Thanks, Ryan. I'm Ben Schmidt, the president and cofounder of RoboticX. So RoboticX is all about helping communities around the world to better understand their roads and their infrastructure. We started out as a a Carnegie Mellon University spin out. So about three and a half years ago, myself and a few of our cofounders, we got started on this idea of, you know, how can we use a simple smartphone to better understand the physical world? And we got started actually with pavement. So we started looking at things like cracks and potholes on road surfaces, and we've sort of expanded on that, you know, a lot over the last few years. And so today, we help about 200 plus governments, the majority here in The United States, but then also a bunch around the world. So we have some in The UK, Australia, a few other sort of smaller ones around the world, that we help to basically deliver for that government to deliver better roads and infrastructure to their citizens.

It's a great question. So I think, you know, this is my my second real start up. I had, like, a a sort of, like, a fake side gig one when I was back in school. Never really went anywhere. And now that I know what I know now, certainly was not ever going to work. So, you know, for me, a lot of it is is kind of that challenge. I think there's the, you know, why do I get out of bed each morning is, you know, being an entrepreneur, being in the start up and the tech world, you know, it moves fast. So there's a lot that you can both accomplish, in a short period of time. There's a lot of opportunity in terms of you know, there's both a lot of, like, capital. There's a lot of really talented people in that space. And the most interesting part is, like, there are so many interesting, useful challenges in the world that, you know, you can really make an impact on if if you sort of focus and dedicate yourself to. So I think that huge part for me is just I love the challenge. I love sort of the thrill. I love the people that I get to work with in this environment, and that's that's just that's my driver.

Sure. So I think in the, you know, when we're talking about, you know, robotics, hence the name is really around roads, we started over the last few months to expand into, a lot of other media or not media, a lot of other infrastructure types looking at, like, sidewalks, trails, curbs, things like that. You know, one of the biggest organizations here in The United States that we found is with ASCE, so the American Society of Civil Engineers. Strangely, this is not my background, but, you know, I'm a member there. They have some excellent sort of both advocacy, media feedback pieces. They're hugely, instrumental in a lot of this. So that that's one, certainly. And then the other one that I think anyone in the transportation world will certainly know about is the Transportation Research Board, which came out of it's it's part of the National Academies of Science, I think. And that's another huge organization, lots of really interesting and influential people in there. And then there's a bunch of just other really great podcasts and media and and pieces out there that, you know, are terrific in trying to understand more about the roads. Most fascinating part about infrastructure is it's one of those things that you'd never notice until you start paying attention to. And, really, the when you start paying attention to it is when it's not going well. Right? So, like, everyone notices a pothole, but you rarely ever notice the, like, brand new repaved road that you're like, wow. This is in really great shape. And it's one of those fascinating pieces. Right? Infrastructure is everywhere. Right? It's it's literally the foundation of our entire society. So I think there's a lot being done there. There's a lot of technological progress. So it's certainly an interesting space and I think getting more interesting over time as as new technologies sort of inundate the space with new innovations and things like that.

Sure. So while today, I think I'm the, you know, president of a company, I'm certainly more on the business focused side. You know? Deep down inside, I still consider myself a techie. It's kind of the ecosystem that I grew up in. It's what I went to school for. It's what were my interest, you know, a lot on the side. I think the startup bug caught me at some point, and so I started to learn more about, you know, business roles and functions and the more, like, entrepreneurial sides of it. But, yeah, certainly, I've had an interesting journey to say the least. You know? But I do think the theme underneath all of it is, you know, it's about trying to solve a problem. And I do think that technology you know, the the math that I learned in graduate school is no different than the math that we use to solve road problems today that I was trying to solve brain issues back then. Right? Like, the the math is sort of universal. The the technology and the techniques are universal. It's just how you choose to sort of apply it from one area to the next. So I I've been very fortunate, and I do think it is a it provides a really fun perspective, in that, you know, you you're looking at ten years ago, I was doing, you know, how are how are brain networks, interacting with each other. Then I started looking at how power markets are interacting with each other. And then today, we're looking at sort of, like, how do you better improve the world's infrastructure by applying really sophisticated technology to a problem that's been with society since the beginning? And I think that's that's really the kind of interesting part about it is is mixing both the business and the technical pieces to achieve really a solution to a real problem. And that's, I mean, that's what modern day, like, tech entrepreneurship is all about. It's super exciting. It's a lot of fun.

So I think, certainly, for me, one of the the strangest parts is my schooling was certainly in more the engineering sides, a lot of, like, bio and things like that, but a lot of math and programming underneath. So I I still consider myself like a programmer is really my my thing. One of the most fun parts of it is that, you know, business and programming have so many interesting parallels. So, like, how you write the architecture for a system as a programmer is very similar to how you should think about setting up a business, how the pieces communicate, same problem with, like, processes. So I think there's a lot of fun analogies in there in terms of the business side and the technical side, and that there's not really that much different. Just one has to do with computers and the other has to do with people. But I do think, you know, along my own journey of sort of sitting from I sit with a computer every day, and that's really what I interact with to, you know, my interactions are now entirely with people. It's people that drives all of it. It's not you know, technology is the fun, interesting, shiny part that's at the bottom of it, but it's all about people. It's the people that are making that technology, the people that are figuring out which technology we should make, the people that want the solution, the people that help us to sell that or to, find new people who want that, the marketing. So it's really you know, for me, it's been an interesting story arc from technology focused a 100% to now it's all about people, the processes, and, like, how can we unlock more of that technology by by focusing on people?

So governments still rely on on people for all of it. Right? So there isn't too much of a technical infrastructure for it in terms of, like you know, we talk a lot about trying to move towards a more data driven future. Now I think business has started that journey, you know, several, maybe decades ago, and is still very much on the journey. Right? Most businesses are still not run as a data driven organization. It's still intuition and gut and things like that. Governments are no different. They're just probably a little bit further behind for the most part. So even you look at things like the smart city solutions, and pieces like that, like, as we try to move into more data driven, process driven kinds of organizations, we have a lot of challenges there. I mean, the first is finding good data, is always difficult. You know, finding good data, making sure that it's it's data that will help you to make your decision. I mean, there's some parallels there with, like, machine learning kind of challenges. Right? You can have all the data in the world, but if it doesn't help you to predict anything, then it's useless. Same thing with with governments in particular is how do they start moving towards these data driven techno and technology in a sense that, like, technology can help make you data driven, but data driven is a policy choice. Right? That's a that's a people problem. Right? You make the the people decision to start making data driven decisions. You don't just sort of, like, snap your fingers, get a piece of technology, and all of a sudden you're data driven. Right? Like, if you ignore the technology or you don't believe in it, then you're not gonna make data driven decisions, and you're just gonna default back to the original scenario. So that's a critical piece. It's making sure that the people in that government, the people in that organization want to start making their decisions based upon good data, and then track those decisions and then sort of iterate on, did we make the right decision? Did we make the correct conclusion from the data we have to the decisions we have? And are we going in the right direction? So does that make sense?

Absolutely. So I think and, again, this is part of that, like, tech focus to sort of people focus. And I do think one of the biggest challenges is that trying to go in with sort of the hammer and convince someone that, like, data driven is correct. Here's a bunch of data. Like, here's a bunch of evidence that suggests why data driven is best. That'll work sometimes. But by and large, the the thing that will convince people to start using data is when their peers, and this is kind of the catch 22, but their peers or people around them have already made that choice and are liking the results they're seeing. So a lot of it has to do with, referrals convincing that, like, you can slowly move into this. Right? You don't just snap your fingers and all of a sudden kind of transform into it. And one big key for us is that, you know, there are lots of governments out there. There are lots of people that we would like to work with. The challenge is that they're all at different stages of that journey from kind of, you know, it it's a much more traditional model to this sort of, you know, newer models of data driven kinds of processes. So that and there's really no end to that. It's not like at one point, you're like, ta da. We are now a data driven organization. We're finished. Right? Like, it's still iterative. And I think the the key with what you kind of asked there is that you could convince someone at the earlier stage to adopt a new technology, But more than likely, yes, you'll get the success of working with them immediately. But if you haven't convinced them that the process needs to change, you know, it's gonna be DOA in in a short amount of time. Right? They will be just completely dead, and they'll use it for a bit, and then they'll sort of, like, taper off. That's not you know, that's both not our business model nor is it sort of our our mission statement. We don't want to just go out and give governments a tool, and then watch them sort of not use it or use it for a little bit and then say, like, yeah. Okay. We're done. Right? We want to be part of that journey with them in moving from one end of the spectrum to the other, and and make that continuous. Right? So we want to find partners and governments, and we have. When I say 200 plus, I mean, that's the really exciting part. Like, I think that's something that, as a society, we should be excited about is that there are this many governments who are already adopting this earlier technology and and progressing towards that data driven, environment. Because, again, like, you know, you live in a town, I live in a town. We all live in one of these municipalities. Like, I would love it if my municipality could tell me how effective are my tax dollars, how good is my infrastructure compared to someone else. Like, what are our challenges? Like, not in sort of the political rhetoric, but rather in, like, hard facts, like data that we've collected and that we've used. Again, it's not gonna happen overnight, but it's this trends that we're going to see. And I I think it's exciting, you know, as an entrepreneur to be sort of moving with that that journey of so many different governments around the world, as they move in that direction and and sort of you know, we don't really know what the the next steps will be necessarily, but we hope we can keep making or having an effect on them and and making progress. And that's really fun. I mean, that's the cool part about it. And and I I think, again, that's the reward for, you know, myself, for, you know, my colleagues in seeing what we're able to accomplish and sort of the the real world changes that are happening as a result of that. That's really exciting.

Absolutely. Absolutely. And I think there's a huge yeah. You think, like, agile software development where, like, you, you know, roll out an MVP or something and then iterate on it. Right? Like, that same process, I think you have that in that same context in even, like, a sales journey. Right? Like, you you can't sell someone the the ultimate solution. That's not going to work. Right? Like, that's too complicated. There's too many moving pieces. Even if you could accomplish it, it's just too great a chance that they're like, you know what? I don't wanna do this, and I'd eat. Right? Or they sort of, like, walk away from it. It's gotta be sort of that momentum inducing, the sort of, like, slow and steady, iterative, you know, get feedback, change it, move again, change it, move again. And we see that all the time. And and I think that's a theme that, you know, that agile software piece, is very much replicated across every area of the business. Right? Like, that's sales, no different in terms of, you know, how do we work with them? Sales account management. Right? How do we work with our existing clients to keep developing both that trust, but then the relationship and then moving it and progressing it forward. And, you know, ultimately, we want to deepen that relationship because that's what helps drive the success for our business. But by deepening that relationship, we're offering them greater and greater value.

If you think about what our our product does, right, we help the government to understand the physical world so that they can take that information and then go make better data driven, more informed decisions next. But frequently, when it ends up happening is, right, if you if you imagine all the roads in the town and you sort them from sort of best to worst, a lot of what happens by default sort of if you don't have our technology, is you get the squeaky wheel problem. Right? So you go in and you find all the people are calling the government and saying, like, I have potholes. I have problems. Those are the roads that get fixed first. Now the really fascinating part in there's not necessarily a one to one physical parallel in software in that roads physically degrade. Right? Like, you build a road today and you leave it outside, and thirty years later, it will be gone. Right? It will be damaged. You'll have problems on it. Software doesn't sort of, like, naturally decay, but I think we all know as software engineers that it does decay. Right? Like, as it sort of, like, builds and, like, there's issues. There's therein lies your technical debt. The fascinating part is that I I think there's a very interesting parallel in that the worst first strategy, the so called right you know, only work on the the things that are absolute critical, you know, on fire right now. Right? So only repair the roads that have potholes and such. Worst possible management strategy that you can have, right, for roads and I also think for software. And I think there's a great parallel in there. And that in the asset management world, that worst first strategy ends up taking all the roads that are that still have a lot of life on them, and you could do a little bit of repair right now, and that little bit of repair could save you several years of life on that road. That keeps it from sort of falling off the precipice into really bad. And the really bad are the things that are the most expensive. Same thing in software. I think the software engineering, you know, the pieces that are failing right now, those are the most expensive. You you end up deploying everybody to fix it. Whereas if you just did slightly better maintenance activities, all of your software systems, all of your road networks, their their lifespan increases. You get a little bit better maintenance and such. So there's a really yeah. Your your analogy is spot on. This sort of idea of preventative maintenance of deploying your resources to areas that are not yet in total disaster, but, you know, are not sort of brand new and and keeping your focus on maintaining those, ends up saving you a ton of money, not immediately, but over the life of your entire your your program, your asset, whatever it is. And that's really you know, when we talk about this this journey towards data driven, that's a a huge part of it. Right? Is that you're trying to move away from reactionary, sort of like go out and and fix the pothole. You you naturally need to fix the pothole, but you'll never sort of solve potholes if you don't start to move your resources further up that chain. And I think that's part of that data driven journey in Rhodes, and certainly the parallel is apparent for for tech debt and things like that.

management, pavement engineering, specifically. So there's a lot of, specific kinds of distresses that you'll see in pavement that are indicative, useful for sort of understanding where that pavement is in its its lifespan. So I think most people know, like, pothole. That is very bad. Right? There's a gaping hole in the road surface. Not good. Like, that's the worst you can have. But there's a whole bunch of signs that occur way before that. So, you can see, like, different types of cracking and there's lots of variations on that are also sort of indicative of, what might be happening underneath the surface, what might be happening in terms of a lot of the the binding agents and things inside that pavement surface. And so by visually observing these, what our our sort of technology is able to focus on, we can then take that what we're seeing, the types of distresses we're seeing, and correlate that with kind of a lifespan, that that one through five rating system. So one's good. Those are, you know, brand new roads to five. Those are where, you have maybe, like, base problems with type certain types of cracking, like an alligator crack, or you've got, like, potholes and things like that, which are really indicative of, like, this road is at the end of its life effectively. So that's that's basically so, you know, our path to achieving that, there are a lot of other existing rating systems. Here in The United States, we kinda have rating systems all over the place. There are some by the army corps, some by, different national organizations, some, cities use their own. And then because of the generally subjective nature of it, it it tends to be kind of like a rubric. So whatever program, you know, that town is using is more like a rubric where you, like, fill in, you know, a person sort of subjectively chooses what's kinda happening and things like that. They sum all that together and they can generate a rating. So we very much took the same concept of, like, a rubric, but then deployed tech to make it kind of fill in the rubric for us, so that we could generate another score. And the, again, the ratings kinda go all over the place. They have, like, one through tens, one through one hundreds, zero through one hundreds, one through fives like we do. And and so, you know, our choice was on going to a smaller scale to make it easier to digest. Right? One of the challenges that we found over and over again is that, that pavement engineer knows a lot about what we're talking about, and they'll know things like PCI and and sort of other rating systems that have, like, one to 100 scales. The challenge is that for most people that need to make the decision right now, and then for a lot of the stakeholders surrounding it, right, that extra complexity, doesn't necessarily, you know, result in a better decision. So we decided on a smaller one. Now we can certainly elongate and add some decimals in there and make it much more refined, or or more sort of focused in on on, you know, how do you compare this road to this road, you know, and they're both like three. It's like, well, we can actually get, like, three point one and three point two. So there's a lot of tricks in there that we can do. But by and large, you know, our value proposition, our mission statement is to make this something that every government on Earth has a system that tells them, here's the current status of your world, and then enables them to make really good informed decisions about that next. Complexity is the enemy of that goal. Right? So we want to make simplicity sort of the the name of the game. So, it it has been a challenge to sort of shrink these things down into digestible, you know, things that you can make decisions about. And I think that's the exciting part. We've achieved that for our customers is to make this easy enough that, you know, we don't wanna make it easy that it's it's not helpful. We want to make it easy that it's incredibly helpful, if that makes sense. There's a distinction between those two things.

Sure. Yeah. So the standard kind of machine learning pipeline is, right, machines learn best when they have lots and lots of examples over and over again of the same exact thing. And in in our particular case, we're kinda talking about, like, deep learning nets, which is kind of the the latest kind of creation of all these. What you have to do is, yeah, you gather lots and lots of examples of, in different scenarios, in different, you know, lighting conditions, in different angles. You know, machines are quite dumb when it comes to variations that, like, human beings are very good at. Right? Like, if I shake my head, humans don't really know that I'm the same person. That's the hard problem for computers at the moment. So we give it lots and lots of examples, use that to feed in, and then we sort of create that pipeline afterwards. You know, by and large, a lot of this is all about, how do you sort of fuse those two pipelines? And I I think this is very much to the theme we talked about at the beginning. Machine learning today is still a very people driven process. Right? It's it's I think it's kind of talked about in the media as this, like, you know, we snap our fingers and then all of a sudden machine is doing everything itself. Like, no. There's there's lots of people involved. Right? There's, data science and machine learning experts who who know how to train and tune and sort of coax a model and the math that goes behind it into achieving its best result. There's the engineers, software engineers involved who are all about, trying to deploy those kinds of models at scale, work with sort of data ingestion and sort of difficulties there. And then there are the, you know, people that are ultimately making and teaching, and then reteaching and updating the machine. Right? And there's there's a lot of people involved with that activity and sort of making sure that the answer that goes out and it goes to that customer that they attempt to make a decision with, is the best possible answer that's a combination of all of those people and all of that technology, that's served to one purpose, which is let's make the best answer that we can and then give that to the customers so that they can go on to make the best decision they can.

It's a funny anecdote. So, like, three years ago well, so when we started the company, it was in December. We had our our first sort of clients through the summer. By the time we hit fall of that year, right, naturally, what happens in fall, the leaves start to fall off the tree and get on the ground, but they also get in the road. Our model started to say, like, every road was terrible. We were like, what is going on here? I don't understand. How is this like like, all of a sudden, it's just stopped working. Like, I don't get it. We started looking at it more, and it turns out that all the leaves the machine thought were actually cracks. So it started saying, like, every all of these roads are awful. Like, there are leaves everywhere. There are cracks everywhere. The roads are terrible. And so what we had to do is basically go back, but, you know, there's a perfect example of overfitting. We were fitting to data that was built in the summer. A little bit before that, it did it had never seen leaves. We didn't have examples of it. And so, you know, we we had to then sort of rejigger the entire piece to say, like, no. No. No. No. No. This is good. This is bad. And I think that, you know, that's the sort of big one, but, like, there have been lots of those examples of, like, small iterations where that that edge case kinda comes out. Right? Same edge case you talk about in software. It's just probabilistic what happens with the the machine learning models and how they pop those out. So, like, you know, sometimes it'll happen in almost the same scenario, but for some reason, it's different. So, yeah, that that process is really the critical one. It's, you know, machine learning is not kind of a one and done, unless you sort of fully know the game. But for any real world application, that's never true. So it's this constant process of, you know, where's the next sort of challenge? Where's that next edge case? And you yeah. The neatest one, you know, people talk a lot about overfitting. You never actually know if you've overfit,

Oh, absolutely. Yeah. If you think of it more like a, kinda like a manufacturing line where you kind of, like, sample the product kind of piece, it's like that's that is a lot of what you need to do in AI in order to make it successful. You can't just sort of blindly trust that it it continues. By sampling, you can't catch everything, so, like, there's trade offs in there. But, yeah, that's that is really those processes, you know, the the sort of parallels of physical processes sort of sit, you know, align well with kinda how machine learning works is that it's by no means magic. Right? It is an incredible technology, but it is not like an automatic technology. It takes a lot of work, a lot of time, a lot of investment, and it's it's continuous. Right? It'll I imagine, you know, years from now, we will still be trying to refine kind of how we measure payment or payments rather just because that's the the nature of it. Right? And and very much, you know, it's the same thing with people. Right? You know, every once in a while, me who spent three and a half years now staring at pavement. Right? I can't drive anywhere anymore without looking at the road. You know, every once in a while, I'd be like, that's weird. Like, what is that? Right? It's the same thing with machine learning. You'll you'll still get that, like, that's weird sort of situations, and you just can't predict it, but you just gotta be aware of it and keep diligent.

Yeah. Absolutely. And it's certainly you know, given that we're our clients are governments who are even more sort of susceptible or more sensitive to these kinds of things. So all the data that we send out is all obfuscated. So we find people, not just kind of like faces, but, like, people and cars and blur the whole thing out. So we make sure that there's really no identifiable information in there. So, like, you can't find license plates. You can't find any of that kind of stuff. And that that was a very conscious decision much earlier in the company's history to make sure that everyone would be comfortable with the data, that it would be collected appropriately. And, yeah, I think very much the parallels with the Google Maps car, not that we are anywhere near the size of Google Maps car, so, like, the parallel is inappropriate in that sense. But, like, in The US at least, and and you know, another part of this is it's driven by our international presence to try and have one system for everything. In The US, you have no expectation of privacy on the road, But it was very much our sort of, like, corporate perspective that while we could do it, we there's no reason or no interest in doing it. Like, it's it's just it's better for everyone to do the obfuscation, make us more privacy conscious, and and try to do our best to sort of not be part of the problem in that way, I guess.

Yeah. And I I think the biggest part for us is we just try to be as responsive as we can. So I I think when as a tech company, as, you know, people in the technology market, we know things are not going to be perfect or going to work every time. Our goal is is not you know, we would like to be perfect, but our goal is not to be perfect. It's it's to be responsive. Right? When we see an issue or we're having challenges, respond back, make the changes, roll that out, deploy that, and get that out there as as best we can. So and that's really been our policy from day one. It's just yeah. Let's just try to keep doing our best. We're not always going to be successful, but we'll just keep trying.

I'm probably the wrong person to ask, and I I can't off the top of my head, I don't know who has, but I many governments have sort of, opened it up publicly. One of the things we even offer is you can take the color coded maps that we have with the rating systems, and they can actually embed that on their own government website. So a lot of governments have done that. You can even take the data we have and sort of extract that out and put it into other GIS systems. And so we we've also seen a lot of bad activity as well, so, like, opening it up to the public. You know? I think, again, in that if we're talking about sort of this journey towards data driven, a lot of people's first reaction and I don't think it's neither, you know, it's it's good or bad, but it's just part of the journey is people don't necessarily want to open up, and sort of be seen to see every single, issue that the the roads have. But I think the the customers of ours that have gone that path, they've really made that leap and are sort of open with the public about it. It's always well received. It helps solve a lot of issues. It it brings in and sort of, focuses people on what the problems are, what the challenges are, and how can they address them. And I think that's really exciting. So and I think those open data portals and a lot of the the data sharing that governments are sort of moving towards are incredibly powerful. I mean, we we have only scratched the surface of what's gonna be possible with, you know, lots of really interested bright minds looking at datasets like that. And, again, knowing, you know, our datasets are are spanning across countries, across continents, across states, you know, we have a lot of really interesting data, that's all collected at the same time in the same locations. So I think there's gonna be a lot more that we can do with that over time as well, that, you know, robotics can do, but also that, you know, we're hoping we can keep pressure on that movement to sort of get that data out there, and make it something that people can react to, roll into their their other ideas, look at different projects for, even, like, whole research dimensions with with open data access. You know, there's a lot of things that I think we can learn because, again, you know, we have a lot of local governments in the country. And by and large, they don't have to communicate with each other. So, like, I think there's a really neat way to sort of span across a couple of them and and identify, like, best practices or or new ways of doing things and things. So it's gonna be a lot of innovation from just the data itself and having it being open and and more accessible.

Sure. So, internally, we've done quite a number of it, especially looking at sort of comparisons. Again, a lot of that was originally driven by making sure that, you know, our systems and our ratings were we have developed a a global and, you know, the reason we did that was we had so many comparisons, a globally true rating systems. Like, a one should be a one anywhere in the world, and a five should be a five anywhere in the world, and not subject to sort of the, you know, whatever sort of, a local jurisdictions interest in the rating is. Right? Like, they can always change and and mold the ratings later, but, you know, we as a a global organization wanted to know how can we build an objective global rating system. So we've done quite a number of those comparisons and correlations. In terms of, you know, I think the sort of next step is going to be more on that predictive side. So looking at so, like, here's a really, you know, a a pattern that you might not see ever unless you're really paying attention to it. So streets in the Northern Hemisphere that generally go east west that have trees on the Southern side are usually in the shade most of the time. Right? So the the road itself is in the shade. The challenge with that is that if you get precipitation, the precipitation doesn't necessarily evaporate as quickly. So if you look at roads that have that where they're in shade a lot of the time versus if they're in the sun a lot of the times, you will see deep different degradation patterns. And you can see this even on a stretch of road that, you know, that whole road is there and there are trees and then not trees. You start to see these interesting patterns around those areas in the shade start to get cracks, degrade a lot faster than the areas that are in the sun. And you're like, okay. That makes sense. Right? Like, in a colder climate, you know, it might not evaporate, so the freeze thaws a little worse. Like, there's a lot of reasons for it. The challenge is is that, you know, that is an anecdote that we've learned over just a few years of watching. Robotics being three years old, but pavement lasting ten, twenty, thirty years, unfortunately, we have, you know, twenty seven years to go before we'll have seen a road fully new to thirty years later completely degraded. There's a lot we can sort of predict and start to get better at in the intervening time, but, like, you know, if you really start to think about this on a longer scale and you have constant, observations of the status of the physical world and you have those frequently in time, now you're starting to get enough of a time series that you can predict a lot of really interesting characteristics of it. Right? You could say, oh, yeah. This road will degrade at this rate because the climate, the shade patterns, the angle of the road, and the material used. Like, oh, incredible. Those are the kinds of breakthroughs that I think will come next when, if you have that solid foundation of you have a good understanding, that snapshot of the world as it is today, if you sort of marry that with a snapshot and then sort of continuing snapshots over time, you're gonna unlock huge potentials on what you can predict, you know, over the horizon now, things you can't necessarily see. And then you take that same concept, and you can apply it across geographies. You can apply it across latitude and longitudes. You know, you can apply it across continents, material types. There's a lot of really interesting characteristics that because that data you know, calling it siloed feels inappropriate because it's not like a database, but, like, this local government in Texas has an understanding, and then another local government in Georgia has another a different understanding. What's the chances the two datasets will ever sort of meet is is zero. But then with sort of this software focus and, like, moving into more data driven, database driven, you know, we can unlock that potential. We can start to find those matching things. So, I think it's a great question because there's there's just so much potential in there. And And I think we're starting to see the beginnings of it, but, certainly, you know, if you look at ten, twenty years from now, I I think I think the entire industry itself, I think how we manage assets, I think how governments operate this this area of service is going to be radically different than it is today. You know, I I I hope robotics is part of that journey. I think we'll continue to press forward on it, but it's certainly ourselves plus other companies are gonna keep that pressure and and really just shift us into a totally different a totally different future, a totally different model of government services, government, infrastructure management, and things like that. So that's fun. I mean, that's that's neat to see sort of the impact on the world.

I don't know. We've covered a lot of bases. I think my my general concluding thoughts are on my own personal journey from more, you know, 100% techie to now 5% techie, maybe 15, but five percent. You know, I I think for me, one of the biggest things that I I I wish everyone can find that realization is it is all about people. It it's both the greatest pro and the greatest con of of business, of technology, of of selling, of, you know, for robotics of the the people that we work with, our partners, our government clients. It all comes down at the end of the day to people people trying to, you know, do the best thing, make the best decisions, especially with hard times. And I think that's really you know, there's a lot of power in sort of understanding that, especially with a technical background. Right? I many years, I I tried to fight that notion. Like, no. I can build the best product, and that will be the thing that works. It's like, it's just not how it works. It's all about the people. It's all about the process. For better or for worse, that is how the world works. So, anyway, I think that is my concluding thought, and I I think that's one. I guess the my last concluding thought is, like, yeah, I think we are really at that that beginning steps of watching a very much government transformation, a public services transformation that I think is gonna be very exciting and I think also will open the floodgates to software and technology being able to help, enable, make more efficient, to better like, all of these really great concepts, within the government space. And I I hope a lot of people, see that same opportunity we do and and come join in this, gov tech world because, you know, you can just do so much to help, and, I think the world needs a lot more of that right now.