Episode 84 – Silicon Valley IoT Innovations: Insights from IoT Tech Expo 2024

The 84th episode of IoT Podcast: Over the Air finds Ryan Prosser, CEO of Very, and Bill Brock, CEO of Peridio, at the IoT Tech Expo 2024. There, they connected with a range of industry experts and thought leaders to explore the world of IoT and technology solutions, and had discussions on generative AI technologies, designing cybersecurity, the innovative applications of AI in radiology, and beyond

Hear from figures including Jason Zhou, Ellen Boehm, and Rene Gubancok, as they share their experiences, challenges, and incredible advancements in connected technologies. This episode is a must-listen for anyone interested in the latest innovations in this rapidly-developing field.

Live from the Valley, at IoT Tech Expo 2024.

Connect with the Guests Here:

• Jason Zhou – Pryon
• Ellen Boehm – Keyfactor
• Ken Kao – Rad AI
• Spencer Huang – Edge Impulse
• Charles Yeomans – AtomBeam
• Virginia Puccio – Fuel AI
• Tomer Greenwald – Otterize
• Eric Tschetter – Imply 
• Joshua Demumbreum – Sesame Software
• Lee House – IoT83
• Rik Attrill – Kontron
• Brian Walsh – National Grid
• Rene Gubancok – HARDWARIO
• Deepal Mehta – InnoPhase IoT

About the Show

“IoT Podcast: Over the Air” is a podcast that features unfiltered conversations with executives about their IoT journeys — the mistakes they made, the lessons they learned, and what they wished they’d known when they started. In each episode of Over the Air, our expert guests make bold predictions, reveal unpopular opinions, and share stories you won’t hear anywhere else.

Buyers Guide – How to Choose the Right IoT Company: Very

Subscribe here:

YouTube: https://bit.ly/3tYPLQI 

Apple Podcasts: https://apple.co/3tZpQZ3 

Spotify: https://spoti.fi/3KTQ7Pr 

RSS: https://bit.ly/3E1YJ4j

Transcript

Hey everyone! Ryan Prosser and Bill Brock here at IoT Expo 2024. Bill, you ready to get started? I think we oughta. Alright, this is a serious tech show. 

Don’t do anything to embarrass yourself. Hey everyone, I’m here with Jason from Prion. Jason, tell me about your company.

So, Prion is all about enabling enterprises to leverage generative AI technologies safely and securely. And we do it through something called RAG, which is Retrieval Augmented Generative. The idea is we want the large language model, the generative LLM, to do what it’s really good at, which is be conversational and creative, but we want it grounded in truth through a retrieval engine that finds accurate answers to a user’s question and then gives those answers to the generative engine so that it can provide that conversational contextual response that’s trusted and secure.

Okay, dumb question, but isn’t ChatGPT, isn’t it a retrieval engine? So ChatGPT does have some retrieval capabilities, but it retrieves from the broader internet. So in the business setting, you probably don’t want a general answer from the internet. You want an answer that comes from your SharePoint, your Google Drive, your trusted repositories, right? So that’s what we do. 

We make sure that these models are grounded in the truth of your organization. Got it. And so we’re here at IoT Expo. 

What brings you guys here? Yeah, it’s sort of two things. Number one is a little bit of evangelism, right? So everybody knows about generative AI, but we’re finding not everybody knows that there’s safe, secure ways that you can leverage generative AI today for your organization. So we want to share a little bit about what that technology capability is today.

The other piece is just to get a sense of what’s going on around us, right? We don’t want to operate in a bubble all the time. So seeing a lot of the other organizations that are out there, a lot of the cool innovations that they’re bringing to the table and understanding where we might fit, essentially partner and expand and grow together. Cool. 

Last question. Give me an example of like a deceptively hard problem that you guys have had to solve to make this possible. There’s so many involved. 

The idea is, you know, when you look at retrieval augmented generative at face value, it seems really simple. It’s, well, I need a retrieval engine, which is basically like search, and I need to orchestrate it with a generative engine. When you get down to the weeds, it’s actually a lot more complicated.

For example, when you look at a document as a human, you’re actually getting a lot more information off the page than just the words you’re reading, right? So a lot of naive retrieval engines will just extract the text off the page. But what we found is, if you actually get the relational information on the page, meaning if I’m looking at a header, and that header is above a body paragraph, and we can interpret, hey, that body paragraph is probably related to that header topic, you get a whole lot more rich information off the page and really a whole lot more accuracy when you start to ask questions against that content. Amazing. 

Jason, thank you for being on the show. Thank you for having me. All right. 

Well, here we are today with Ellen Boehm with Key Factor, and welcome to the show. Tell me a little bit about what you do. Yeah, so I lead IoT strategy at a company called Key Factor.

My background has been in product development and engineering for smart cities and smart homes, and now I’ve kind of moved into the cybersecurity space. And so I am all about understanding how to design in cybersecurity from the start when you’re making a new product. How’s that transition been for you moving into the cybersecurity space? Has there been a lot of learning, drinking from the fire hose, or did you already have a good understanding? IoT security is the most important thing to start.

Yeah, so I had some experience in the past when we were working on a connected street lighting system. So over a decade ago, we were putting in certificates and keys and flashing them in with the firmware. So I was familiar with that as a way to uniquely identify and address devices.

But definitely as time has evolved now, I’m seeing more and more adoption of that sort of security at the edge in a variety of industries, whether it’s smart home or automotive or energy meters. And so I think best practices around cybersecurity, it’s still emerging, and I think we in the cyber community are trying to help promote best practices and what to do. But it’s definitely not completely solved, and it’s definitely a partnership that we have to do, I think, with designers and developers and cyber vendors and product people.

Absolutely. What would you say is your bread and butter vertical? You came from smart cities, is that primary focus? Are you expanding further out? Yeah, so we’ve seen pretty well-established knowledge around cyber within medical devices, and I think that a lot like in the U.S. has been driven by the FDA saying, you need to have secure embedded pacemakers and secure communications so that you can’t change the operation of a medical device and harm a person. So I think from a medical standpoint, there’s been embedded security for a while, automotive, so like EV charging, autonomous vehicles, we’re seeing the need to have, again, security built into those, the same thing.

You don’t want to have a person on the road be harmed because a car starts driving away, shouldn’t because somebody hacked into it and changed the way they operated or pushed code that’s malware code. So any and all of these things can be protected with a level of authentication and authorization to know who’s allowed to change any of those things. So I think medical, automotive are kind of more established, but I’m seeing adoption of using keys and certificates and that sort of signing and verification across much more many more industries like smart home, industrial, energy.

IOT Expo, what brings you here? What are you excited about? So I’m excited about seeing what’s happening in the industry in terms of what are people developing, what new applications, what new technologies are being used. And from my perspective, again, we’re a very focused cyber security company that is just a piece of the puzzle when it comes to designing IoT. So I talk about thinking about designing security in from the start and not waiting until your product is ready to be launched.

And then, oh, by the way, you need to get some way of getting security in there. We should think of it holistically just like we do when we think of electrical, mechanical, software design. Security is also a big piece of that to start from the beginning.

Great. Well, thank you for joining and we look to have you back on soon. All right. 

Thanks for having me. Great chatting. Hey, everyone, I’m here at IOT Tech Expo 2024, and I’ve got Ken Cowell from RadAI.

Tell us about RadAI. At RadAI, we apply GenAI for radiology. What we do is we can actually save radiologists time around dictating the reports up to 50 percent.

We can automatically generate the report based on a few bullet points that they dictate. The second thing we do is we actually improve follow-up management. What we see is we can improve follow-up management by over 2x.

And a lot of these follow-ups from radiology exams are actually people who should come back but don’t come back. So by using our tool, we actually see that we have hundreds of people who end up coming back who got first-time cancer diagnosis from our tool. What that means is had they not used our follow-up management tool, they would not have realized they got cancer.

And that is something that we believe we’re very passionate about, improving patient care. Okay. So this is about improving the quality of the diagnosis.

Is that the idea? There’s two things. One is improving the quality of diagnosis. The second one is improving patient care by increasing their follow-up rate for people who should come back.

How did you guys… So I want to understand what brings you guys to this show. But first, the name RadAI. I was researching you guys before you were on the show.

I thought it was like radical, like surfer, whatever. No, this is radiology. Yes.

How did you land on radiology as a market vector? I love how specific you guys are. One of our co-founders, Jeff Chang, is a radiologist himself. He also has a bit of a background in machine learning.

And so combining his expertise in radiology and machine learning, RadAI’s idea was born. Cool. And is there a product name to the product? Or is it just RadAI? Yeah. 

So we have RadAI reporting. And the other product on follow management is called Continuity. So those are our two main products.

I love it. Okay, cool. So now to the second question, what brings you guys here? Yeah. 

One of the things we’ve noticed is we’ve built a fantastic brand in the radiology community. What we haven’t done as much is get in touch with the AI community. You know, AI is taking over the world.

AI can be done, can be used for good. We truly believe we are a positive force to the world using AI. And we want to be able to share our story to others, both for recruiting and to also encourage others to use AI for more mission-driven use cases as well.

Interesting. So this is about being here is about, because I would think you guys would dominate the radiologist conferences and so forth, because those are your end users. But here you’re trying to kind of raise awareness about what you’re doing in AI.

Yeah, exactly. We do go to radiology conferences. That has been our bread and butter the last couple of years.

Starting this year, we started going to more AI conferences and tech conferences because we want to raise awareness. We want more people to spend time building tech and AI for good and not just optimizing on ads. I came from Meta.

Optimizing on ads, for example. Ooh, dig on Meta. I love it. 

Okay, cool. Last question. I believe I can understand how you guys are.

I mean, this seems like a great use of AI. What are some of the difficult technical problems you guys have had to solve to bring this product to market? There’s a few things that are a bit tricky. Number one is the handling of PHI, patient health information.

So making sure we de-identify and anonymize it in a way that’s secure, that’s trusted and vetted before we train. This actually gets into a bit of a complicated machine learning training problem because the more you de-identify and remove data, the less the training quality is going to be. So you have to find the right balance of de-identifying the right set of things but still maintain the high quality of training.

The second part that’s really important is the bar for accuracy is much higher in the medical context. If GBP gives you a wrong answer, you laugh at it and then you say, ha, that’s funny. But in the medical context, it has to be very accurate.

And so we need to make sure that our model is much more accurate than what you get in a consumer-based product. And so that requires a lot of fine tuning, a lot of using different types of adapters, and also a lot of human feedback built into our product so that we can capture how radiologists respond to the product output. So you mentioned Meta.

One of the things that I’ve heard in tech for a long time now, 15 years at least, is to displace an incumbent. I think like Uber displacing taxis, for example. To displace an incumbent, you’ve got to be 10 times better than what the incumbent solution is.

AI in a radiology environment, you know, you’re kind of replacing human eyeballs. How close are you to that 10 times better moment, do you think? So first I want to share, we don’t think of ourselves as a replacement for radiologists. We see ourselves as a way for radiologists to be more efficient.

And so what we’re really going up against is industry incumbents that own radiology reporting software. So because we have our AI-powered software that can automatically dictates, sorry, not dictate, automatically generate the report, we can save radiologists up to 50% of their time. So it’s not quite the 10x you’re seeing, but given how overworked radiologists are in the U.S., being able to save each of them up to half of their time means they can process more cases.

It means more patients can be processed and more patients can get better care. And so we believe this is a huge advantage we have over the incumbents that don’t have AI in their reporting solutions. All right, Ken, good luck.

Thanks for being on the show. Thank you. All right.

So here we are with Spencer from Edge Impulse. Welcome to the show. Thanks for joining us.

Tell us a little bit about what you do. All right, well, thank you for having me. We are a software platform that makes it super easy for our customers to build, optimize and deploy machine learning models for edge hardware.

That’s a lot. So I’m going to just take a pause because your listeners are probably saying, what the heck is all that? Yeah, yeah. So what the heck is all that? So you guys are at the intersection of like nearly every buzzword in the space.

Yeah, I know. I know. But you’re doing it all, and you’re like real and in production with a lot of customers.

So what are some actual use cases? Yeah, so let me tell you, I took a step back because what’s causing this to happen. Today, there was this big push for the cloud. Everyone was like trying to pump all this data to the cloud.

And now there’s a new trend happening, right? All this data, especially, you know, in factory settings, manufacturing, or just like I’m wearing a wearable, right? You can’t always be connected to the cloud. Where’s the data being generated, Bill? Let me ask you. On the device.

On the device and on the edge. And so that’s what’s driving a lot of this. It’s the cost.

It’s the customers need real time. They can’t have that latency. And so for us, we’re like, okay, how do we make it easier for our customers? How do we help them build models that are optimized for smaller footprint? Because it’s not the cloud.

You can’t build a 200 gigabyte model and expect it to run on like a small little piece of hardware. And so we’re the software platform for that. And so your end customers, are they product managers? Are they CEOs? Or more like engineering and DevTools? Yeah, we love engineers here.

They’re all engineering at heart. Our customers, our users are engineers. Now, I recognize that sometimes these engineers, they can do it themselves.

So what we’re doing for, what we’re solving for is, yes, you can string together 20 different tools. But again, we make it easy. We allow the engineers to focus on the higher value stuff.

We’ll do some of the grunt work around data labeling. I mean, it’s all the necessary stuff that needs to happen so they can focus on the more important things. And so a lot of times what they’re trying to solve for is, they could be product companies.

They want to make their product, we like to say, smart. These are some of the use cases we’re hearing a lot about. It’s maintenance, fault detection, things that you really want to get a handle on as soon as possible because you don’t have time to send it to the cloud and back.

Yeah, yeah, very cool. What are you most excited about this show? Is there anything jumping out? If you look at this show, I mean, it’s kind of a weird show in a sense of, in a good way. You got like IoT, you got hardware people, you got security people, you got everyone here.

But what I like about this show is like, I don’t think there’s any show like this that brings together like the brain power of all these different industries. And it’s just the thinking and the innovation. Like when you have hardware mixed with software, mixed with security, mixed with edge, mixed with AI, big data, right? There’s some really good, really great conversations that are happening.

I agree. I was walking the show with Susie earlier and we were a little concerned. I felt like the conference was actually spread a little thin.

It was like not concentrated enough, but like actually getting in and talking to a couple of the vendors. Really a lot of great conversations, a lot of good interactions. We’ve seen our, this great, we’ve got customers coming by, like I was surprised, like, hey, we have prospects coming by.

Of course, a lot of partners come by. So it’s been a good show. And of course, new people that have come by that we’ve just have not heard of us or the best ones are, we’ve heard of you guys.

We were just referred to come by to talk to you. And I love those conversations. Awesome. 

Well, thanks for coming on the show. Check them out at Jim Pulse. It’s been following for a while.

Very cool product. Thank you. And I’m going to be the last pitch.

We have a free version. Come to our website. Check it out.

It’s all free. I promise I won’t be spamming you. It’ll be someone.

I’m just kidding. Just try it out. You’ll love it.

Thank you. All right. So welcome to the show Charles with Adam Beam.

Yeah. Thank you very much for joining. Tell us a little bit more about what you guys have going on.

Yeah, sure. So what we do really is we take data and we transform how computers think about it. We make it into something else.

We make it into code words. The whole concept here is to drive dramatically efficiency in computing. We can quadruple the amount of data that goes through any pipe.

People just need to send more data or they need to cut the cost of sending data. We can do that by sending four times more data to the same pipe with almost no extra power, without almost no extra latency. And not only that, as a side benefit, it is natively secure.

Let’s go a little bit deeper. Tell me about some early use cases or customers. So one good example is Viasat.

So Viasat is a partner of ours. They’re very interested in working with us because how do you get, if you have a half billion dollar satellite and you can’t send up a repairman to upgrade it, how do you make that satellite more productive? Well, if you use our stuff, you’re sending the same information to that satellite, but at the same time, you’re getting four times more capacity out of that satellite. And so if you’re, for example, say you’re a big heavy equipment maker like Caterpillar, you have an enormous amount of data that you need to send from some remote location over a satellite to a cloud.

You can cut that down by 75%. 75% less. Or you can send 75% more data.

Depends on what you need. But 75% less money or four times more data. It depends how you look at it.

But either way, the amount of efficiency gain plus the security is huge. And so where does your software actually run and how would a customer go about thinking about using this? Good question. So it is normally implemented as a SaaS.

So it sits on the cloud and somebody would, say, buy an Internet gateway, an IoT gateway, a thing that connects to the cloud. And all it does is the SaaS, our SaaS, the NURPAC SaaS that Adam Beam has, pushes out to that device a software application. And now it builds a code book automatically and it pushes the code book out to the device so that there’s matching code books on both ends.

And now all that happens going forward, and it’s all invisible to the customer. So all it happens is there’s communication between that device and the cloud. And as far as the customer is concerned, they don’t see it.

So Ericsson’s Cradlepoint unit is working with us now to integrate it into their devices. And pretty soon you’ll just be able to buy a Cradlepoint gateway, and the Cradlepoint gateway will just do it. And you won’t even know it’s working, except that you see a lot more data.

Yeah, amazing. So you guys are at the very forefront doing a lot of cool stuff. What else seems to be very successful? This is really exciting.

What other companies here have you seen that are really interesting to you at the show? What’s jumped out to you at the show this year? So one of the things I think that is happening that’s going to be a big deal is having satellites directly communicate with cell phones without being modified. I mean, Global Star over here took a big investment from Apple, right? And that money they’re using to launch more satellites. Why? So those satellites, 85% of that capacity of that satellite is reserved for Apple.

So guess what? That’s not just pretend. That is real intention to move from just emergency use of direct-to-satellite to, okay, I can do texting, okay, I can send emails, et cetera. I think that that’s going to be a huge trend.

You’re now the first person on the show that’s been talking about satellite. There’s a lot of exciting stuff happening there right now. Yeah, very cool.

Well, that’s it. That’s a wrap for today. Thank you for coming on, and best of luck for the rest of the conference.

Thanks very much. And come to AdamBeam.com and check us out, AdamBeamTech.com. Hey, everyone. I’m here with Virginia from Fuel AI.

Virginia, tell us about your company. All right. So we are a two-sided marketplace.

You can think of us as DoorDash for data. We connect people who take photos with their smartphones to AI builders that are building a company that’s building an AI solution for first-party data. We gamified the system to make it a little bit more fun.

So we have bounty hunters on one side, the supply side. And on the buy side, we have companies that are building an AI solution. They might come to us and ask for audio files, video files, images, the type of data that can’t be stored in spreadsheets or in Word documents.

And then we put those requests together, we call them bounties, and we put that out to our bounty hunter network. We have about 8,000 bounty hunters right now in 20 different countries. And they decide, just like if they were driving for Uber, do they want to go and collect that data today or do they not want to work today? So if they want a little bit of extra cash this week, they can go ahead and, you know, collect the data.

And if not, they can pass up that opportunity. Sounds like data that doesn’t lend itself to easily being stored in a database. Is that kind of the idea? That’s right.

Absolutely. Very unstructured data, very raw data that AI builders can use over and over again as well. Can you walk me through a, for instance, of client engagement? You could strip the name out, just say, like, you know, we, they were looking for this kind of data and we were able to go get that for them? Yes.

So we had a company that was looking for 10,000 images in, 10,000 in each country, and they were looking for five different countries of floors. So I’m sorry, continents. They wanted floors in five different continents, indoor floors.

So we just, they wanted rugs. They wanted hard floors. They wanted area rugs.

So we, we, you know, put that out for our bounty hunters and they go out and take the photos and then they submit those and they get paid. I don’t know if you have the insight, but can you, can you like dive a little deeper? Like, what would they be trying to solve for with that image? So they might be building a video game that they are trying to, you know, sort of capture what really looks like, especially if you know, that person in the video game, the avatar is kind of walking through a space. And especially if you’re doing using the Wii or if it’s something you’re capturing on your phone, if you’re playing on your phone and it’s like a lifelike video game.

So tell me, what brings you guys here? Tech X. So I’ve actually been here before with a different company before I found in this company. Tech X is huge. There’s a lot of companies here that are looking for data.

There’s a lot of interesting people to meet. So, yeah, this is, this is a great place to be. I love Tech X. I’ve been before, like I said, and I really enjoy it.

Last question. I have a special place in my heart for marketplaces. I know I’ve worked at one.

I know how hard it is to build, basically build two companies at the same time. We talked about some of the challenges that you guys have had to overcome building up both sides of this marketplace to make this possible. So, you know, when you build up a marketplace, it’s kind of like a seesaw.

If you can think about Uber, if you had, because that’s a two sided marketplace, people can sort of understand. If you have a hundred riders and you have a million drivers, your drivers are going to be sort of upset because they’re not getting enough rides and vice versa. So you’ve got to do this seesaw thing where you’re constantly building up one side and then building up the other and then building.

So you’re moving back and forth. That’s been one of the challenges. And then just going out and mobilizing those 8,000 bounty hunters.

We really got our beachhead market at universities because university students typically have a very flexible sort of sort of flexible schedule. I mean, they need to be in class, but outside of class, it might be challenging for them to work in, say, a coffee shop or an office because those offices have specific set hours. But this is something they can do on their own time.

They can kind of do it when they don’t have class, when they’re, you know, don’t need to study. So that’s when one of the challenges is just building that seesaw up from side to side and making sure that both sides are satisfied. Makes sense.

Virginia, thank you for being on the show. Thank you so much for having me, Ryan. All right.

Welcome to Over the Air. We have here with us today, Tomer from Autorize. Welcome to the show.

Thank you. So can you tell us a little bit about what you do? So Autorize is a workload die-in automation solution. It builds for platform engineering teams and security teams, allowing them to manage secure access between workloads and infrastructure, and catering to any security needs if they’re running through PCI compliance or zero trust initiatives.

I’m picking up on a little bit of an accent. Seems like you might not be from this area. Tell me a little bit about where you guys are from.

We’re from Tel Aviv, Israel. That’s where the team is based at. We’ve got an engineering department there.

And they’re the best team that I could hope for. That’s amazing. Yeah, a lot of security in Tel Aviv.

So what brought you to the show? What are you excited about? We’re giving a talk here tomorrow about adaptive compliance. And we’re launching our new security suite, targeted security teams, allowing them to run through compliance requirements without having to bug engineers or stop any product in workloads by saying, no one talks to anyone. That’s how we can be secure.

We want to offer them a new solution and a new approach, saying, you can stay secure, but still allowing new engineers to do whatever they want, make any calls they want. Got it. Yeah, a lot of times, security will slow things down and you guys are fighting that battle to not let that happen.

Yeah, very cool. What industries or sectors are you guys seeing early traction? Any use cases that are really interesting? Yeah, so we’re seeing a lot of traction in the financial and health care industries. And in any regulatory industry or enterprise customers who are looking to scale a solution to hundreds of cloud accounts, and they can’t have anything that’s scraped today, so homebrew.

They just want a solution that works and was built for their engineering departments. So tell me about the solder. That’s how it is.

Our mascot, he’s named after our company name, Authorize, and Authorize is how we say Authorize in a Hebrew accent. Okay, got it. Very cool.

Great. Well, thanks for coming on the show. It was great to meet you, and I hope you guys have a great conference.

Thanks for having me. All right, Eric, thanks for joining us on the show. Here we have Eric with Enfly, and we’d love to learn a little bit more about what you guys do.

Yeah, thanks for having me, Bill. I’m here at the studio at Tech Expo representing Enfly. We are a company that gives kind of broad visibility into data sets, and we allow you to see data at any arbitrary zoom level, as we like to call it.

You think of it a little bit like the Google Maps, but for any arbitrary data set coming from your sensors. Okay, so IoT and sensor data? IoT and sensor data. Well, IoT sensor data, we came from digital advertising as well.

That’s where we actually started, and so lots of ad impressions, marketplace auctions, stuff like that. We’re used pretty heavily in product analytics as well, understanding what people are doing on products, but we’ve seen a lot of uptick in sensor data coming from IoT, people trying to understand, you know, where the sensors are, who owns them, what’s going on with them, and get a global view of what’s going on rather than just the point asset by asset view. Interesting.

Okay, and so do you guys, are you running on the edge, or are you running in the cloud? We’re running in the cloud. Okay, cool. And what clouds typically is just like a managed AWS marketplace type solution, or how does somebody run your product? We have AWS Marketplace.

We are actually announcing at the time that this goes on air, we will have announced full support for SaaS on Azure. And we also have a bring your own cloud offering so the data can stay entirely in the customer environment, but we can get our software in there and running, and then there’s on-prem, of course. And you mentioned early routes in digital advertising, moving into the IoT space.

IoT, I’ve been doing IoT for a long time, it’s a broad, broad market, so what are some industries, some use cases that you’re really excited about? So some of the customers that kind of helped bring us over here, let me talk about them. We’ve got an EV charging company. They’ve got EV chargers all over the world.

Superchargers, you plug up your car, you run them. Turns out they have problems every once in a while. People call in, they’re like, hey, the charger’s not working.

Now, sometimes those problems are, they’re ephemeral, they go away, everything starts working, sometimes there’s something wrong and it needs to be debugged, it needs to be figured out, sometimes it might be with the charger itself, it might be with the equipment, all of that. This customer was having problems sending people out on service calls when they weren’t needed, they went out, they didn’t have the right equipment, they didn’t know exactly what they were fixing, and so by getting that telemetry going into our platform, they now have a remote capability to understand all of the telemetry, everything that’s going on around those chargers, around what’s going on for the whole station, what’s going on in that geographical area and around the world that gives them the visibility to understand okay, is this a problem with the chargers, is it actually something larger, is it something else where they can compare chargers to each other, and they’ve basically been able to optimize the sending of service people out and only send people out when they actually need to be there and have them sent out with the equipment needed to repair it ahead of time. So what’s the actual user experience on the back end? Is this, you have a full blown on dashboards UI UX, is it more of a command line, like a power console type deal? I’m sure.

So the data core, it accepts data in a schema-less manner, and so whatever data is coming in, we tend to be pulling in stuff from streams, connecting to Kafka, AWS IoT Core, something Kinesis, something like that, and whatever is in the data we will then make available, the database can be queried by a SQL, so you can just do SQL with it, our product offering then adds a visualization layer on top of that, that is built specifically for the exploration of kind of wide, varied data sets, giving you the ability to see at that most granular level, as well as aggregating up to the global level and everywhere in between, which is those zoom levels I was talking about. Well, it makes perfect sense. I’ll have to run by the booth and check it out.

Yeah. Great. Well, what have you noticed from the show? What’s jumped out at you? It’s really interesting.

I’ve noticed that AI and ML are on almost all of the booths. It is everywhere right now. Okay, okay.

We unfortunately don’t have AI and ML in our booth, but I’ve got a bit of FOMO right now, so… Kind of sharp. Okay, okay. We can go write it down there.

But yeah, so AI and ML definitely seems to be taking over the world. This is a little bit of a biased take on it, but as I’ve been exploring the IoT platforms, I’ve been noticing that there’s a common thread where everyone’s focusing on connecting the device, getting two-way communication with the device, making sure that you can do command and control over it. And they haven’t gotten to that analytics layer, and so to me, I’m like, hey, this is great.

This is a perfect opportunity. We can add that analytics layer on top of all the IoT platforms and kind of help people obtain more value out of what they’re doing. Wonderful.

Well, thanks for coming on the show and I wish you guys continued success. Yeah, thank you. All right, so day two at IoT Expo, and here with us we have Josh with Sesame Software.

Josh is also a Nashville local, so we might get into stories on Broadway by the end of this episode, but not really. What do you do? We help speed up time for completion of projects by automating the data movement and migration of application data for analytics and speed up the time by shortening mapping projects. So tell me a little bit more, mapping projects.

What does that mean? So instead of having to write out the entire application, mapping data types between databases from like Salesforce or NetSuite down to your local database, we can just take credentials and automatically build that database for you and go in between databases, automatic aliasing, auto schema adjustments and auto increment so we can move only the data that’s changed and keep your analytics up to date. Great, very cool. Where do you guys see interaction? What are some industries that are excited about Sesame? A lot of government, finance are some of the industries that we work with, mainly insurance, education started popping out more, people wanting to keep their data secure and control it in their own environments.

Anything in the Nashville area? You know, we’ve got a lot of new industries coming into Nashville, a lot of people moving down from New York, like operating standards building to oracles moving over to Nashville. So I think there’s a lot of growth coming simply for that region. Yeah, very cool.

So what’s it like fellow, a brother and a tech guy in a not tech city? Do you enjoy that aspect or do you like really enjoy like getting here to the Valley and being kind of upfront and up close and personal with like where everything’s really happening? You know, it’s great to be first to a region. You get to set the tone, right? So you’re setting the tone and being able to start those meetups and introduce people to the city as they’re coming up. And I feel like the people moving into town, the big businesses moving into town, are going to bring a lot more tech to the area.

And so getting to be here first is a great opportunity. All right. Well, thanks for coming on the show and tidying up.

All right. I know. Hey, folks, I’m here with Lee House from IOT-83.

Lee, tell us about your company. Well, IOT-83 is an application enablement platform built for enterprise-grade solutions. And tell me what’s unique about you guys.

What’s unique about us is we’re solving a problem. The ecosystem of IOT really hasn’t addressed before. What customers want is affordable solutions, affordable manageability of the solutions and fast time to market.

But they also want custom. So that has been a real problem. What we’ve built is a really solid platform that is easily customizable by the customer by managing only a few lines of code in order to deliver unique applications that have cybersecurity, scalability, reliability, and flexibility.

I love it. Okay, cool. And so we’re here in Santa Clara, you guys’ home turf.

What brings you to IOT Expo? Well, we want to demonstrate what we think is a game-changing solution for industrial OEMs and enterprises and to learn what else is going on in the industry so that we stay current and engage with all customers. I love it. And last question, talk to me about, like, you know, I hear a lot of people say they want custom, but they want fast.

You guys have pulled it all together. Talk a little bit more, if you would, about, you know, some of the technical problems you guys have had to solve to bring a product like this to market. Well, what we did was we built a really solid underlying architecture that has all the scalability, reliability, cyber security, and function.

And then we put on top of that some 35 platform services that can be reused. So all of this is highly reusable. The customers really don’t touch or manage this.

But on top of that, we’ve got a more generic IOT application and that’s where the magic happens. Pull that apart, look at the SDKs, make modern modifications, and all of a sudden this application is unique to your particular requirements by only changing a handful of lines of code, leveraging the hundreds of thousands of lines of code below it. So it’s a pretty powerful story.

Is it kind of an 80-20 split between product and then the last mile customization, or how do you guys think about that? I think so, because you’re leveraging all of this. We’ve had 130 engineers for five and a half years working on this platform, and now only by modifying a few lines of code, you’re leveraging the whole thing. So it’s maybe more than 80-20.

I love it. Lee, thank you for being on the show. It’s a real pleasure.

Today we have Rick here with Contron, and tell us a little bit about Contron. Well, yeah, Contron is pretty well known for hardware. If you look at our website, there’s lots of hardware on there, but also, we have a lot of software capability, right? Okay.

Right from embedded software, right down at the sort of BSP level, working on Android and Windows and Linux and QNX, even bare metal devices, right up through the stack, right up into the cloud, producing SaaS products and hosting and managing them with the DevOps team. So because we’ve got all that software capability, we can produce entire kind of IoT solutions. And so what we tend to do is combine our hardware products with our software experience as well as software products that we’ve created using all of our kind of common experience with customers to get into complete IoT solutions as customers need them.

Very cool. And so you guys, are you a solutions provider? Are you a platform? IoT space is a lot around. I think we like to think of ourselves as a solutions provider, but underpinning that, we have lots of products, right? Because we’ve seen lots of common use cases.

So the kind of, the core kind of things we see are device management, where people are really interested in making sure the devices are running correctly, making sure they’re up-to-date with software or content or even the OS, and being able to kind of remotely access them to make sure they can fix any problems, to avoid truck rolls, that kind of thing. And then the other kind of main group of use cases is telemetry-based, where people want to kind of get information out of devices so they can make operational improvements, so they can look at things like energy usage, all kinds of telemetry really. So we try to build a whole lot of flexible product pieces that we can combine and mix and match to solve solutions.

And yeah, customers tend to engage a little bit, and they’re like, they focus on one particular use case, which they want to kind of prove out the ROI with, and then it kind of snowballs. We can bring in more products and give them a bigger solution, and solve more and more problems, and make them more and more efficient. So what would you say is your ideal embedded runtime? Like, what’s your favorite operating system? You mentioned a lot, so… Yeah, well, I think, you know, we also sell Windows IoT instances, so that’s pretty good because it gives us additional revenue, but we’d like to be agnostic, right? We’d like to use the right tool for the job.

I saw on your website you have your own operating system, your own OS. Yeah, we’ve got a control OS. It’s a Linux-based OS.

It’s got the ability to kind of update yourself and roll back when things go wrong. So that’s one of the products we can use as part of the bigger solution. And what is the most interesting application you’ve shipped to the Edge in the last year? We’ve got lots going on.

One of the most interesting customers we have at the moment is a company called Arcus, who are a facilities manager and company in the UK. So they look after a whole load of customers’ facilities at their places in the UK, and primarily what happens is when there’s a problem, they send out the right sort of engineer. They triage that problem and then send out the engineer.

And we’ve been working with them for a long time on cloud-based solutions to enhance that experience, right? Because they not only look at problems, but they also look at potential opportunities to improve efficiency, so energy usage, that kind of thing. So we’ve worked with them for a long time to create these cloud solutions that allow them to provide a better service to their customers. They’re very forward-thinking, which is great.

And quite recently, we’re working with them now to put in place sensors and gateways into these, what they like to call tech-poor sites who are like, they haven’t got lots of smart equipment, but as soon as you start putting in a few sensors here and then you can start to detect problems early, which means that you can solve problems really quickly. You can send out engineers really quickly before the customer maybe knows there’s a problem. So we’re working with them to really kind of revolutionize the facilities management market.

And it’s really great to work with a company like that who is really forward-thinking. They’re really brought into the IoT idea and they’re really always looking for opportunities to kind of leverage partners like us. And what are you most excited about with this conference? Just seeing all the different kind of use cases for IoT, really.

Seeing what people are up to, what they’re finding value in, you know, whether people are, there’s lots and lots of AI, which is really interesting. Like, everyone seems to be really looking onto the kind of latest generative AI stuff. And so I’m looking at it, seems to be the key to things.

Yeah, just seeing what people are doing, really, and what challenges people are facing. It’s really interesting to see, I’ve been in software for 20 years, and it’s really interesting to see how IoT has kind of defined itself and then kind of how some of the more recent technologies, like really low power sensors and AI are kind of changing the story a little bit. Great.

Well, thanks for coming on the show. You enjoyed the conversation and we’ll talk to you again soon. All right, thank you.

All right, cheers. So, Brian, thanks for coming on the show. Sometimes when we’re at these conferences, we just pick up interesting people off the floor that aren’t presenting, but I know you’ve given a few talks at the conference so far, and I’m curious, what are you doing here? What are you talking about? Yeah, it’s a great question, Bill.

I’m just glad for the opportunity to be here. So there’s something like 7,000 people here over the course of two days in Santa Clara talking about IoT, cloud, security, and then all of the normal buzzwords like digital twins and AI. And what I’m constantly curious about is how can we build better tech? So I work for National Grid.

We’re the utility provider in Massachusetts, upstate New York, Long Island, and England, and we have a tech stack that works well for us, but we’re constantly curious what could work better. And so having opportunities to have all these vendors here, to talk to them, to understand what they bring to the table, including these companies yourselves, it’s very interesting because it gives us opportunity to continue to explore possibilities. You guys are in the middle of a major rebuild and rollout right now.

What’s going on? We’ve been given approval by the state of New York and the Commonwealth of Massachusetts to replace existing electric meters with smart, connected, they’re called AMI, Advanced Meter Infrastructure, devices. So these devices, they connect back to cellular and they connect to each other, hop to hop to hop as well. So they really, the key there is, give people more information about their utility, about their energy spend, about their energy use, so that they can continue to save money, make choices, use more energy when it’s more affordable or when it’s more green, curb energy when it’s more expensive or when they choose that it’s more fossil.

Give them those opportunities where they can make those choices with real data about where their energy is coming from. Really quick, has there been anything like really surprising in this journey on this project or any key lessons that you’d like to share? One of them, Bill, that probably sticks out more than most is the partnership. So for us, we are a utility company.

We’re not a network company. We’re not a cellular company. We’re not a metering company, but you’re doing all of it.

But we’re doing all of that, so we need to rely on our partners to do that. And there are some capabilities that the partner is exceptional at and some capabilities that the partner is not exceptional at, but they’re continuing to mature. So how do we continue to give best-in-class guidance to them as a partner without pushing a contract on them and saying, you didn’t meet the SLA? Because that doesn’t work.

Yeah. Anything on the show floor that’s really jumped out at you? IOT, AI, digital twins, they’re all buzzwords. And it’s frustrating because then you talk to Boston Consulting or Ersin Young or Deloitte, and they’re like, here, take a pitch deck of all the buzzwords.

Well, that’s good. And it’s helpful to get accomplished what business cases we need to do, but really we need to learn how to build better tech that’s closer to the business, in my opinion. So what are your thoughts on, you know, what you’re learning here, what you’ve seen in the last few months, and you know, you do this a lot.

Yeah. So what I’m most excited about, and we’ve talked a little bit about this on the show already, is just the like edge computing really coming of age. And folks truly needing to understand that it’s not just a central application server that they’re managing anymore, but you have hundreds, thousands, tens of thousands, or hundreds of thousands of computers in the field now, and they’re not just simple temperature sensors phoning home data.

Your application is probably, and itself is probably more nuanced than that. And maybe the first pass you’re like, hey, we’re just doing temperature reading or sensor reading, but now we need to bring in more advanced functionality on the edge. So I’m really excited to see more capabilities on the edge, whether that’s like what we do at Peridia, which is device and fleet management, but also a lot of custom databases happening, new communication protocols.

But yeah, there’s a lot here. It’s been a very interesting show. The conversations have been really good.

I think it’s been a good convergence of a lot of different interesting areas, even through security. Yeah, it’s been a lot of fun. It’s a good show.

Well, like you mentioned, the coming of age. I think a good way to look at that is, you know, just physically looking at the different corners. There’s an area in each corner, right? Yes, like six of them or five of them.

They’re very intertwined. In the past, I would say, cloud, IoT, edge, data, they’re really separate. Like, they need to be separate organizations.

They need to be separate leadership in the company. And I’m starting to see these really converges, like, this is tech. Well, awesome.

Yeah, well, thanks for coming on the show. It was a pleasure. Thank you.

IoT Tech Expo. Today, we have Hardware I.O. And this is Rene. And he has traveled a long ways to come to California for the show.

Yes, that’s true. Yeah, we came all the way from Czech Republic. So, yeah, ready to be here and very happy to be here.

Awesome. Well, you got a very interesting product line here. And I was wondering if you could just tell us a little bit more about what you guys do.

Yes. So, we are basically developing and designing cover of Gateways. Yeah, we develop, design the hardware, mainly the Gateways.

We also do the custom engineering projects, let’s say. And the main focus for us is basically to collect the data, digitalize it, and send it to the customers. And so, your customers, are they looking for more a complete solution that they can take to market, or are they running their own software on your devices? How does that typically work? Normally, our part is, let’s say, to design the product, the hardware, and bring it to the partners who make the final dashboard applications, and then bring the data, let’s say, to the final customers.

So, we would like to look for a partnership with the companies who will analyze this data, do some, let’s say, final dashboards, et cetera. Got it. So, we are basically collecting the data and sending data to their applications.

What has jumped out to you at this conference? What’s been some big takeaways, or anything that you’re excited about? It’s good conference for American market. It’s kind of our focus to be more successful in North America. So, we are happy to be here, let’s say, to present our solutions.

We have some existing projects in North America. This is our project with our partner for monitoring the grain silos. So, basically, it’s, again, our platform, Chester, connecting the one-wire temperature and humidity sensors.

And, basically, with this kind of concept in the grain silos, we can monitor the volume of the grain and also the quality of the grain. Great. So, a friend of mine mentioned your company to me just earlier this week, and I looked up and saw your booth here.

So, really excited to see you guys at the show. Fantastic products, and thanks for coming on. Thank you.

Thank you. Bye. Good.

Hey, everyone. I’m here with Deepal from Interphase, something about your company. Hi, Ryan.

So, thanks for having us here. So, Interphase IoT is a startup, privately funded company, focused on ultra-low power Wi-Fi, BLE, chips, and modules, built specifically for battery-operated IoT devices. So, we shine when people want low power, long battery life.

So, for smart home use cases like video cameras, video doorbells, or your typical sensors in the phone, window shades, thermosets, fire alarms, things like that. Also, we focus on industrial markets for vibration sensors, medical, as well as commercial buildings, you know, where you have operation, occupancy sensors, air quality sensors. Wherever you want sensor information connected to the cloud.

Do you guys think of yourselves as a, are you a product company, or are you doing custom development work for other folks, or? We are a product company. We build our own chips, which are manufactured at the SMC. We package them.

We build modules around it. We build reference designs with our solutions. So, a customer can get a complete reference design solution, hardware software, ready to go.

And then we have a design partner program. So, we have a lot of IoT design services companies that take our reference design, productize it for the end customer, customize it to their needs. And essentially, you can get a complete IoT device solution from onset to deployment in six to nine months.

What brings you guys here to IoT Expo? So, it’s an opportunity for us to show our latest technology and capabilities. So, we have been in production since 2020. We have had reference solutions.

And actually, we do have customers in production also. There are many more getting into production in a quarter or two. But we want to showcase our newer use cases.

So, today, we are showing our Matter 1.2 certified solution. You mentioned Matter cameras, headphones. Do you consider yourself specialists in the consumer-facing side of things? Consumer is definitely a big focus.

That’s where a lot of the volumes are. Things in the industrial world, medical, commercial space also. That makes sense.

I see that you brought a couple of things here for show and tell. So, for folks on audio, you’re going to have to paint a picture. But what are we looking at? So, we sell chips and modules.

That’s where we get our revenue. But to make it easy for our developers, we have reference designs that we built. So, I’m showing this an evaluation kit with Arduino interface with standard built-in PC, PCB antenna and all that.

So, you can check out the power, the performance that we deliver. All the sensors are already built into this evaluation kit. The SDK is already programmed in.

So, within an hour, you can check all the performance and power, cloud connectivity, firmware update, all the features. And then we have built this scalable. This you can connect to an SD micro nuclear board through this interface and get a complete device-side solution.

The hardware software is degraded, done with the SD micro, iQube ID that they have. We are also showing the camera reference design. So, rather than saying, hey, you can build a low-power camera, battery-operated, we have a complete battery-operated camera reference design built with our chip, Wi-Fi, with the video chip from InGenic, as well as the cameras and image sensors, showing not only low power through our chip, but also complete power management of the whole device.

This seems like the next thing. You know, pushing everything to the edge, compute at the edge, transmit as little as you have to. This is the next decade or 50 years or beyond.

Deepal, thank you so much for being on the show. I really enjoyed talking to you. Thank you very much.

All right, folks, IoT Expo 2024. That is a wrap, Bill. Three key takeaways. 

You go first. The edge is real. I’ve been talking about this for a long time.

Compute on the edge. Quit sending stuff to the cloud. People are taking full advantage of that here and just lots of applications.

For people who wrote its obituary a little early, California and the tech scene is still very, very real here. This is the most tech of all the tech shows that we’ve been to so far. Very impressed with the show.

Back to San Jose. Back to San Jose here in a few weeks for Sensors Expo. Play us out with your third takeaway.

Yeah, security at the edge. It’s an unsolved problem. Lots of folks still innovating in this space and trying to deliver high-security applications and get out of engineers’ ways.

All right, folks, 2024, that’s a wrap. See you next year. If you’re out there trying to figure out how to get started with your IoT project, I’ve got some news.

Very Technology has just published an e-book called How to Choose the Right IoT Company. It’s a guide designed to help you ask the right questions, narrow down your search, and find the right partner. You can find it on our website at VeryTechnology.com.

Discover valuable insights on how to find the best IoT partner to help you create a solution that aligns with your business goals here.