and, you know, sustainability and clean tech and all these things, you know, on everyone's agenda today. And you, we need, I think more people like [00:33:00] yourself who are like motivated and passionate about solving these real problems. Now, besides, you know, what, what the, What you're working on and what you do at Enlighted.

Mehmet: So what like other, I would say, intersection of the technologies that are getting you the most excited? Um, is it like a, a, a, a intersection of all these technologies together? Which is kind of a cliche, I know. Or is it something like next generation? I don't know. Uh, Robots or bots. I don't know how we should call them that can work on on these smart iot platforms What are like the things that are exciting you and you are seeing them coming in the near future?

Colm: Yeah, really good question Um, you know, it's quite an exciting time to be alive Um with everything that's happening, you know, I think thinking about the intersection [00:34:00] between ai and iot Which is something we spend a lot of time thinking about You One thing I'm really excited about is, uh, environmental monitoring.

Colm: Um, and there are, there are many different, you know, use cases for environmental monitoring. One that we think a lot about at Enlighted is, is energy efficiency. So, you know, in a smart building, if you understand with, with high granularity, the environmental conditions. You can understand better where you might have issues with with energy efficiency or faulty equipment Um, and so that's that's definitely a big one Um, but then you know just going a little bit further if you look at how energy is generated to smart grids So AI and IoT when put together really contribute to the optimizing the energy distribution based on real time demand and supply data.

Colm: And so this reduces energy waste and [00:35:00] enhances the Integration of renewable energy sources, which is one of the biggest challenges with, with, with decarbonizing the grid, is the fact that, you know, a lot of renewables can't just be turned on, uh, like you can, can fire up a gas turbine. Um, and I think, you know, on energy efficiency, there's another big one around, you know, renewable energy optimization.

Colm: So IUT can really help you monitor environmental conditions. So that you can optimize the positioning of solar panels and wind turbines. And so you really maximizing energy production. Um, and you know, there are other use cases for, for environmental monitoring on like air and water quality precision agriculture is another big one.

Colm: So. You know, how can you really monitor in real time soil conditions, crop health and weather patterns, um, so that you can do, you know, precise irrigation and [00:36:00] precise fertilization. Um, and I think, you know, on the technology side, so the question is, okay, well, why is this easier now? Why is it, why is it changing?

Colm: I think it's really never been easier to implement AI and IoT in your product. So, you know, we tried this recently. You know, you can host a, uh, recurrent neural network. So an LSTM type model on a microcontroller, uh, with something like TensorFlow light, and this will help create a ton of real time monitoring use cases, which have, you know, ultra low latency, uh, fault tolerance, because you're doing that computation at the edge.

Colm: So close to the problem and, you know, which generate a low number of, uh, Of false positives and false negatives, you know, across a bunch of industries. So, you know, that's, that intersection is particularly exciting. And then, you know, with the advancement of battery life performance, you know, the form factors for, for sensors are [00:37:00] becoming smaller and smaller and cheaper and cheaper.

Colm: And so, you know, I think we're at a point now where, where IOT is ready to go mainstream. Um, and then I think, you know, the other big trend I'm excited about, it's not really an intersection, but it's really firmly in the AI domain, is I really think we're at the start of the era of AI agents. So, you know, autonomous compute engines that conform perform, you know, moderately complex tasks.

Colm: And I think it's clear that this will transform, uh, industries like every industry, uh, especially those that have a high complexity in software development. And so one of the big, you know, VCs. In the san francisco bay area, which is where I live I I was reading an article that they were they were saying that you know in the world that we're living software is actually Becoming labor.

Colm: So they had this equation which says, you know capsule buys [00:38:00] coffee it buys engineers and it buys gpus And out of this comes code that that takes the role of labor and you know Whether you believe it the timing for this is now or whether it's still off in the future Even if that's close to being true, I think it's equal parts You know exciting and terrifying.

Colm: Um, but if you do if you think about industrial processes You know, I spoke about the problems that we have in smart buildings with normalization of data. Um, you know, the problems around data that AI can help with are really specific to the industries, and sometimes they're specific to the companies who have the problems.

Colm: And so, I think the great promise of AI agents will be, it will allow us to solve these problems At an order of magnitude, uh, more speed and, and, and more ease, um, it will make these problems a lot easier to solve because right now, you know, you would [00:39:00] need to throw a team of five people at a really specific problem and a really specific industry.

Colm: Whereas if an AI agent can help you, you know, pre integrate, pre normalize all the data, and then even apply a model to get a certain outcome. I mean, that, that's, that's a really exciting, uh, uh, possibility.

Mehmet: Indeed, exciting times and I'm a big believer in, in, you know, the AI agents approach across all industries because, um, I read like a couple of papers, uh, since last year and you know, the year before actually the, by the time actually chat GPT appeared and all this, and to your point, like, it looks like, yeah, so, so we're going to have the agents working for us in form of, of, uh, software or code.

Mehmet: Yeah. Uh, which really excites me also as well, because, you know, if you, I think even about the simple processes or [00:40:00] tasks that if I think an agent can do to me, you know, and then I give just what I want, the end goal, and then it goes and, you know, uh, categorize, you know, the task that it has to do, and even it can actually utilize other agents, you know, to, to, Complete some tasks.

Mehmet: So this is really fascinating to me. People thinks that I'm with you, uh, Colm. I think it's, it's, it's now, it's not like, uh, something futurist, futuristic, because I've seen kind of proof of concepts for, for few things in the past few months, which is like really blown my mind across different industries, whether it's like in retail tech I've seen in, in construction, in different.

Mehmet: You know, uh domains which is really fascinating Um, so curb almost will come to an end finally, you know Maybe words you want to share with fellow ctos And where people can find more about [00:41:00] you and connect with you.

Colm: Yeah, you know final thoughts I mean, I don't have all the answers right I have my experiences, but you know, I found um, That having values is really important Uh, for your leadership.

Colm: So it's surprising sometimes how easy it is for my teams to tell when I'm passionate about something and maybe when I'm going through the motions. And so, you know, when reflecting on sustainability, um, You know, it needs to come from the heart. And so, you know, we can always borrow examples from our private life.

Colm: I recently had, was lucky enough to have a, my first child, uh, with my wife. I have a daughter who turns one, uh, next week. And I think this has really changed my outlook as well, because, you know, when you have a, I think anyone can care about sustainability and around leaving the planet in a good shape for the next generation, but when you actually have a child.

Colm: Uh, that's when things really come home to you because you start thinking about the life that they're going to have [00:42:00] Um, so that would be my advice for people is You know have those core set of values which which are true to you and which everyone should know you by Um and and try not to deviate from them Um, you know, people can connect with me on, on LinkedIn.

Colm: Um, I live in, in California and in the San Francisco Bay area. So I'd be happy to get to know, uh, some people that are part of the community, other CTOs, other technology leaders, or especially founders. And, and hear, you know, what you're working on and, um, very open to that.

Mehmet: Great. Thank you very much for, you know, sharing your experience and for being also, also not only, um, you know.

Mehmet: sharing the technology aspect, also the human aspect, which is, uh, I really appreciate it. And to your point, like the whole thing, what we try to do, at least I'm talking about myself also now is how we can, you know, use our experiences to [00:43:00] leave the planet better than the way we found it. And, you know, to your point, think about.

Mehmet: The future in a way that is livable, uh, our kids can survive it and make sure that they, they have, you know, an easy life better than the one maybe that we, we, we had. Uh, so I really appreciate this Colm and being so transparent on that also as well. Uh, and, you know, I gonna put the link it in, uh, Profile link in the show notes so they can connect with you.

Mehmet: And again, this is how I end my episodes. This is for the audience. I really appreciate your feedback on this episode. And if you just discovered this podcast by luck, thank you for passing by. I hope you enjoyed it. Please give us a thumb up and subscribe. And share it with your friends and colleagues And if you are one of the people who keep following me, thank you for doing so and sending me your suggestions and comments I read all of them keep them [00:44:00] coming and as I say always thank you for tuning in.

Mehmet: We'll meet again very soon Thank you. Bye