Hello, and welcome Works With Virtual 2024. We're thrilled to kick off the fifth annual IoT Developer Conference from Silicon Labs, where innovation in wireless technology and IoT solutions come to life. This year, Works With expanded globally. In addition to the virtual event, we're hosting in-person events across major tech hubs like Austin, San Jose, Hyderabad, and Shanghai. This expansion underscores our commitment to empowering IoT developers worldwide. Works With is the signature event series that unites a global network of device manufacturers, wireless experts, engineers, and industry leaders. Over 500 companies have joined us this year alone in person, making it a powerful hub for the latest in IoT technology. Today, we're honored to have Matt Johnson, CEO, and Daniel Cooley, CTO, take the stage to discuss how AI is accelerating the IoT, a transformative topic that's reshaping industries across the world. Additionally, NVIDIA will join us to share insights into connected health trends and the role IoT plays in advanced healthcare innovation. A huge thank you to our platinum sponsor, Arrow, for their continued support and helping make Works With possible. Thanks again for joining us at Works With. Let's dive in, and let's get inspired together. Ladies and gentlemen, please welcome the President and CEO of Silicon Labs, Matt Johnson. Everyone, welcome to Works With 2024. If you've participated in events at Works With in the past, you're familiar with our goal of providing attendees with in-depth technical content to accelerate your connections, whether it's advancing your product development or strengthening relationships with ecosystem partners. Before we begin, I want to take a minute to express my gratitude to all of our customers and partners here in the room. The last few years have been an unprecedented time in our industry, and your trust in us has been crucial in helping us navigate these times. We've come through this together and stronger, and for that, I want to say the most sincere thank you. It means a ton. As you know, at Silicon Labs, we're not just participants in the wireless ecosystem. We're leaders and innovators and proud of this role. As one of the few founding members of Thread and Matter and a key partner for Amazon Sidewalk, we've been at the forefront of driving critical wireless protocol advancements for a long time. Our commitment to open source and our active involvement in standards such as Bluetooth, Wi-SUN, Wi-Fi, Zigbee, and Z-Wave through various industry alliances reflects our dedication to pushing the boundaries of what's possible. Our breadth, depth, and focus on the IoT is unparalleled. Works With began back in 2020 as our way to bring the industry together, the best minds in this industry all in one place. This event is all about connecting people, products, partners, and ecosystems. We saw the embedded wireless space being developed and adopted on a massive scale, starting in smart homes with devices like Google Home and Amazon Alexa. We saw the industry's need to create a space where developers and partners and products and ecosystems could connect. We then expanded our focus to include smart cities such as Wi-SUN technology and then integrated Amazon Sidewalk into the mix. And this year, we're excited to add connected health to the lineup, recognizing the crucial and growing role of connectivity in healthcare. So now, let's talk a little bit about the IoT. Everyone here in the room has a remarkable opportunity moving forward because of the IoT. By 2050, the Earth's population is projected to reach an astounding 10,000,000,000 people. It's difficult to process or imagine that scale. That's driving growth and unprecedented trends in economics, urbanization, the environment, and technology. And all of these are incredible challenges, but also a catalyst for incredible innovation and progress. And this is particularly true in the world of connectivity. Today's application developers in the cloud have more reason every day to connect anything they can, not just phones and computers, but also embedded systems that are all around us. The number of connected devices, the things that touch virtually every aspect of societies and economies, will reach over 100,000,000,000 devices in the next decade. 1 00,000,000,000. These connected devices unlock unparalleled capabilities, productivity across a vast range of industries, from manufacturing and retail, transportation, healthcare, energy distribution, fitness, and agriculture, all revolutionizing every one of these sectors. As we look at the state of the IoT today, we're sometimes asked how we see the potential of this space post-pandemic compared to pre-pandemic or pre-supply chain crisis, however you want to think about it in our space. The answer is that we believe the potential for IoT adoption has expanded for a few reasons. First is that the underlying technology performance and reliability is at a level that is ready and able to support broad adoption. The second is that our customers are seeing good returns in business models adopting IoT technology. And there's so many examples of this. Some that we talk about a lot would be digital shelf labels or electronic shelf labels or smart metering. The economics there are allowing faster adoption and accelerated adoption. The third is the unprecedented capabilities like Matter and Amazon Sidewalk. These were all fundamentally created to drive the proliferation of connected devices. That is essentially why they exist. And the fourth and last, but not least, is artificial intelligence, a trend that is likely, I believe, the biggest and probably the most transformational technology we may experience in our lifetimes. It's also a trend that we can't ignore in the IoT, particularly the symbiotic relationship between AI and IoT. I'm going to explain what we mean by that. As part of that, we're going to talk a lot about AI in this presentation, but I also want to be clear that we're not up here trying to rebrand the company or say that we're an AI company now. We're not saying that at all. What we're saying, there's actually enough companies out there changing their strategy because of that. What we're saying is our goal of being the undisputed leader in IoT wireless is unchanged, full stop. We're making great progress towards that goal by being the largest company in the world dedicated to IoT wireless. What we see with AI is the catalyst for accelerating global adoption of Silicon Labs solutions and, more importantly, accelerating the global adoption of wireless and connected devices, and I'm going to walk through the three reasons why we believe this. The first is AI can help make the IoT smarter and more intelligent. Everything in the IoT has always been smart something, right? Smart homes, smart buildings, smart cities, smart everything, but in reality, a lot of these things and devices are connected, but not as smart and not as capable as we wanted or needed them to be. This is in part because they don't work together as a cohesive intelligent system as well as they could or should, lacking contextual awareness, natural language interfaces, and meta-intelligence beyond the individual sensors and devices. AI is going to change that. It's going to allow devices to be a lot more useful and impactful, fulfilling the original promise of the IoT, a smarter IoT. The second reason that you hear a lot, and it's because it's true, is data. The sheer volume of data already being produced is staggering and virtually impossible to manage without intervention from AI. AI enables analyzing this often unstructured data in real time, identifying patterns and anomalies, producing insights, predictions, and intelligence that was never before possible or so readily available. More importantly, not only does AI bring the ability to take advantage of existing data, but the need for more data will be insatiable, driving the need for more devices and connectivity than previously thought possible or imagined. So a bigger IoT. And the third reason, AI inference at the edge will continue to grow and accelerate. AI processing will not only take place in data centers. It will take place at the edge, our edge for everyone here. Today, Silicon Labs has production solutions on our Series 2 platform, our current platform that enables industry-leading machine learning inference in battery-powered applications. According to the latest MLCommons benchmark, which is publicly available, our xG24 delivers more performance per unit of power consumption than any other wireless SoC. These products enable our customers to have entirely new features and capabilities. And even more exciting, Daniel's going to share how our next-generation platform, Series 3, will take that already industry-leading capability much, much further. In short, our commitment to the IoT is absolute and only becomes more exciting with AI. Artificial intelligence enables us to provide even more capable products. At the same time, it will accelerate the number of connected devices and enable capabilities and experiences that were previously not possible. So with that, I'm excited to introduce our CTO, Daniel Cooley, who will take these concepts even further for everyone here. Thank you. Hello, everyone. Thanks for joining us at Works With 2024. It's really good to see you today. Like Matt said, the IoT is all about data. After all, one of the fundamental reasons IoT devices are more valuable is because you can shuttle data on and off these devices at will. You can do this quickly, cheaply, and reliably. The result is that data being generated by internet-connected devices is increasing exponentially. Next year alone, it's estimated that up to 80 zettabytes will be created by internet-connected devices. That's 80,000,000,000 terabytes in one year alone. Now, we're all part of one big ecosystem generating, processing, transmitting, storing this data. And how we actually use it will kind of be the key to the future here. We'll come back to that in a minute. But first, we have to understand why is all this data being generated? How can it add up to such large numbers? The reality is the explosion of connected devices. We're not just talking about what you think about as IoT. It could also be computers, smartphones, and a bunch of other things, but it's driving social media usage, expansion of e-commerce, digitization of industries and services. Every click, every swipe, interaction, vibration, temperature change, badge read is a piece of data. It contributes to a vast ocean that holds all this potential. And going forward, the barrier to getting that data is coming down, comes down every single year. It's really up to our imagination on what we can and should do with the data once we get it. Things that were either not viable or, in some cases, unthinkable are a real part of our future, and I'd like to share an example of that. Kind of going back, my grandfather graduated from the University of Kansas in 1950, and it's a while back, but he spent the next 60 years designing power plants and power distribution systems here in the. and in Asia. Late in his career, when we started talking about this IoT, I asked him what he thought about the smart grid, and he said that, in his opinion, the success or failure would hinge on kind of a couple of things. The first is, do you trust your ability to acquire data real-time? Can you secure the system end-to-end? This made him nervous because lives were on the line. He also acknowledged that power distribution systems were mostly at an impasse. They had really optimized so much that the only way to go further was to get real-time usage all the way down to the meter. They had simply optimized everything up to that point. It all comes down to data and trust. Not surprising, but sifting through the data is a pretty big task. It's not because we don't want to do it. There's just too much. This has been said many times in many different forums in the last couple of years, but we're going to have to take advantage of the benefits of machine learning and artificial intelligence to get everything we want out of it. Fortunately for Silicon Labs, we've been thinking about this for a long time. We're not new to the IoT or where it's been going. And although it's still early days for AI and the IoT, I like to say machine learning is what wireless was 10 years ago. We were the first company to launch IoT SoCs with wireless at their core. And we are the first to launch machine learning-optimized solutions in those products. These aren't standalone dedicated machine learning chips. This is market-leading performance embedded in the same wireless SoCs that you used to buy from us. Earlier this year, when we launched our xG26, that is an example of a product with this and its core tenets. So we're doing for machine learning in the IoT what we did for compute, for wireless, for security. You do not have to sacrifice cost, power, or latency to implement machine learning inference anymore. Our solution is leading the market in benchmarks, like Matt said. The MLPerf Tiny is the one he's referring to, and you can go to the website and check it out yourself, or better yet, buy a kit and get started today. We're bringing a great solution, but we do not own the data. You do. Companies like Silicon Labs play a crucial role in helping you get that data onto and off of the devices. We help you connect them together and bridge to the internet itself. This crucial data pipe is the very essence of the IoT. From me to you, I thank you for trusting us and helping you keep that wireless link alive. Beyond that, we help you secure your products, the secure lifecycle from key injection to commissioning to firmware updates and all the way to decommissioning. We take this responsibility really seriously because without serious engineering, it's my belief the IoT will crumble like a house of cards. You want to protect your products, your IP, and the data flowing around these networks. Lifecycle management is one of the pillars, one of the key foundational elements to a scalable Internet of Things. But it's not done. To take it to the next level, we have to make these devices smart. Processing data locally will be a massive piece of this. Once you fine-tune your ML inference engines, you simply won't have to transmit that raw data anymore, or you may simply transmit the action instead of the big data dump. The end product will simply act, and this has a lot of benefits. It extends battery life, radios are power hungry. It reduced cloud computing costs, cloud computing storage, and the associated liabilities from them. It increases spectral efficiency so we can get more devices on the air, and of course, it increases the security and privacy because that data is processed locally, and only then it will decide if it needs to transmit it further. I also believe the IoT will be a very important part of the next generation of ML training. It's only in the IoT that you're going to get the corner cases, the information that's useful for that next round of training. When that inference engine is 50-50 between resolution on some classification, that's a very important piece of data. You're going to want to pluck it out of the field, put it in the cloud, which we'll hear about in a second, and run it in your next round. The winners in ML will be the people who have IoT installations. You won't subsist on synthetic data anymore, and you won't subsist on open-source data models. So Silicon Labs is just one part of this ecosystem. On the other side of those data pipes sits a very different company, one who is leading the shift to AI and in many ways defining it all together. To tell us more about that, we've invited David Niewolny here. He is the Senior Director and Global Head of Business Development for Healthcare and Medical at NVIDIA, and he's going to talk to us a lot more about our AI future. Thanks for being here, David. Thanks, Daniel. Good morning. I could not be more excited to be with you all here at Silicon Labs Works With 2024. Such an amazing event connecting all aspects of the IoT ecosystem, and for me, frankly, it's been a nice little kind of homecoming and reunion. My name is David Niewolny. I lead the business development for healthcare and medical at NVIDIA. I spent 15 years in the embedded space with Motorola, Freescale, and NXP, a majority of those years leading their connected health efforts. I had the opportunity to follow the data to Amazon Web Services and help organizations structure their data and work with customers to figure out how to build SaaS business models to make good value out of that data. For the past four years, I've had the opportunity to work with NVIDIA, working with healthcare, med tech, and digital health organizations to accelerate all of their AI initiatives. And today, I'm super excited that we're going to explore how AI and IoT together are transforming industries. It's so amazing to see so many leaders from many of the world's most important companies represented in this room today. Just as Matt and Daniel said, there's something absolutely amazing happening right now. As you heard, industries, companies are being transformed. The computer industry is being transformed. It's hard to argue that the computer isn't the single most important instrument of society. It's powering fundamental transformation in every single industry we work in. AI by itself is driving the need for a 2x increase in computing power every six months. This is unprecedented. Accelerated computing was born out of necessity. Accelerated computing has been a game changer. It's really paved the way for the rise of AI and most recently, generative AI. AI is truly transforming everything from enterprises to healthcare by making data-driven intelligence a valuable and manufacturable resource. This is truly driving a new and next industrial revolution. At NVIDIA, we're committed to accelerating organizations' data and AI strategies safely and effectively, ensuring that all of you in this room can leverage its full potential while maintaining high standards for safety and data integrity. With AI, enterprises like yours now have the ability to unlock new capabilities, optimize processes, create new novel business models, all thanks to the power of AI platforms. Just as previous industrial revolutions were powered by things like steam, electricity, this new industrial revolution is powered by IoT devices, data, accelerated computing, and AI. Industries from healthcare to automotive to manufacturing and telecom are all embracing AI-powered solutions to generate insights and drive efficiency. The AI infrastructure that all of these innovations are powered by is what we're calling an AI factory. In an AI factory, data is processed into valuable intelligence, which in turn powers this next wave of AI applications. And this wave is what leads to real economic transformation across a number of different industries. To seize this generational opportunity, companies like yourselves, countries around the world are building NVIDIA-powered AI factories to both process, refine, and manufacture intelligence from data. Intelligence that now comes in the form of tokens. A token, it's a new kind of commodity, a completely new kind of product that's going to create revenue opportunities for all of your businesses and generate more than $100,000,000,000,000 of new revenue in the industry. So much of the AI we're using today involves using digital text. That digital text can be fed into this idea of an AI factory where it becomes compressed into a large language model with trillions of tokens that become billions of parameters. And those billions of parameters are really what become your AI. This is essentially reading, learning, and knowing how to respond on the information that it just learned from. This requires today a single computer, a giant single computer, one with tens of thousands of GPUs, a single computer at data center scale. But what's most important for all of us in this room is the next wave of AI. And that's really where I want to focus a lot of the conversation today. We need to understand what's going on in the physical world. The physical world AI is going to be the next wave. It's so important to put together something that is much more complete than just an AI factory. What you need is three purpose-built computers. The first is for model building. That is the AI factory we've already talked about. The second is a dedicated computer for simulating the physical world. This is how you optimize the models. And third, as Daniel and Matt mentioned, inference at the edge is important. You need a computing infrastructure for inference at the edge. And these edge devices for inference aren't just for inference. They're also for aggregating data from the billions of IoT devices that you guys are all deploying. It feeds that data back into the AI factory. The models are continuously improved, all via reinforcement learning. And as Silicon Labs knows so very well, these IoT devices capturing data at the edge are a critical part of this three-computer physical AI flywheel. They enable the real-time inference. They enable the data collection. They enable the continued evolution of the AI models that you all are building. Just as you need a mining operation, or AI is now the new gold, data is the new gold. Just as you need a mining operation to extract that gold, the same sort of infrastructure is needed for data extraction. In the IoT world, the IoT devices that you guys are building and deploying are those miners. They're capturing critical real-time data from devices and environments while applying machine learning to both filter that valuable data from some of the data that's less valuable. I think Daniel articulated that very, very clearly in terms of not all data is valuable. Now the data collected from these devices flows back to both the edge inference engine, where it's instantly processed and valuable insights are derived, but it's also sent back to the AI factory. That ensures the continuous learning and updating of the AI models. All of you in the room today and your organizations are absolutely critical to developing the components, the systems, and the complete solutions needed to enable this three-computer AI flywheel for development, simulation, deployment, and edge inference. I'm a little bit biased, but I don't think there is a single opportunity for AI that's more significant than what we can see in the healthcare and life sciences area. In digital medicine, AI is bringing robotic capabilities to all medical devices. It's providing clinicians with copilots from surgical navigation to real-time clinical decision support. Advanced sensors throughout the hospital are helping calibrate and locate these medical devices wherever they may be. In digital biology, AI is transforming the biopharma industry by accelerating the development of new compounds. This ends up reducing the cost and accelerating the time to market by over 35% of new drugs. In digital health, AI agents provide personalized insights based on historical and real-time data gathered from wearable medical devices, as well as a variety of other data sorts. It's crazy to think that 30% of the world's data today is generated by healthcare. That's growing at a 36% annual rate, more than any other industry, more than manufacturing, more than financial, more than telecom, even more than automotive. As these connected healthcare devices become smaller, lower power, and more connected, the sheer amount of data that we are generating from the 8,000,000,000 people on Earth is enormous. I think Matt mentioned we're headed up to 10,000,000 by 2050. If you look at this data, combine it with AI, it truly offers the potential to make healthcare more personalized, more efficient, and much more predictive. Now kind of taking a little bit more of a futuristic view. This isn't a long-term future. This is something that we are going to see, and we're seeing it develop today. I don't think there's a single innovation that's going to happen in healthcare and life sciences more than the development of personalized human digital twins. This isn't just a personalized twin of the human. This is a dynamic, real-time digital representation of an individual's biological, physiological, and potentially even cognitive state. These twins are powered from data streams from physiological sensors, many of which are IoT devices, wearable wireless devices. They're going to capture metrics like heart rate, ECG, glucose levels, physical activity. I mean, as fast as the sensor world can innovate, you're going to continue getting more sensors. As these connected health technology continues to move, the quality and breadth of the data captured is going to continue to increase. This continuous flow of real-time data ensures that human digital twins reflect a person's current health status, which now gives us a live digital model that can be used for simulations, predictions, and highly personalized medical interventions. Highly personalized, personalized medicine is the future. The potential of this human digital twin and personalized medicine is huge. For instance, wearable real-time monitor data allows us to move much more from reactive care to preventive care. Think about a case where a human digital twin could detect subtle shifts in heart rate along with other biomarkers and predict a healthcare event before it happens. How did they do this? By analyzing real-time data in the context of each individual's historical health, each individual personalized medicine. This is now going to help physicians make predictive, better personalized decisions. Now take that to chronic disease management, such as something like diabetes. In a Digital Twin, you can now adjust a patient's blood sugar in real time, adjust their insulin dosage based on that blood sugar data. Is this from a general model? This model is personalized for that individual. This personalized adjustment is far more effective than relying on any general treatment guidelines. It leads to much better patient outcomes and, in general, improved quality of life. The last area I want to hit on in terms of human digital twins is talking about clinical research. There are huge opportunities in clinical research. These Digital Twins can now be used to simulate how a patient would respond to treatments. This is key for drug development in terms of enabling much more targeted therapies, personalized therapies for each individual. However, I think Matt and Daniel both highlighted this. There are challenges. First, data security, privacy, and ethical use of all of this sensitive healthcare data. A human digital twin, as they evolve, we must ensure that the AI-driven insights and these models not only deliver accurate insights, but also safeguard patient trust. We're entering a future when the human digital twins being developed with data derived from wearable IoT devices is revolutionizing healthcare by enabling the idea of truly personalized medicine, something we've been talking about for a long time. But the future is now. We are making this happen. This shift is going to change the way we deliver care from both disease prevention to now real-time treatment. That's going to make healthcare much more precise, much more efficient, and much more effective for us all. So let's take a look at a couple of real-life opportunities of AI in healthcare. First, I want to talk about Artisight. Artisight has a smart hospital platform. They built future-ready infrastructure that automates tasks, processes, and simplifies communication. It enables healthcare providers to streamline their clinical workflows and elevate care at the bedside. Artisight's platform takes all of the data that sits statically in healthcare information systems, electronic medical records, and gives it context by combining it with real-time IoT device data from cameras, microphones, medical devices, wearable IoT devices. This combination of computer vision, a multi-sensor IoT device network, and industry-leading and defining AI creates an ambient intelligence platform that learns and responds to your exact environment and workflow. Personalized. The Artisight solution highlights the need for this three-computer architecture. It takes data from both static and real-time data sets. And though it's not actually moving to a digital twin of a human or a patient, you can see how Artisight's technology is now paving the way for digital twins in other areas, creating a digital twin of an operating room, creating a digital twin of a patient room, creating a digital twin of a clinic. All of this helps optimize workflows and improves hospital operations. So from the operating room to the patient room to the clinic, Artisight smart hospital platform is just a great example of how AI can start using this three-computer architecture to drive efficiencies, save time, and ultimately provide better outcomes in a healthcare setting. And now one of the more exciting companies that's really starting to show the capability of this human digital twin. We're making great progress with a company called Hippocratic AI. Hippocratic AI is a pioneer in building healthcare-specific large language models and combining that by creating AI-powered agents that leverage patient-specific data to assist in a variety of clinical tasks, patient-specific, personalized. What they end up doing is taking multimodal AI, and they integrate all these different types of data: text, video, sensors, inputs, medical device data, and they combine it to create a cohesive understanding of the patient conditions as well as the hospital operations. These agents can make real-time recommendations based on current patient data, which helps ensure that the medical staff is following the absolute best practices and protocols while reducing their overall burden of documentation. Specifically, as this relates to human digital twins, you can see Hippocratic's AI technology ensures that the massive amount of real-time data feeding into this human digital twin is not only processed, but interpreted in a way that is actionable for healthcare providers. Let's take a look and see Hippocratic's AI solution in action. Hi, Sarah. I'm Rachel, your AI care manager. How are you doing after your appendectomy? You know, I'm good. I'm glad to be home with my dog, Dennis. It's great to hear that you're back home with Dennis. I noticed that Dr. Chang has prescribed you some antibiotics, specifically ciprofloxacin and metronidazole, due to your ruptured appendix. Were you able to get these medications? I was. I did forget to mention to the doctor that I am allergic to penicillin. Is it still okay to take the medications? Great question. The antibiotics you've been prescribed, ciprofloxacin and metronidazole, don't contain penicillin. So it's perfectly safe for you to take them. Can you begin taking them today? Yeah, I certainly can. Do you have any tips for faster recovery? Given that you have diabetes, keeping your blood sugar levels in check will be beneficial for your wound healing. Your hemoglobin A1C was at a healthy level of 5.5, but your glucometer readings from yesterday were a bit on the high side, so keep taking your metformin and try to steer clear of foods high in carbs. How does that sound to you? I think I'll be able to avoid the carbs for the week. You know, I need to go right now, but it was great talking to you. I'm glad we had this talk. I'll pass along your progress to Dr. Chang. Goodbye. Goodbye. Always fun to see that and know that that type of technology is only capable with all of you in the room. We need to be connecting all of these IoT sensors, IoT device data to be able to continually make that AI agent smarter and more intelligent. So a new era of computing is upon us. One that I think we all agree gives us unprecedented access to both data and compute. Every single industry, every single company in this room is going to benefit from this next industrial revolution. But the impacts to healthcare and life sciences, I believe, will be one they're going to let us all benefit. Every one of us is going to benefit by improved access to care. We're going to lower the cost of care, and we're all going to have better clinical outcomes. The future of personalized medicine has never been brighter. We've been talking about it for a long time, but the future is now. This idea of a human digital twin, it's not just an idea. It's something that is becoming a reality. So with that, I'd like to thank Matt, Daniel, and the entire Silicon Labs team for the opportunity to speak with you all today. Together, we are powering the next industrial revolution. We're making devices smart. We're leveraging the power of data, and we're moving AI inference to the edge. And our work together is going to change the way we live, the way we work, and most importantly, the way we manage our health. So thank you so much for having me, and have a great Works With Conference. Thanks, David. You know, it struck me as I was watching this talk, it took NVIDIA technology to make generative AI, Gen AI, possible at all. And I don't see a way to make the IoT smart without a similar partnership. You know, the generative AI was mapping the internet to data center compute. The next wave is going to be mapping data center compute with all the data collected in the real world. So thank you for that. So these emerging workloads like AI are demonstrating the value of the IoT and can be an integral part of solving problems and making advanced applications a reality. But AI isn't the only application where IoT can play a role. And let's wind things back a bit. The IoT, in many cases, or maybe even most cases, is the last mile to your customers and to that very critical data. In fact, adding wireless to these products at all doesn't just open a pipe for data harvesting. It makes the product better. The number of use cases you can contemplate goes up an order of magnitude from what you can with an air gap device altogether. And all over the world, simply connecting devices together and to the internet is transformative on itself. And this was the basic thinking behind our Series 2 platform. Identify areas where connectivity is needed and connect them. Pretty simple. But as you all know, sometimes it's not quite that simple. Once you add that wireless connectivity, you have a whole new set of constraints and care about to think about. You have power consumption, bandwidth, standards evolution, keeping up with the application, protecting your IP and data from malicious actors. But the benefits are worth it. For example, agriculture. This industry is being transformed by technology. In the 1950s, 1960s, 1970s, we were all worried about, can we feed ourselves, and they grew better plants. How do we do that going forward? Well, stop by the John Deere booth at the next CES and see what I mean. Here in Texas, we have the U.S.'s largest working ranch, King Ranch, which covers 825,000 acres, but that really pales in comparison to some of the other places. Take, for example, the Mudanjiang City Mega Farm in China, which has 22,000,000 acres and over 100,000 heads of cattle. At this scale and these operations, the only way to make it work is through automation, through sensing. The IoT offers a range of solutions for this, from irrigation systems, monitoring irrigation systems, tracking livestock movement and health, securing perimeters, and more. Servicing the number of applications in an environment like this with protocols, distances, environmental concerns, security, it's a hallmark of what we do. It's not just a lab scenario. It's not just a limited number of devices. It's at scale in the field for a long time. And we go to extreme lengths to make sure that we are thinking through that with you. Now, to make that real, I have something else to share. One of our Silicon Labs field applications engineers, Mark Yumina, took that spirit to the real world in testing our performance of Z-Wave Long Range. We wanted to show it's not only cheaper and open source compared to LoRa, it's also better performing. And I have a short video to show what I mean. I wish I could have ridden that bike. But to be fair, the lab is a sacred space for me, so I might be the guy who's happy to sit in the lab and turn the dials. But that video just, I think it shows a lot about what we do. It really brings the lab aspect to Silicon Labs. And it reminded me of The Wire, the opening. How many people have seen that show? And so you guys know that wasn't a commercial. I mean, we're performing those tests real time as we were filming. And the data streams you saw are real. The locations are real. This is the kind of stuff we do. Now, sometimes we need to know what matters to you so that we can go do that. But we are happy to go show these real-world use cases because that's what matters. It's not just the lab. And so to make that real, here's a heat map. This is the real data that we captured. You can see the distance between the nodes, the hotspots of strong and weaker connectivity, the obstacles that drove changes in that. And we fold this back into the next version of whatever we're building. So this goes way beyond checking the box of protocol and FCC certification. It's these kind of capabilities, you know, that we bring to you when we engage. Take Matter, for example. Here at Silicon Labs, we are all in on Matter. We're a founding member of the alliance. We're the number one Silicon Code contributor to the standard. And we're bringing all that expertise into the standard every single day. We want to make the standard better so that we can all win for embedded devices. And we're balancing things that matter to us, like code size, power consumption, security, longevity. And that's the kind of thinking that wins over time. You have a lot to do. You have to develop your product, deploy it, manage it in the field. Maybe 10 or 20 years. I was with a customer last night. These are 20-year kind of deployments. We put all that thinking to work when we were developing our MG26 and BG26 SoCs. We took the last generation and we grew it in every way: RAM, flash, flash, RAM, GPIOs, and more. And it also brings the same dedicated matrix math processor that we're talking about for machine learning inference so that you can help offload the MCU from those very time-consuming, powerful tasks. So what comes next? Last year, we told you a little bit about Series 3. This year, I have something to share. I don't know if they can see this. This is actually our first Series 3 silicon. There it is on the board. I don't want to brag too much, but I will for a second. It's a badass product. It's sampling now. And it brings a lot of breakthrough innovations that you're going to see in Series 3. It's in connectivity. It's compute architecture and security. You know, we're not just a one-trick pony just because that's really, you know, key to the value prop that we bring. This chip, it has the world's most flexible IoT modem. It allows you to live simultaneously on up to three wireless networks with fast channel switching and seamless detection of the right packet at the right time. We're talking microsecond-level switching between these things. It has the world's most secure memory interface here. In many cases, external memory interfaces can be the weakest link in securing that application and the IP. It's the first one people go after when they're trying to break these things once they get physical access. So it's important that you kind of protect your IP there. Also, crucially, it has the best scalable memory architecture in the market. We're not long from operating systems and embedded applications being the norm, not the exception. You can call it RTOS, OS. You know, there's a bunch of flavors that are out there. But bottom line, these applications are getting advanced enough. It's impossible to do the development on bare metal and keep it running. The Cloud doesn't want to interface with bare metal anyways. It wants to talk OS to OS, and it always has. This memory architecture allows you to update and upgrade your code base over time. It's expandable. You're not locked into one kind of memory anymore. And we have encrypted it. We have all the accelerators, the caching that you would expect on any kind of expandable memory architecture. And I believe it's going to be foundational for getting the IoT to where it should be, which is running authenticated code, just like every other major compute platform out there. You cannot run unauthenticated code on PCs and mobile and in the data center and in a car. IoT is next. Once you break the air gap, you have to get it authenticated. So this chip brings all of that. We utilize the latest 22-nanometer fab technology to keep your power down, your cost down, and the size of what you expect. And it's these kind of innovations we're bringing to Series 3. In each technology pillar, we're going to bring something to the table. In connectivity, Series 3 is going to be able to connect just about anything. We'll bring dozens of products to market across all major protocols and frequency bands: Bluetooth, Wi-Fi, Thread, Zigbee, Wi-SUN, more. In compute, Series 3 devices are all multi-core, upgraded Arm processors, and dedicated code processors for the application, the radio, the security, and the machine learning subsystems that you're relying on inside of those products, and we know you need to get a lot done, and we built the compute engines to segment and separate those tasks in the most energy-efficient way possible. It can be very hard to orchestrate all these processors running together, and we think we've got the best way to do that, and we're also very aware that not all products are going to use machine learning. We've talked about it a lot today, but it's a stretch. It really is to say that a contact sensor or maybe a light switch are going to be running the latest and greatest machine learning inference engines. But a ton of applications will: cameras, occupancy, anomaly detection, health monitoring. We just saw the future from NVIDIA. For both classes in Series 3, we're going to have the right machine learning compute capabilities so that you don't break your budget on power and you don't wait around forever for these engines to resolve, and of course, in security, all Series 3 devices are going to include our Secure Vault High. We're going to take that to the next level with more advanced cryptographic engines, more advanced differential fault analysis, authenticated execute in place for on-the-fly decryption of that code that's sitting on that flash outside, and we're going to bring this in. We've got to start talking about post-quantum eventually. We're going to be bringing post-quantum ciphers into the portfolio. So to put it plainly, Series 3 is the future of the IoT. We're really excited about telling you more. We're going to have some dedicated Series 3 round tables today. Please bring your questions and your feedback. We heard you two years ago when we did this, and we worked that in, believe it or not. It's these kind of products that are going to make the IoT smart. Yes, we will have AI factories in the cloud, but they need to be interfacing with real products in the field that are very capable engines on their own. This is where it's been going the entire time. First, we had processors woven through products. Your washing machine used to be discrete. Now there's processors inside of it. Then we connected them. Then we secure them. And now we're making them smart. So next, I'd like to invite our platinum sponsor, Arrow's Aiden Mitchell, to share a few words with us. Aiden? Good morning, everybody. I'm Aiden. I lead product at Arrow. And I think what we saw today was Silicon Labs really demonstrate that you've got a wireless leader that's truly devoted to the software experience. If you think about everything they're doing to affect the protocols, but also the understanding of the systems and applications that each of you are doing, they're deeply invested to make sure wireless technology works where it's deployed. And I've also seen from each of you that you are great product companies that are moving fast in creating applications that are breakthrough and that's unlocking now a services economy both for you and your end customers. So what's our role in the middle of that? We're right between Silicon Labs and you. We're in the middle, and it's our role to help get you to market. But once you go to market, scale. How do we do that? Arrow, we're deeply devoted to the engineering know-how in what's becoming a very complex AIoT environment. We've got over 1,000 field application engineers and now 4,000 designers and developers under our eInfochips brand that's there to work with you and your R&D teams to integrate the mechanical, physical, and digital environments to make sure that complexity around the sensory network, the wireless technology, the intelligence at the edge, and what you're doing to get to market that we can augment your efforts. We are designing, developing, creating, and integrating the software with you. It's all your IP, but we're helping you do the work so that you can get to market on time, even right down to pre-certification and how we can test your products in the right chambers and the environment so that you can get through certification reliably. Our engineers are all around the world, and we bring them to market through this roadshow, Works With, because we really do value the great breakthroughs you're doing. And when you get to market and you need to scale through a complex supply chain, especially one that's becoming highly diverse and where it's really changing all the time, diversifying beyond a China to a Vietnam, a Malaysia, a Guadalajara, wherever you are building, making sure that we're getting the product you need, segregating that inventory for you when required, moving it to the factory that you provide preference on, and doing that through a range of models to help either manage it on your behalf when you're owning the working capital or when you need us to own the working capital and the financing, creating models to do that for you too. So it's our delight to be able to engage with you. We've got teams from eInfoChips here today. They are showing the intersection of AIoT between our partners, NVIDIA and Silicon Labs. We have an autonomous robot that's here that's showing what a true autonomous AMR robot can do in a connected framework with Silicon Labs, the power of where Wi-Fi is going to take that autonomous workloads inside environments like this, and then we also have our supply chain resources here to talk about how we can help scale, move product around the world on your behalf, and ensure there is no disruption in supply, so with that, I really want to thank the Silicon Labs team and perhaps bring Matt back on the stage to wish you all a really productive afternoon. Thank you all, so in closing, like everyone, we do a lot of events every year. But Works With is the one that we look forward to the most, not just because of our time together with our customers or partners or the work that goes into it, but also because over the last five years, it's become a milestone to reflect on how much has happened and how much is to come in the future in our space. Think of the realization of Series 3 and what that means for developers and device makers that Daniel shared. Think how far Matter has come in the last few years trying to tie the industry together to improve developer experiences and consumer experiences. Or think about everything we shared today around artificial intelligence and how it's going to accelerate the IoT and really how it's going to change the world. These are all incredibly exciting. Our commitment to you, our customers and partners couldn't be stronger. IoT connectivity is what we do. Thank you for your trust in us, and like we like to say here in Austin, Texas, y'all have a great time. Thank you again. Appreciate it.