Loading summary
Transcript84 lines
- [00:00]
Lee House
You still have to understand that these things are all different. And so your, your solution and your approach at an industrial scale has to be, all right, I understand that this is different. So part of my software paradigm is that I've built in to my solution, this, this understanding of difference. Right.
- [00:19]
Darren Pulsford
So, yeah, that's very complex.
- [00:22]
Lee House
I don't know how to say it. Hyper connectorization.
- [00:26]
Darren Pulsford
Welcome to Embracing digital Transformation, where we investigate effective change leverag process and technology. This is Darren Pulsford, chief solution architect, author, and most importantly, your host. On Today's episode, Revitalizing IoT with special guest CEO of IoT 83, Lee House. Lee, welcome to the show.
- [00:53]
Lee House
Great to be here, Darren.
- [00:54]
Darren Pulsford
Hey, Lee. Before we drop into a passion of mine which is, you know, edge and diversity of edge, I just finished my dissertation on this about a year ago on specifically on edge security and the whole bunch of things around that. But before we get diving into there, everyone that listens to my show knows that I only have superheroes on the show and every superhero has a background story. So, Lee, what is your origin story? What's your background story?
- [01:27]
Lee House
Oh, my goodness, my background story. And throwing that into a superhero context makes it.
- [01:33]
Darren Pulsford
Well, yeah, I wouldn't have anything less on my show. Why would I?
- [01:36]
Lee House
Well, so thanks. So, you know, I guess, I guess Lee's superpower, right, if I have one, is to. Is to understand the context of business needs in the moment of current technology. So. And we've got all of this.
- [01:58]
Darren Pulsford
That's a pretty awesome superpower.
- [02:00]
Lee House
Well, you know, people, I grew up at IBM as a, as an engineer, and people would tease me and say, lee, you don't sound like an engineer. You sound like a product manager. So it was always this, you know, being on one side, what does the customer need? And then being on the other side, what are the available technologies? And so it's really been fun for me throughout my entire career to try to create that, that product market fit, you know. And it's also interesting because the sort of edge of the archetype, if you will, is always changing. So it's constantly moving forward. The needs of the customer are constantly moving left or right. And many times we find ourselves trying to use technology to solve a problem that is not a real problem. So, you know, that whole thing has been a lot of fun throughout my career.
- [02:51]
Darren Pulsford
It's like, I've got the technology, I'll find a problem that I can use it with. Right.
- [02:56]
Lee House
Well, I don't want to get too off topic, but at IBM, we built the first wireless lan and we were the first people not paying attention in meetings because we were on the Internet and we could not give it away. We tried to give it to the highway patrol, we tried to give it to fleet management, we tried to give it to warehousing. Nobody wanted it. Two years later, everybody had to have it. So there you are.
- [03:19]
Darren Pulsford
There you are. That's, that's crazy. Yeah, no, there's, there's lots of interesting technologies that never saw the light of day until like 10 years, 20 years later. And then everyone needs it now and you just go, well, yeah, but that's been around, right? Why didn't it succeed the first time?
- [03:39]
Lee House
We just weren't ready, I guess.
- [03:40]
Darren Pulsford
Yeah, I guess so. So, Lee, incredible background and I, I love, I love your positioning because we do need people that can help bring the technologies that are there today to solve real problems. So let's, let's drop into the real problem space that we're dealing with with IOT. I remember 10 years ago, IOT was going to revolutionize the world and change everything. And in Industrial Revolution 4.0 was going to just, you know, change economies and everything. And it kind of did, but it kind of fizzled a little bit.
- [04:23]
Lee House
Well, I think we had a long, a long gestation period and some of the early hype, you're right, we were a little bit ahead of ourselves. I mean, there was a lot going on that would cause us to be excited. I mean, there was, the cost of edge computing was dropping like a stone. There was all sorts of edge sensors that were available. The, the gosh, I guess at that same time frame, what I'd call the Raspberry PI revolution was, was happening as well. So you had this sort of convergence at the edge where there's cheap, you know, abundant sensors, low cost compute, and then this, this availability of edge processing that just sort of made everyone understand how quickly these solutions could be, could be brought together and how much that could be, could be done with them. So everybody saw and understood the opportunity. But then pulling all of that together became tricky. So it's one thing to build a prototype, it's another thing to really look at what does it take to build enterprise class solutions, Especially with the whole notion of IOT where it's all about scale, scale, scale.
- [05:37]
Darren Pulsford
Well, I'm glad you brought that up because the technology was way far more advanced than the processes to manage those could keep up.
- [05:46]
Lee House
Right.
- [05:46]
Darren Pulsford
I even look at my own home. I've got IoT all over, all over my whole house. I have 72 devices in my house that are on my WI fi, right? And there's no real good central management of all those devices. If I look on my phone, I've got like seven or eight apps that manage, you know, 72 devices. Why, why can't there just be one? Right. I can't even imagine what it would be like in the enterprise. I mean, my house is my house. It's not that, it's not that complicated, right?
- [06:19]
Lee House
Well, interestingly, there's a different dynamic going on in the enterprise and certainly to your point, it is extremely complicated in the enterprise and we can dig into that. But there are commercial reasons that you know, that the consumer is gonna, you know, is gonna be divided and you're gonna have Nokia, you're gonna have Samsung, you're gonna have all of these different, these different, you know, providers that have a vested interest in their brand. And so at the residential and consumer level that's hard to fight. And, and you know, candidly, we never went there with, we'll talk more about IoT 83 later, but we never went there. We always, we're focused more on the enterprise solutions where there is a really strong motivation to solve this diversity problem. Right.
- [07:11]
Darren Pulsford
Isn't the diversity problem similar though in the enterprise space in that all the big players in Edge and IoT devices want to lock you into their brand in your life?
- [07:25]
Lee House
That's a really good point. And so you've got this sort of these, these hierarchies of needs. So you know, the, the gateway provider, for example, the gateway provider wants to maintain their, their brand. You know, maybe it's, maybe it's Dell or maybe it's Intel. You know, they want, they want you to buy their product and of course they want to be able to differentiate at that layer and that's important. And same thing is true of like you know, sensors, right? You, you've got sensor solutions that can be attached to various assets and those brands want to be shown. But in the end the enterprise has to be able to pull together a solution. And we were talking earlier, about 10 years ago, IoT came on the stage and what was the delay? Well, the delay was part of this conflagration of all of these different vested interests and trying to harness this all into something that could actually be used. And so again I think it's the commercial need of these enterprise class solutions to be able to overcome the different, I don't know, brand awareness and hierarchies and biases in order to use standards to build solutions to overcome this.
- [08:47]
Darren Pulsford
You mentioned the S word standards, right. Are there standards that everyone's now following or are they still kind of doing their own proprietary protocols? Because I know that's been a problem in the past, right, that, hey, if I buy something from Siemens, it doesn't work with GE stuff because they're using different protocols on the back end. And, and it's almost like they do it to lock out competition from moving into their space.
- [09:13]
Lee House
Yeah.
- [09:14]
Darren Pulsford
Is that. Are there standards out there now that people are like, all right, we'll, you know, we're going to move. We're going to be the first ones to move into the standards and make it more accessible and easier to integrate across all these different devices And.
- [09:30]
Lee House
Yeah, yeah, well, you know, I think less. My view is that the sort of proprietary protocols are less of an issue than the brownfield protocols. So, you know, you've got, you've got like Modbus and, and, and, you know, opc, ua, and, you know, all of these, all of these sort of, you know, legacy protocols that exist and they're all, you know, in industry products are made to last 20 or 30 years. Right. So you've got this long tail of products that have to be integrated into these, you know, modern protocols. And so, yes, I feel like the modern protocols, the IoT protocols have really come a long way and people are embracing them, particularly mqtt, Right. TLS security. Mqtt, that's really become the sort of common ground, even though there's a lot more, you know, HTTPs, et cetera. But I think solving this Tower of Babel has become more better handled now at the edge. So we've got a lot of edge gateways that can take all of these various protocols and bring them in and normalize those protocols into protocols like MQTT over tls, for example, and bring this Tower of. Solve the Tower of Babel. But then you've still got the challenge of diversity from what is the data mean? Right, so you've got.
- [11:08]
Darren Pulsford
Yeah, yeah, so I like that you brought that up because being able to, being able to get everything to at least come into like MQTT or HTTPs, I got the gateways to help me to do that, but a simple thing. And this, this caused multibillion dollar failure on a Mars Rover. Temperature. What am I measuring it in? Fahrenheit? Kelvin Celsius. Right. Well, I mean, what, what, what, what is it in. And without knowing that and having context around that, this is a real problem.
- [11:49]
Lee House
That's right, that's right. That's right. So our first problem, you know, we're going back to what did it take to get iot, you know, out of the sort of the doldrums? First was solving this, this Tower of Babel. But then the next thing is getting to sort of design pattern understanding of what needs to be done in order to contextualize data. And so everybody started talking about digital twins. So the digital twin is a great sort of and very understandable way to talk about contextualizing data. So when you have data coming in from a sensor, it could just be a single data point. What does that sensor mean? What is the digital twin for that end device? So you're not thinking of a reading, a telemetry reading as much as you're thinking of this is the asset, what is this asset doing? What is this asset for? What does this data mean? And so building a digital twin is of course the digital model of whatever that asset is. And the beauty of the digital twin is that it can contain near real time data, it can contain metadata that provides context for what is that data. It can, it can contain your units, you know, what units is temperature measurement in.
- [13:09]
Darren Pulsford
So are you finding that most of. Well, for brownfield, that's different, right? Because digital twins did not exist. So I have to build those digital twins for all those brownfield devices that are, you know, maybe only three years old or four years old.
- [13:24]
Lee House
Right.
- [13:25]
Darren Pulsford
Are we seeing, with new devices coming out that the vendors are providing digital twins with the physical device as well? Are we seeing that?
- [13:35]
Lee House
Well, it depends. So, you know, this is back to hierarchy. So if you're, if you're providing a, I don't know, a relatively simple device that is used in a larger context in a, in an industrial system, probably not. You know, you're just providing the, the product itself, the asset and then the data feed associated with that because that asset is then going to be used in a lot of different contexts, right? So the foundational layer, you know, of a digital twin might be there, you know, what is this name, value pair, you know, temperature and, you know, what have you, and some level of context about what is the release level of the, of the product that is reporting this data, etc. But then the cool thing about digital twins is that particularly as we move from sort of thinking about the edge to thinking about the cloud, right, or a more complex industrial IoT solution, whether it's on the cloud or on the edge, you're starting to bring data from individual assets or sensors right into systems. So now you can build a digital twin for A system and start to apply more and more context to whatever it is that you're actually trying to. Trying to optimize.
- [14:51]
Darren Pulsford
Okay. So that's good to differentiate because I don't necessarily need to have a digital twin for every single IoT device in my system. Right, right. Sometimes I can, like. I don't need a digital twin for a thermostat.
- [15:06]
Lee House
Right, right, right. In this case, you're really more interested in a digital twin. I don't know whether. Whether you're. It's a digital twin for your H VAC system. Right. Or digital twin even for the entire building. Right. Because these, you know, you can, you can even build a digital twin and. Well, we do build digital twins in. In smart building that take into account, you know, what is the, what is the temperature, the ambient temperature outside today and what is the, the. The thermal leakage, you know, of the H vac relative to the thermal radiation from the sun? Right. So we take all that, you know, you can take all that into account, or it can be relatively simple. Right. Just. Just trying to optimize the, the servicing of the H vac system.
- [16:02]
Darren Pulsford
So that digital twin kind of gives me the context a lot better. Right. Between these individual IoT devices that are sitting out there, you know, a lot of them are some types of sensors. There are actuators out there as well, but a lot of them are sensors.
- [16:18]
Lee House
Right, right.
- [16:19]
Darren Pulsford
That are saying, here's my data, here's my data. Here's my data. And they're dumb. I hate to say we. We call them smart edge, but they're really dumb. Right.
- [16:29]
Lee House
They just, they're not smart until you do something with the data. Right?
- [16:32]
Darren Pulsford
Yeah, exactly. So having the context. And that digital twin gives me that context on, hey, these sensors are these two thermostats, which I've got. I've got six little sensors I have throughout my house that tell me the different temperatures in different rooms to try and balance out everything. It's really nice if they know where they are with relationship to each other. Right. That's context. That's that digital twin concept. Right? Is that.
- [17:03]
Lee House
Absolutely. And so there's sort of multiple dimensions of this in terms of creating value. So, you know, you've got the. We talked about, you know, is it just a data point, a single sensor, or is it part of a system? So what is the context in the system? And then, you know, what level of, you know, as you bring in more pieces of data, you have the opportunity to say, okay, well, what matters in this system, you know, what threshold crossing events have relevance, what analytical measurements have relevance? What might we apply AI or ML modeling to? So you continue to take the different pieces of these digital twins to add, moving really from sort of a monitoring, initial monitoring level to more of an analytical level for creating more reliability or optimization or even asset protection, right? You know, in your home, in your, in your example, temperature gets to a certain point in my God, that's on fire.
- [18:07]
Darren Pulsford
Yeah, yeah, yeah, yeah.
- [18:09]
Lee House
You know, so, so, you know, asset protection is part of this as well. Somebody buys a multimillion dollar generator, you know, you want to know what is, what is, where is it running off the rails to protect, protect my asset. But then taking it still further into, you know, we talked about the various levels of intelligence. You know, whether this is, you know, just, you know, basic rules and threshold crossings or algorithms that we write because we understand the system, or AIML models that we deploy because we understand, you know, certain components of the system, derived data that we can compute or large, large language models, what have you. We start to create this deeper and deeper understanding of this data that is coming in. But then on another level we're starting to add this piece of this system influences this piece of this system influences this piece of this system. So we're trying to build all of these things together into a more complicated total solution. And then once that's done, so you've got the, you know, the depth of data, depth of knowledge and analytics and then you've got the, you know, the system level view and then you've got the business level view, right? So now how do you tie all of these, this information and data into your business? Because in the end, what are you doing it for, right?
- [19:38]
Darren Pulsford
Yeah. You're doing it to make money, right? Yeah.
- [19:40]
Lee House
You want better service, you want incremental revenues, you want again, capital protection of devices that were purchased, you want smoother operation.
- [19:51]
Darren Pulsford
I got to back up just a second because this is very different than software development that we're used to, right? Because and I'm a software engineer, so when I develop something, I can develop it once and deploy it anywhere. But these types of systems are not like that, right? Because every single building. Let's talk about smart buildings. Just in general, I can't cookie cutter a smart building, right? I can't just say everything's going to be identical in every smart building unless everything's painted the same, everything has the same square footage and same rooms and all that. That's why these systems are so complicated, is because they're unique.
- [20:35]
Lee House
Every Single spin. And we go back to your original, your original premise of, you know, 10 years ago we were talking about IoT, industrial IoT and how it would solve so many things. Well, and it is, and it is, but this is exactly the problem is this uniqueness because, you know, in, on the one hand, everything is unique, but on the other hand, let me discuss this a little bit. There are very distinct design patterns that repeat for different, for different, you know, vertical types.
- [21:03]
Darren Pulsford
Types of vertical. Yeah, I know, I agree.
- [21:06]
Lee House
You know, back to our earlier premise, we've got all of these different suppliers of gateways and sensor devices and all, all of that. So there's an area of uniqueness that has to be overcome and that's overcome mainly through standards with security practices and protocols. Right. And then people leaning into this notion of digital twins. He makes it easier for you to produce a digital twin because you can better understand the context of the, of the device. Okay. So that, you know, tends to make it easier, but you still have to understand that these things are all different. And so your, your solution and your approach at an industrial scale has to be, all right, I understand that this is different. So part of my software paradigm is that I've built in to my solution, this, this understanding of difference. Right.
- [21:55]
Darren Pulsford
So yeah, yeah, that's very complex.
- [21:58]
Lee House
Connect. I don't know how to say it. Hyper connectorization becomes a really important thing. You know, N to M to M to, to, to O to P. You know, you're connecting all these different things in a secure way. So, you know, that's, that's, I think, you know, really, really interesting and important.
- [22:24]
Darren Pulsford
Well, and we haven't even talked, we haven't even talked about all this. Variability in systems is really difficult to manage because anytime I have variability, and I learned this when I was in MBA school, I'm a dimming. I'm a dimming zealot. Right.
- [22:39]
Lee House
Yeah, yeah.
- [22:40]
Darren Pulsford
His whole, his whole thing was to, was to decrease the amount of variability as much as you possibly can in the, in the process and do that up front if you possibly can. But these big industrial systems and, or smart cities or smart buildings by definition have high variability in some of the systems, whether it's weather, because we're dealing with the real world. Right, right. It's the physical world throws a lot of curveballs into things. Right. Then you add people into it and oh my goodness, now you've got, you know, mass chaos going on. So how you talked about driving variability out, but through standards, through digital twin, what are some other techniques I can use to drive out some of the variability or do I embrace the variability which complicates my software packages?
- [23:37]
Lee House
I think you clearly want to drive it out where you can. And we can come back to the design patterns a bit, try to leverage the design patterns, but I think you definitely have to embrace it. When you look at, when you look at IoT industrial IoT systems, you've got several areas of high churn, right? So there's high churn in turn in the, at the, at the connectivity, right? What, what sorts of. And the interesting thing about IoT IoT is it is, it is not just IoT, it's. It's. You're connecting to software systems, you're connecting to databases. You got all these different sources of data, right, that are important to this overall solution, right? So areas of churn at the, at the data connectivity, right? So what do you do? Embrace that. Say, okay, I'm gonna, I know I'm gonna have these columns stand, this set of standard protocols. So I need to be prepared to support that. I'm gonna, I know that I have to have security, you know, absolutely. In industrial solutions, so secure connectors everywhere. And I know there's going to be some sort of oddball things that I need to connect to as well. So you're having that capability to integrate, you know, custom edge, or rather custom connectivity solution. So area of Churn number one, area of Churn number two is in the custom business logic. So you're going to, you know, all of these solutions are going to have something unique about them because, you know, there's these are being sold by companies that want to have differentiation, right? So.
- [25:12]
Darren Pulsford
Yeah, yeah, yeah.
- [25:12]
Lee House
Something specific and unique about their value proposition that they have to be able to produce. So having the ability to add custom business logic and custom analytics. And so that can be as simple as threshold crossing and rules creating. The ability to create that, embrace that. Because it's going to be different, right? The ability to create and instantiate algorithms, right? You're just code, code written by people that understand how a system works. And then the ability to instantiate AI and ML models, right? Because that's an important part of it. And then the ability to pull in all of the manuals and user logs and different information that might be needed in LLM models. Right. Because that's really important in where we are at this stage with IoT as well. So second area of Churn, custom business logic. And then the third area of Churn is in the user interaction. So the custom dashboarding the, all of the visualizations, the role based access control. Because these things, these systems are multimillion dollar systems. They're not just being used by one type of user, they're being used by someone who's responsible for operating the building, someone who's responsible for fixing the broken valves, someone who's responsible for rolling the truck out to the customer site or someone who's responsible for managing the overall enterprise. So I think you really have to embrace.
- [26:40]
Darren Pulsford
Yeah. Something that hit my mind on that was in my own house. I want, I want our bach in my own house. Roll access based on the thermostat.
- [26:54]
Lee House
You don't want anybody else to change it?
- [26:56]
Darren Pulsford
Yeah, yeah, yeah. I want my wife and I the only ones that can change it. The others can look at it.
- [27:01]
Lee House
Letting your wife change it.
- [27:03]
Darren Pulsford
Well, yeah, well, yeah, absolutely.
- [27:04]
Lee House
Right.
- [27:05]
Darren Pulsford
But when my kids come and visit or my grandkids come, they're all messing with the thermostat. Like who's messing with the thermostat? We have this all set up so I, I get that Rolex role access based is really important.
- [27:21]
Lee House
Well, it's critical based access management when you're, when you're talking about industrial and so we talked earlier about the evolution of these systems and how they become more and more and more capable and powerful. You know, as you, as you move up this curve, you want to allow more and more levels of automation and autonomy. So you absolutely cannot do that unless you strictly control, you know, this role based access control. Only certain, you know, we have to, we have to have, you know, in these industrial IoT solutions you have to log any of the changes, you have to understand why anything was changed, who changed it, when did they change it. You know, all of this stuff is essential and then you need to be able to control who has the authentication, who has the authority to make these, to make these changes. But at the same time, don't make it so hard that this uniqueness of the solution becomes a burden.
- [28:28]
Darren Pulsford
Well, I love that last part because I think that's been the problem in the past. We've made it so difficult to do this that people said, well, I'm not going to do it. They buy IoT devices but they don't integrate them into full systems.
- [28:45]
Lee House
Right, right.
- [28:46]
Darren Pulsford
That's some of the frustration that I have with my home automation is I don't have a central control. I've got seven or eight different apps. I'm like, I'm even going to try and tie them all together because I'm not right Unless there's a single place that I can go and say when I'm on vacation, make sure all the lights are turned off, all the doors are locked, the AC is set appropriately or the heating or whatever, I want to just say I'm on vacation. And it do all this instead of me going to five different apps, which is what happens today. And I go, and my wife goes, are you sure you turned off the lights? And I'm like, no, I'm not sure I turned off the lights. I gotta go do that. So I, I see the, the purpose in this now for, for you guys. Lee, do you guys, with, with your solution stack, are you offering more of that centralized view of all of these, that systems view? Is that, is that your approach then?
- [29:42]
Lee House
Oh, yeah, yeah, absolutely. I mean, what we, what we have to be able to do as a, as a, a software provider to these industrial systems, large OEMs, right? We have to, we have to have sort of a, gosh, I have to offer solutions that allow them to be extremely sophisticated in some areas and sort of beginning to walk in other areas because the different parts of the business will have different criticality and they'll have different going to market. So you'll have some pieces of the business that are sort of on the asset performance management, asset life cycle, basic end of the spectrum. Some pieces of the business that have built these complex digital twins and are beginning to add more and more systems to the digital twin and adding AI and ML and LLM solutions. Right? So it's everything in between. Some systems are only concerned with the, with the, with the, you know, connection of the telemetry data. And some systems are tied to multiple other enterprise applications in order to make the most value. So it's, it's all of this, right? And, and you know, we talked earlier about, you know, the challenges of how do you make it flexible yet not, you know, so hard that you can't use it, or expensive that you can't afford to run it. You know, so this has been one of the big problems in this industry is, is, you know, first thing on, the first folks that showed up on the scene, we sort of call them the mega OEMs, right? You know, the IBM with their predicts. I was actually at, I'm sorry, GE with their prediction.
- [31:27]
Darren Pulsford
GE with predicts. Yeah, yeah, I actually helped with that project a little bit.
- [31:31]
Lee House
You know, Hitachi, Siemens. You know, these are great solutions if you have a lot of GE or Siemens or Hitachi.
- [31:39]
Darren Pulsford
Exactly, yeah, yeah.
- [31:41]
Lee House
But, but trying to use those to build Custom solutions was just horrible, right, because you're just, just not appropriate for that. You know, millions of lines of code that are specifically oriented towards certain, certain, you know, paradigm. And so the next thing that show up on the scene is the IoT as, as a service, right, from, and Azure and you really, you know, from a, from a, they, they look fabulous. And in terms of building a prototype of something, you can do it almost over the weekend, right, Build a build and you know that these things are scalable. So you can build something that is, can, you know, you can construct virtually anything using the components from these systems. But you have to realize that they themselves are components, they're not really platform, the components to build a platform with. So if you embrace that, you've just embraced a long term maintenance debt, technical debt, going down that path and not only that, you have to hire all of the specific skills, et cetera and so on, but you've got the extreme versatility, right? The next thing to show up on the scene is the low code solutions, right? So they're more or less canned solutions and they are, you know, focused on industrial solutions. But you add the, I talked about the areas of churn, right? You've got the connectivity, you've got the custom, you know, the custom logic and you got the dashboards. Well these, these solutions are very focused on that. But the large OEMs find that they don't have enough flexibility or versatility or scale to be able to leverage that. So you know, what we have tried to do with IoT 83 is we've tried to, you know, zero in on this sort of challenge, right? You've got, on the one hand it's really expensive, long time to market, high maintenance costs to leverage the sort of IoT as a service solutions. On the other hand, you've got the low code solutions that you just can't really build what you mean to build, right? We tried to find the sweet spot in the middle that gives you all the flexibility that you need for differentiation and value creation without all of the technical debt and cost and risk and time to market. It's been a challenge. We're on the sixth version of the thing, so.
- [33:55]
Darren Pulsford
Oh, I bet, yeah. No, no, this is a very complex, much more complex than any other type of software that is done because you're, you're bridging a gap of a very, oh, I'm not going to use the word archaic because that's not the right word. Very well established industries that don't see eye to eye on things Software. The software side and IT side and the OT side. They just don't, they don't even understand. They don't even use the same languages.
- [34:28]
Lee House
Well, you know, it's interesting. One of our. We, we joke around and it's really not that big of a joke. Our biggest competitor is the intern internal engineering teams. Cool. And they want to do it themselves. But you know what we, what we. Where we are successful in convincing the business leaders that they should leverage folks that do this all the time. Yeah. It just accelerates their innovations. They're not reinventing this things that you've already resolving this problem that we've been focused on for almost 10 years. They're. They're just leveraging it for more innovation. Leveraging their expertise for what they really know how to do and create.
- [35:12]
Darren Pulsford
Well, that. That's right. That's absolutely right.
- [35:15]
Lee House
Right.
- [35:16]
Darren Pulsford
Eli, if people want to find out more about you and IoT83, where do they, where do they go?
- [35:23]
Lee House
Well, IoT83.com we've got a comprehensive website. There's all sorts of information there. I also publish weekly articles on LinkedIn that are focused on different topics around enterprise grade IoT solutions and AI and ML. All of those things ultimately get published on the, on the website as well. But those are both good, good places to learn more about IoT83.
- [35:51]
Darren Pulsford
Hey Lee, this is, this has been great. I'm glad I don't get to have guys like you on the show very often. I need to, I need to do this more because this is a passion of mine. Right. This is an area that I think is extremely important. So thanks for coming on the show and sharing with my audience, but it.
- [36:09]
Lee House
Was really fun for me. Anytime. I'd be happy to do it again.
- [36:14]
Darren Pulsford
Sounds good. Thanks a lot, Lee.
- [36:15]
Lee House
Cheers, Darren. Thank you.
- [36:19]
Darren Pulsford
Thank you for listening to Embracing Digital Transformation today. If you enjoyed our podcast, give it five stars on your favorite podcasting site or YouTube channel. You can find out more information about Embracing Digital transformation@embracingdigital.org Until next time, go out and embrace the digital revolution.