Ambarella, Inc. (AMBA) Earnings Call Transcript & Summary

Well, good afternoon, and welcome from all of us at Ambarella to our CES Briefing and Technology discussion. I hope your first day of CES was productive. I'm sure for many of you online, in the room, one of your New Year's resolutions was to learn about at Edge AI and how Ambarella fits into it. So we're definitely going to help you check that box, so you can move on to your other new years solutions. But some of you visited us earlier in the day. We have in the adjacent rooms, more than 30 demonstrations of Edge AI at work. And if you're not registered to visit us later in the week, you're welcome to contact me afterwards and we'll fit you into one of the groups. Also for the online participants, we'll need some time to prepare a virtual CES experience, but we should be running that in March, and you can also send me an e-mail if you'd like to participate in one of those events. So Ambarella has been, I thought -- I was going to say 10-plus years, and I was just talking to Fermi. Ambarella has been at CES at this location in a much smaller footprint initially for almost 20 years now. So this is a very important event for us to share with customers and partners, our new products and technology and to talk about our Edge AI platform is evolving. And that's what we're going to achieve here. But we felt that in addition to all the demonstrations that we can provide to you, given all the developments in the AI market that providing some additional perspective would be very helpful in terms of what's happening in the market, our view of it and our strategy to tackle this space. So that's the purpose of the meeting today. Before we proceed, I do need to read some forward-looking statements. I'll make it very short. I apologize. I wanted to get through it quickly, but I do need to read this part. Today's discussion will contain forward-looking statements that are based on currently available information and subject to risks, uncertainties and assumptions. Should any of these risks or uncertainties materialize or should our assumptions prove to be incorrect, our actual results could differ materially from the forward-looking statements. These risks, uncertainties and assumptions as well as other information on potential risk factors that could affect our business are more fully described in the documents we file with the SEC. So we got through that. In terms of the agenda, we're going to run with management presentations for about 75 minutes. We're going to try to save 15 minutes for Q&A. One thing that's unique about this presentation is that you'll be hearing for the first time from 4 Ambarella executives that have not been accessible to the investment community before. So you'll hear them present today. But also after this event, will be hosting a reception immediately outside this room with some refreshments and I think some snacks. And you'll have a chance to meet with some of these executives as well as some of the partners in the room that we'll introduce later in the presentation. As a reminder, this is not a financial event. So Ambarella's fiscal year, fiscal '26 ends in 3 weeks. And in our earnings call at the end of February, we'll be providing financial updates at that time. So with that, let me turn it over to Dr. Fermi Wang, Ambarella's Co-Founder, President and CEO, to go through the rest of the deck.

Thank you, Louis, and good afternoon, and thank you for coming to this important event for Ambarella. As you know that CES is always important event for Ambarella to show our new technology products. And this year, with the fast evolution of AI technology market, we feel critically important for us to provide a deeper insight into our Edge AI strategy platform road map as well as demonstrate the capabilities of our new products. Compared to the AI data centers, AI at Edge serve different purposes and calling for a fundamentally different silicon architecture. And on top of that, the AGM market and products needs unique requirements, like, one, power consumption is most critical; two, low latency and the privacy or must haves; three, much less bandwidth is available, not including the communication bandwidth as well as DRAM bandwidth; four, this market is supported not only by the business CapEx, but also the consumer spendings. With all of the silicon hardware and sulfate optimization needed to achieve those market requirements as well as drive meaningful volumes at Edge, we believe that this Edge AI market still at the very early stage of commercial development. And this is the opportunity for Ambarella. And this is where we are focusing on today. And at Ambarella, we are very proud that we offer a very comprehensive Edge AI platform, featuring both AI silicon and the software designed to support our customers not only for the design for the customer not only for their -- to meet the market requirements, but as well as help them to scale their business. And we already shipped that we will ship more than 50 -- sorry, 40 million units SoC across different application is the best proof for that to that statement. And today, we are happy to show you that we're going to expand this platform in a way that we have never done before. and also helping us to tap into new opportunities. As you know, Ambarella is recognized as the best high-quality video perception system at age, original design for human viewing. And later on, this proprietary video perception system, which get a significant amount of data from various sources fit into our customer cost and AI accelerators. And this setup helps machines to operate autonomously, sometime partially some time fully on 1 single AI SoC. And this evolution between the video perception system as well as AI technology really help umbrella define multiple generation product cycles already. In 2004, and we found the umbrella with one simple idea that the digital video content will become popular. And consequently, that we develop our first-generation video perception system. And in 2012, with all the AI paper, research paper published such as AlexNet, we came to the conclusion that the CN type new network will become the foundation of AGI, and we develop our CV2 family chip to seize that opportunity. Today, 80% of total revenue came from CV2 family chips. In 2017, the transformer-type new network is published, which led to the development and creation, our third-generation architecture, which support both and transformer type of network. And with that, we create our CV3 family of chips, of course, based on our third-generation architecture, which really help us to address new applications both new physical AI application like aerial drones, auto driving and other robotic locations. So that was the past history, how we get here. Today, we're going to show you how we're going to improve and improve the product offering in 5 different ways. First, we're going to really provide the update how we're going to improve our Edge AI strategy and platform. Two, we're going to announce our first 4-nanometer AI SoC being sampling to our customers. Three, we confirm that we on time tape out our first 2 nano meter chip to Samsung foundries. Fourth, we're going to discuss a brand new go-to-market strategy that will improve and increase our revenue generation. And fifth, describe the significant evolution our Cooper software development platform that will help us address the market need at the edge infrastructure market as well as the endpoint market. I think many of you have seen this slide at the bottom of the slide, the last row of the chip is our video processor, which is average ASP of single-digit dollars. On the second row, which is our second-generation AI silicon and UC CV2 family like I talked about, the average ASP is ranging from $10 to $75. And third generation is on the third line and the average ASP is from $200 to $400. And today, we'll add 3 new members into this family. First one, as I said, we are sampling our first 4-nanometer chip we call CV7. 48 hours that we received the sample from Samsung Foundry, we successfully brought up of 2 demos in our lab. One is 4K P60 video and AI processing for video streams. Second demo is 4 channels of video input at 4K P30, both for video processing AI. Both of this configuration has a performance level that nobody else can reach in the market. And those 2 configuration is going to be critically important for many different applications for our customers. And we can achieve this kind of engineering the deals within 48 hours of receiving the chip, show 2 things. One, Ambarella continued our tradition of excellent engineering execution. But also throughout years, our silicon architecture and silicon road map plus our software platform becomes so mature that we can easy to tape out a chip and even though it's a complicated chip, that we can deliver those kind of demo in real time. We expect CV7 will be in production at the end of the year into certain different applications. Second family we add to this page is our first 2-nanometer chip. We talked about this before. This is a chip going to be -- a tape out to Samsung foundries. And this is a -- the first customer is really helping us to pay for this chip is a semi-custom design for letter application that we haven't talked about, but we're definitely going to provide you more updates later on. And this chip will be used by our first customer into production in first half 2027. And the third thing I want to talk about today is in the future -- in the past, Ambarella focus on edge endpoints. But just a few quarters ago, we announced our 1 family chip to address each infrastructure. We talked about that we had our first design win 2 quarters ago and the first design win were turning to mass production in first half of this year. And on top of that, Muneyb going to show you, talk about how we're going to evolve our go-to-market strategy for this age infrastructure market by including our GSIs and ISV partners to help us to expand the software offering to customers quickly. And this is a new strategy, go-to-market strategy going to implant for not only for the aged infrastructure, but also for age end points. And I think this is important for us because in the history of Ambarella, most of our revenue is still driven by direct sales. And this channel sales, I think, will be beneficial for us by increased our potential revenue creations I think throughout the development and customer engagement of CV2 and CV3 family cycles. One thing become very clear that all of our customers, all of them telling us that they need more and more AI performance while they don't allow us to increase the power consumption numbers. So first of all, let's talk about how we -- what kind of workload will trigger this kind of AI performance at edge. First of all, the CP workflow be comparable AI processing. And this fundamental change really pushes us to implement our silicon architecture in a way so that we can really have to increase our AI performance and horsepower in a big way. The second thing I want to point out is due to the need of low latency and privacy at edge we have to move some of the workloads, particularly for the application like physical AI, drones, driving car. You have to move those workloads from the data center to the edge. It's not just limited by the physical location. There are many, many more application networks start seeing our customers, enterprise customers, moving their AI workload from the cloud to the edge. The third thing is probably one of the most important in the explosion of the transformer type of the workload, particularly for GenAI and the vision language model being widely used by our customers today. And they are trying to figure how to implement those new models into their products, and we are helping them to do that. And throughout the process, in the past, at the C level, we're talking about maybe a few hundred K parameters today, we're talking about $2 billion parameter as a minimal running on our 2 out chip. That just shows you how much more AI performance we need to add to our silicon to address that need. And in addition to that, all our customer have said we need to attach more and more channels of sensors into our silicon, not only just camera sensors, but radar sense all the other type sensors. All those new channels of sensors requires new AI performance. And to make it worse is in the past, we are handling 2 megapixel cameras of the input. Today, we are addressing 8K P30 camera as a video input and the data among the data that comes into our chip increased significantly over the years. And because of the 5 reasons, you can see that our new products continue to add more AI performance, therefore, bigger die size, therefore, bigger, better ASP for us to address our customer needs. One thing I can say is, for example, in Q4 last year, the average corporate ASP is $15. Today, every chip that we are developing or sampling right now have average -- higher average ASP than $15. That just show you how we're going to capture more value per AI unit. In the last several quarters, we start introducing brand-new AI applications, take -- helping our customers to take them into mass productions. And we get asked many times by our investors what's the additional R&D cost to enable 1 extra new applications with our platforms. Today, I want to address this question by showing you the flexibility, programmability of our silicon and our software architecture. On the left-hand side of the picture shows this is a software stack and some of them done by us, some of them done by our customers. And MunibMuneyb going to go through the detail of that. I just want to point out the core of the software stack is our corporate development platform. which is identical for every product that we ship in the market. And let me allow me to use the right-hand side, the CV5 as an example. In the last 2 years, we used CV5, our first 5-nanometer chip, take them into 7 different type of applications from automotive to enterprise security, drones, portable video cameras, perception system for robots and so on. see totally different applications. But if you open up those products, you will found that a bottom of that product is CV5 silicon plus a corporate development platform, identical -- the only difference on top -- for those 7 applications is the application level software that our FAE help our customer to develop. So the extra cost to enable 1 new application is really that small FY '18 dedicated for each customer for their application. So you can see that the extra cost to -- for us to enable new application is limited -- and therefore, we can start showing significant ROI when we increase the number of new locations. And that's why we never fear to talk about new locations because the ROI from our investment. But I also want to point out -- this benefit of programmability and flexibility applies not only to Ambarella, but to our customers. For example, any customer who use CV5 to implement one application. If they want to go to lower end products, they can port their application level offer easily to achieve like CV75r for the low-end product porting to CV7, our new 4-nanometer product because they are sharing the sen silicon architecture and the Cooper development platform. So our customer by using our system they can significantly reduce the R&D dollars and improve the ROI also -- and that's why -- and the reason I mentioned the previous slides is because we start seeing many, many new AI use cases. 5 years ago, when we get together, the most significant application for Edge AI for us was enterprise security. But today, all of the boxes I show here, we have either engagement or design wins in those spaces. Let me use the last year as an example. Fiscal year '26, we not -- while our enterprise security market continued to grow in a very healthy way. The growth, our revenue growth last year also country by 2 new markets. One is a portable video market, the otherwise telematic market that we have not -- we never addressed in the past for the HI applications. And with this new application, it really helped us to significantly increase our revenue last year. And in addition to the 2 new applications, we think there are 3 large opportunity for us, in front of us that we can tap into in a short period of time. One is robot. In this particular market, we focus on aerial drones because this has a huge volume opportunity for us. We also focus on, for example, MRs or manufacturing automation, human noise. They are all important markets in the future, but I think all of us still at early stage of the revenue generation. And also, we think the edge infrastructure like we talk about is important. We have first design win. We're working on more design wins. We think that potential is huge moving forward. Of course, the third one is autonomous driving that we have been working for a few years, and we continue to commit to this market and working with OEMs and Tier 1s for the -- to secure our first major OEM designs. And beyond those 3 big opportunities, there are many green shoes here. All of them has shown some potentials, but it will take time for them to develop their revenue generation. But I will point out that all of these new opportunities that we put out here -- all of them are using Cooper development platform and leverage an silicon architecture so that to enable each one of them won't take a lot of new investment on the R&D side. With all the discussion so far, I think it should not be a surprise to you that fiscal year '26 was our record revenue year in the Ambarella history. Here, we show you our revenue for the last few years and the blue bar is our -- the revenue generated from our traditional video-only product line. And the green bar is the revenue generated by our -- if you look at the overall -- our revenue performance, we generated 12% CAGR in those period of time. But if you discount, if you remove those company revenue generated by the company who impacted by the entity list. For example, in fiscal year '20, HiVision Dahua, DJI was 45% of our total revenue. And they are almost 0 today because the entity list. If you remove that, our company revenue CAGR was 18%. And -- but that's -- there are nothing compared to our Edge AI revenue CAGR at the same period of time, which is [ 60% ]. So I think the only conclusion you can draw from here is the transition from a human viewing company for Ambarella into a machine perception, plus autonomous decision-making company is complete. 80% of our revenue come from our market -- our Edge AI solutions. That also shows the most important IP in this company today is our AI accelerators, which is really well defined and has better performance per watt than all of our competitors out there. And we believe that this is going to be the focus for us continue to leverage our traditional on the perception side but focus more and more on the AI accelerator to make sure that we can continue to deliver the performance per watt for our customer in the future. At the end, I want to wrap up my presentation to make one statement, which is Ambarella is a leader of Edge AI market. I don't think that's an empty statement because there are support -- the statement is supported by the following numbers. We shipped more than 40 million Edge AI SoCs so far, cumulatively. And we're putting $1.3 billion of investment, R&D investment in the last 10 years into this product line. We helped 370 unique AI customer projects in production and also report over 200 unique AI model architecture, not AI model architectures for our customers. So all of this I challenge anybody outside who trying to come to AI to provide a similar matrix and provide the comparison. And with those numbers, I'm proud to say that Ambarella is the leader of AI. So with that, I would like to introduce Muneyb to give you a more deep presentation into our strategy on the marketing and customer engagement. Munib joined us as the executive team of the customer growth office 6 months ago. The first time we met, we came to -- quickly came to a conclusion that we both believe that the HCM market is at early stage and worth our time and effort to make it bigger. And because Muneyb was -- spent his first 20 years building data centers and cloud products at a company like VMware. And he joined us from Intel as a Chief Marketing Officer for their network and the Edge IoT business. With that, please welcome Munib.

Thank you, Fermi Thanks, Fermi, and thanks, folks, for taking your time and coming out here. I appreciate your spending here. And folks online. Hopefully, you can watch and follow us. So I have a longer agenda here, but let me kind of get in and thanks for -- as Fermi said, when we first met, we talked about the opportunity at edge about to explode. And I've spent 25, 30 years of my life just working in the data center and cloud and arrive at that conclusion that edge market is going to stand up and proceed just like data center and cloud was 3, 4 decades ago, right? Because I think -- in the data center market, initially '90s and 2000s, it was about transforming companies. There was data automation and people who are building clients or architectures. And I see a lot of investors in the room top companies in the market, in the top where system providers went from silicon CPU as the market cap to system providers like the Ciscos and the world then came the cloud native market started moving towards the architecture was very much more mobile apps and developers building this. Of course, market caps of SaaS companies and cloud companies went to the roof. Where we are looking at, we believe, will be where Edge will stand up, and you're actually now transforming not people now you're transforming things because everything is going to talk to each other to the AI market. And we believe that is going to explode and at the same level. It's not one against the other. There will always be a continuum, right? So there will be a continuum where there's still 30% of the workloads in the data center, 40% in the cloud, but new workloads emerging at the edge will be 30%. That mix is about to happen over the next decade, and that's the opportunity we're talking about, right? And how do you go capitalize on that, and that's a huge opportunity. But let's kind of double-click into that because the segmentation is not clear. As you have data center folks moving towards the edge, they're building edge servers, which are in the scale of servers and platforms and racks. Whereas Ambarella, as you can imagine, IoT endpoints, we're seeing adjacent markets come in 2 ways. One, physical AI, where you're taking multiple sensors, vision, audio, infrared, radar and putting together cars or robots, which multi-sensors put together on one end. On the other end, up the network, we're seeing more and more kind of sensors and inputs to be processed, but not at a server level, more at an appliance level. Now add AI to the mix that starts getting interesting because AI processing and model sizes vary. If you're in the cloud and data center, now you're looking at hundreds of tillions of parameters and models. As you move those towards the edge and you have big edge AI inferencing solutions and chip providers who look like wafer size chips, we're not talking about that, right? So a big, huge still a AI inferencing market, but then you're looking at more simple, smaller. So these are more parameter sizes of $500 million to $50 million or in the multimodal language, it's about 50 billion parameter model that you're going to do with. So that kind of sizing is where we think that edge AI market comes up. And that -- but designing for edge, going on what Fermi said,it's a very unique architecture. You can't just take data center silicon software and move it very quickly to the edge. You have to spend a lot of different cycles because privacy and security very important. You can't stick a huge load balancer in front of an edge endpoint right? So that's not that easy, real-time processing. You can't deal with 30, 50 milliseconds of delay in the cloud and data center, you have to be really real time because you do it real with real-time things. Reduced network is not like a land with a backplane of 10 gig you have to deal with really hard network at the edge and low power consumption. I don't have massive power facilities like data centers, it has to be embedded low-power and I think these are the synergies kind of help Ambarella position ourselves because we're designed for all these elements and capitalize on this, right? So now what does that do for opportunity? Now I'm making a hypothesis that workloads are going to move to the edge, and they're going to run on these different devices rather than in the big data centers. Therefore, that infrastructure build-out is going to look from first our TAM, and we always talk about our current SAM. But our TAM in '27 is about $12 billion. But as you look at this new edge infrastructure market that's expanding, hardware, software services, that's a big pie. And our silicon can scale and capture some of those early wins in the network and gateways, which are already seeing the design wins we've shared about, but that's a potentially large TAM that we are able to tap into as we go after this market space, right? So that's kind of a big opportunity for us to kind of scale and products that you go into the infrastructure market. But you do need a full stack solution. You need silicon system, software applications all coming together to kind of deliver on this. And the good news, as Fermi has been talking about is we have been as Ambarella for the last several years building that stack. We've not come forward and presented that stack as a whole solution. We've been going after vision and different aspects of that. But we have that stack and then just what you'll see in a lot of our demonstrations is as Fermi articulated, we have a full kind of range of silicon first, second, third generation but also designing the systems from endpoints. Now we're starting to see the infrastructure being built in. We do so a lot of time building software. This software supports multiple operating systems from real-time operating systems, robotic operating systems, embedded Linux, QNX, all of these, but a common set of SDK in tool chain. That's super important because now you can write an application for 1 chip and easily move to wherever you want. And that's super critical for applications in the outbound interface in the software. And that ability for us to do that has been tremendous on back of which we stood up Cooper Developer platform, functional safety, ASL kind of very compliant platform and real-time platforms on top of that to support our customers. Now we're also embracing open source because you see open source framework. And this is an important part because we believe open source will take off and agentic layers kind of helps the scaling and automation for those at the edge. So we have can announce today, and we'll get into the details of some of our developers one, but we're embracing an entire can open source framework and publishing, be on a set of algorithms we've already built. So at this point, I probably want to invite our experts on stage from the company. So I want to have a panel discussion with Bob, who's our Chief Architect, come on to Bob and then [ Mohar ], our software principle. And Alberto Broggi, who is our General Manager of his lab to kind of start getting into unpacking that stack because they spend a lot of time working on these and you want to kind of understand how this works.

So thank you, gentlemen. Thanks for coming up. Let me kind of start with Bob and say, Bob, you would have talked about AI acceleration evolution. My first question always comes when I can always on discussing with you is, are you really an AI kind of architect? Or are you an environmental scientist?

Well, I think that's a funny question. Truthfully, I'd rather be a computer architect. But honestly, at the rate that power is scaling for these AI applications, I don't really have a choice. I mean, there's a fundamental scaling problem here. When the AI, there's a lot of exciting things that we can kind of imagine that AI can do and a lot of people are just throwing MAX at it. So a number of parameters and an AI may be doubling every 6 months. But if you look at the actual power and that requires more power to both train and deploy those type of systems. But globally across the world, like our ability to generate enough power doesn't scale at those rates, maybe global power generation despite all our investments in clean and abundant energy is maybe only growing 1% or 2% a year. So fundamentally, those scales don't match. And so the consequence of that is something has to be done. The edge in that sense is inevitable. I think that -- so let's look at from -- I'm a computer architect. So let's look at where power is actually spent. If you look one in of these devices, the amount of energy that it requires to compute is actually like fairly small. What actually is expensive is actually moving the data around. So maybe a Mac might cost like at most like 1 pico joule. But then once you need to move it either in some large kind of rectal limited chip or off to some DRAM memory DRAM or off-chip or some high-speed interconnect network, that can be 10 to 50 pico joules per second. So much larger kind of power just to move data. And so then the natural consequence is, hey, let's get that computation, let's get that intelligence closer to where it's actually perceived, closer to the video of the image sensors, closer to the perception, closer to where we have some compute to generate some local intelligence. That's really critical. And I've been working at Ambarella for a long time. One of the things that's really exciting is we've been building edge products from the very beginning, all back to A1. So we have expertise in image processing and video compression and those are really exciting. And I think based with this energy challenge, one of the things that I'm really passionate about is architecture really matters and looking forward to future chips.

Yes, having been an architect myself, I'll tell you that's always constrained and balanced, like how do I architect something that is wonderful, but then I have all these little limitations. Well, it's reticle size is too big, it's too small. How do you kind of trade off between constraints and balance really?

Yes. I mean computer architecture, a lot of times are -- you talk about constraints. I mean power, we've just talked about is a really important constraint. -- latency is a constraint, compute is constrained. All of those are really interesting to kind of think about. But those are really inputs and constraints alone don't actually make a good chip, actually balance does. One of the things -- when I see Ambarella chips, what I get really excited about from a computer architect point of view is when you look at all the different components that are inside our SoC, the image processing pipeline, the AI, CV flow as an example, DRAM controller and even ARM processing. When all of those are running at full capability and are fully utilized, then you know you have a balanced chip that's quite beautiful and very effective and also attractive from a business point of view because you don't have any kind of slop in that. Inside, if you have an unbalanced chip, you can have a chip that looks very good on the slide. It hits like a benchmark, it hit some peak number. But then when you actually look at real kind of workloads, most of that time, it's not used. You don't have enough of any particular resource to actually use it. And especially when you get into the image processing pipeline or any kind of real-time application. When you overload a constraint, the system fails. If we have to drop a frame, you've lost the shot. So that's critical. Hitting balance is really important. So actually, I think one of the -- an example of a well-balanced chip is CV7. We've looked at that, and we're looking forward to seeing it deployed. So Greg?

Thanks, Bob. And I think the natural question then I'll get this all the time is, hey, you're pulling accelerators or SOCs. It's like there's a lot of folks building accelerators very dedicated for the market, high speed, but the balance and constraint leads directly to -- should we be doing for the edge market, accelerators or SoCs. , what do you think we should be doing, right? Well, we're an SOC company. So obviously, the question is the answer should be SoC, right? But why I understand that, right? So look, once balance is really important, then SOC is really the only answer because when you -- balance is really required, if you don't actually achieve that, then we leave here's the problem. If we don't actually build an SoC, then we leave the balance problem to someone else. And really, that's not where Ambarella provides value, right? We want to be able to balance all of these heterogeneous components. Those components that I've talked about, the image processing pipeline, CV flow, DRAM controller, all of those have different requirements and all of those have to be in balance. So if we just do an AI accelerator, then we're just leaving problems for somebody else to solve. And I think the interesting thing is when you look at these different markets that you identified things like physical AI, AI bought edge boxes, smart camera applications. All of those are going to have slightly different type of balance points. And I think when you look at the number of chips that we've actually designed, maybe 15 or more based on our CV flow generation. All of those represent a very important and different sort of balance point that we've been able to achieve. And we do that from an algorithm-first approach. Basically, we study where does that balance point need to be? What are the critical applications that we need to solve how do we do those effectively and at low power and then build a chip around that. So in an architecture sort of but it's really -- when you see our Ambarella chips that are working properly in the full SoC, it's also very beautiful.

Last question for you. I'll have 1 more after this. But really, are we are compromising between flexibility and scale in here? Or is it a choice because of all the flexibility, am I losing scale or are we able to do the boat, right?

Yes. So I mean, flexibility comes at a cost, and I think that's important to keep in mind. -- a lot of times in these large kind of cloud systems, maybe the flexibility that you get is fairly small. It doesn't even show up on your kind of balance sheet. But then once you build these sort of edge -- smaller edge devices then building general flexibility is not for free is not free. And everything has to be in balance and be cheap in order to make sense from a business perspective. So we -- how do you build flexibility in your system so you can cover a larger range of balance on -- and I think Ambarella, we have an approach of building flexibility into the architecture family, meaning we build CVflow. And then once we have CVflow, we can scale that up or scale that down to hit some specific balance point that we actually need. Now there is some local kind of flexibility. Obviously, we want to -- we're looking at algorithms first, as I mentioned, -- we're looking ahead and seeing the road map of where things are going to go, and we build enough kind of local flexibility, but it comes at a much cheaper cost than it would be to build general flexibility into our chips. An example of this is when transformers and LLMs and ChatGPT networks really kicked off, we're like, "All right, we need to demonstrate this. So we look around at the different chips that we have, we picked N1 that was at a very good balance point. And then with the power of our software stack and the tools that we have available, we were able to go and execute that demonstrate that you could run N1 at LLM inference on N1 very effectively and do that in a short period of time. So that was -- that's sort of an exciting kind of prototyping aspect and then we can use it effectively.

Great. Thank you. So kind of to summarize what you're saying, is Bob is our huge differentiation is not these trade-offs that a lot of other vendors have to do. Our huge differentiation is we can actually do an amazing AR detecture with power consumption, have the constraints, but balance it right and SoCs and accelerators, we can get that blend ride. And finally, the flexibility and scale giving us to almost 400 million chips that we have shipped with that type of balance. And Fermi's slide especially takes B5, go into 7 different application markets, very easily and programmable. I think this is a big differentiation for what Ambarella has. A lot of people trying to understand where that fits, but this is our key differentiation from an architecture perspective. . Thanks, Bob. I kind of move over to you, Mohar. Now I think we also talked about the full stack. Now going from the silicon architecture, we talked a lot about Cooper development platform. But before we get in there, can you tell us like what are the trends we're seeing in system software identical, right? .

Right. So as Fermi mentioned, transformer explosion happen everywhere. So we can see transformers even in our traditional like a task like object detection and everywhere, that's already there. We have a CNN Gen tool, which is quite proven. It has been there for 10 years. And we had this thought that put metric multiplication into our third generation, which is what Bob and our architects and team did. So we were able to quickly bring up the transformer model in our CNN tool set. And that basically enabled us a lot of new applications like all this multi-model application that even the traditional ADAS or all the software being and with the transform or the new AI kind of -- the models came up. The AI is kind of slightly different because they are auto aggressive models. So we came up with the new tool LMG, that's part of the Cooper platform also. So we were able to quickly bring up the VLM Gen, and we can bring up all kinds of models with our PLM software. The N1 example that he gave is basically that's how it is. The other thing that I wanted to actually mention is Bob also mentioned about scale up and scale down of our architecture. With our tool chain, we can actually do it automatically. So it's not -- even though we see a lot of chips, the CNN happens to be one tool that can scale up and down easily, okay? So that's about the current thing that we have the foresight of having this metric in multiplication, which enable all kinds of transformer applications, right? Now if you see moving forward, the inference time scaling is becoming very important now. Now inference time clinging all these reasoning models are coming up, all this chain of thoughts are coming up or all this things that even the agent in workflow, if you see the context size keeps increasing and the amount of data that is being generated keeps increasing. So inference time scaling is a challenge that we are working on. that's what Bob and the architecture team is actually looking forward in our new architecture. So that's what it is. People outside also know and the bottleneck that you guys mentioned earlier about the DRAM bandwidth and all that. There are new techniques to people at have core and a lot of other companies about like 4-bed conization to controlling, all these things, we also keep a watt. And that's basically some of it we can apply today with our VLM. Some of it, we are actually going to -- looking at it to do it more efficiently in the future director. So if you look at -- you also mentioned earlier that we are embracing the open source community. So AI as such is in still nascent format. And new things are coming up every 6 months, you will see new framework. So agent workflow is very popular right now. So going from one LLM, another LLM, defining the memory, define the contact. That whole thing like LLM is an application by itself, right, having all these things to work together. So we are keeping a watch on that one. And if you look at our VLM Gen tools and all that, we are actually -- we are embracing hugging phase, Hugging phase, and actually most of you probably know, hugging phases where almost all models are released nowadays, and we can quickly convert those models from hugging phase to run on our chip. And with our tools being common, we can go from C72 to even C7 demo that you see outside, we already can actually execute some of those things with our tools. So it's a scalable architecture, right? So with the new optimization techniques we can incorporate, thanks to being embracing with the agent to hugging phase. And the last, but not the least, so physically I that for me and you were referring earlier, we cannot ignore the security, the cybersecurity and the safety, right? So there are techniques in the -- that are coming up in the LLM world even to address that. And we also -- I mean, with our automotive, we also have experience with ACL and other stuff that we have also worked on -- that experience helps us in the cybersecurity and all that. So that also is -- we can leverage that also today.

So that's kind of our Cooper dev platform like I know you kind of threw a lot of the tool change, I wanted to kind of in there so you can talk to it, right? Yes.

So let's look at the holistic one, the Cooper development platform, right? If you look at the foundation is our metal is our chip is what Bob and the team has developed. And then we have these core libraries. The core libraries are actually based for the vision, like what we are doing, the streamer libraries and all that we released. And the critical part of it is the tools that I mentioned, CNN Gen tool and the VLM Gen 2. We are also -- today, we announced model garden. So some of these models that are converted and optimized with our tools, we are already publishing on our website on the aging phase as well as on our website, right, so the customers can directly use that optimized model, so they don't have to go to the exercise of optimization. On top of it, we also have a run time environment to run all these models, right? Now one thing I wanted to stress, I think we also mentioned earlier, we have one as Cooper platform that runs across different chips. So once we invest for 1 chip or writing an application, it can easily transform from CV22, let's say, CV75 or let's say, CV7, depending on the balance point that you choose for your application, we quickly can develop have an application. Then we have all other software stacks that are necessary for robotics like Ross and all those things. that also we have developed and those are part of the Cooper development platform. So Cooper development platform is a holistic, which includes all the libraries that are the drivers, which are interacting with the silicon, which has some open-source model on the model garden, on the hugging phase on the cloud. We have the tools, which are going to convert it and then the application example that we have for various markets. So we have one platform that can address different needs as such.

Thanks, Mohar. I think you go to Alberto. Now the opportunity to kind of double-click into -- well, we talked about a horizontal silicon and horizontal stack with Alberto who's been our General Manager in Italy for Vislab, an acquisition, like how are we seeing -- let's go deep into one vertical, like autonomous driving software stack, but go ahead, Alberto, yes?

Yes. Actually, the trends -- the more the trends in autonomous driving are mainly going towards real end-to-end. And that doesn't -- it means that the full pipeline, so the data acquisition, processing, the perception, the fusion, everything is based on AI. And that does not necessarily imply that we're talking about as one big network, which receives pixels and radar echos and delivers just set points of speed and steering because training that big network would require very large data sets -- and when you talk about large data sets, you mean a lot of energy for the training, a lot of time for the training. Large data sets also implies in these data sets. So moving towards end-to-end means having big data, big data sets and our solution is in the network that I was talking about in multiple networks that are just one. So instead of having just one network, you have multiple networks, you can train them more easily because you can use data sets and more focused data sets just for that specific purpose. And the network should be overlapped and sharing some latent space. And by doing that, you're ensuring that you are preserving and you're propagating the features which are actually intrinsic feature, non-explicit features, and you can propagate them throughout the whole stack. So that's the real interesting part for the end-to-end system. So that's what we do. And for example, in this slide here, you can see our current version of our stack, which is based again on multiple networks all connected together that deliver the actual autonomous driving. But actually, the stack itself alone is not enough. So you need more than that. So you need a complete ecosystem around it. So again, the stack itself and the architecture is so important, it's so critical, but you need something more. And because training a network or multiple networks, again, it's required a lot of data sets, diversified data sets. You need to have large different scenarios covered. And you need to have also grown truth connected to those data sets. So again, the more diversified data sets the more scenes are connected are covered. So that means that you're in large your ODD design domain. And if you enlarge the ODD, that means that you are increasing the maturity of your stack. So if you really want to scale up your in to have lots of data sets, ground truth for them and selecting the data sets. And this is what we're doing actually right now if you go to the next step. So this is what we're having. So we have a complete tooling. And the magic here is that all the tooling needs to be automatic. So if you really want to scale up, you do not have to have any human intervention in the pipeline. So starting from the acquisition of the data, we have vehicles that drive around acquire data sets and the driving behavior of the person driving the car. And then you pull this data into our automatic annotation pipeline that attaches the precise ground tooth to these data sets. Then once you have the ground tools, you need to select the data sets. You need to create a right balanced mix of data sets in order to be able to cover the whole ODD. And this is done by another tool that we have in the pipeline, which is this VLM base. This is based on LLM and is able to classify the data sets. So data size whether night day or pedestrians, tracks and so on. And then based on that and the statistics, you can create the right data sets for the training. So once the training is done, you put these networks, multiple networks on your car or on your simulator and then you get statistics about the performance of the system and these statistics are also used for prioritizing the next acquisition campaign because you understand how much data you have of some specific scenarios if you are lacking some scenarios. So these information are also used in the full stack. So this is actually the full tooling, which we have been using in our project in L4 autonomous trucking project we have with [ OMV ], which is formerly Continental that is going to be SOP next year.

No, I think that's great news. So I think it's good to see these things going into production pretty fast. Now taking a tangent with you, a lot of the learnings, so there's a lot of nice about robotics and movement in AMRs at CES this week. How can we take some of these learnings and apply it to robotics?

Yes. Actually, we started with robotics at the very beginning. So we started with vehicle automation for big vehicles like heart moving vehicles and agricultural vehicles. So we started with that. And actually, there are strong commonalities between many fields in the robotics and the autonomous driving that we've been working on. And the basic architecture and principle is the same, is very similar. And the core functions that you have in a robotic system, well, they are core functionally mean the perception, data fusion, planning. These are more or less the same fundamental blocks that you have in a robotic system. Plus, I would also say that maybe some differences could be in the kind of sensor data that you're using. So maybe in robotics systems, you have a strong usage of cameras and maybe other sensors like textile or pressure sensors that you don't have in a module course, maybe some difference in the precision of the planning that you have to do. But despite these differences, I think that, again, the underlying structure is exactly the same. It's largely similar. And one can also benefit from the well architecture that we have been developing for the autonomous driving, which is kind of similar. I didn't mention before, but actually, our architecture is based on 2 stacks. So the first one is what we call the fast stack. So that means the one that received the input data and provide the steering position and gas position. And then we have on top of that, an LLM based or ELM, stack, the slower stack. That actually synthesizes higher-level information about the seeing the context and drives the lower level. So even in this way of moving towards higher level of automation. That's -- I mean, can benefit from that.

That's like human brain. Like there's a popular book called like in slow think fast, which is about how the brain works, there's a natural thing you do fast and something that you learn and get fast. So that's great. Yes, that's pretty close to what we're thinking. So well, I think that was kind of our kind of panel section and coming back to the stack. Before I can let you folks go short 30 seconds, what does the future look like, Bob, for architecture?

I mean that's kind of a dangerous question because maybe I know a little bit too much the that I can really say. What I would say is I think that we've had a lot of experience now looking at deploying all of these transformer-based networks and on our architecture. And we have a VLSI team that works very, very fast. And so the important thing is to make sure that they have interesting things to do. So we're looking at what are those opportunities. Now that we've understand and see how the systems operate on our hardware, what does the next generation look like? And I think that's what we're focused on and in sort of even more efficient and more power -- low-power devices in the future.

You're an environment.

Yes.

That's right. Alberto, what about you? Where is the trench going?

I'm going echo what Bob just mentioning, actually having the possibility of having these 2 layers architecture going together, mixing the 2 things together instead of having just 1 slow stack in 1 at fast stack, having them together, thanks to the higher power that we can get. That's the future.

So t 2 brains, 1 brand. Finally, Mohar, why don't you -- where does this agentic and where is the software going to go, right? .

Software as such is how quickly you can enable customers to bring up this SLM or SMALL Language model that you are referring to, the VLM and LLMs, quickly through agenticworkflow, we can bring it to the edge, the small model. That's what it is -- the key part is going to be moving forward.

Thanks, folks. And I'll let you back in as you can head back. Thanks for your time. Yes. . As the folks head back, I'm going to give you a quick update on our kind of business and portfolio updates. And I think so you get a quick view. I know we're going to run on time. So I'll give you a fast kind of part big parts of our business. IoT, which is kind of one of our largest kind of business portfolio in 25 is almost 70%. As Fermi pointed out, there's more and more AI that's happening. We got a nose presence in enterprise security. From enterprise security, we're seeing it move into public home safety kind of markets and smart home markets and a lot of kind of announcements along those but also going into portable video. You see drones that are flying outside. You're seeing robotics, as we've been mentioning that there's adoption of these edge IoT kind of environments that are happening on one end, which is end points. But on the other hand, we're starting to see emergence of this edge infrastructure applications that are also saying, "Oh, I need a small box of compute where I can aggregate all of this and compute at the edge without a massive server environment." So we're seeing huge uptick of this business that is going with the range of articulated from first gen, second gen and third gen and the attributes just that Bob and Mila kind of mentioned it from a unified SDK, a family with a common set of software One thing I know we take for granted, but we have some of the best image quality, like leading AI ISP in the market. and I'll talk about CV7. But that's huge, but we take it for granted, but a lot of people don't realize how good and high quality and how advanced we are. Power efficiency, I think Bob's job continues to be an environmentalist, but we have a very power-constrained. And the GenAI models, like I know like last year, DeepSeek came out within 2 months, we had it running, right, because of the flexibility and balance that we're looking at and doing this. So a whole range of chips C7 is something we announced yesterday. So CV7, as an example, goes from CV5 with 2.5x more AI and the CVflow 3, 2x more encode, 2x for CPU, of course, is 4 nanometer. So lower power kind envelope as well and supporting all the applications, Fermi that slide with applications on CV5, we're already seeing the designs of CV 7 in a lot of these kind of environments. And what's kind of really interesting is we announced yesterday with our CV7 and the performance, we had competitors announced a similar chip and new -- like new chip from them, but it's still -- we're 2x performance at launch with those folks. So I think that's what some of the -- a powerful thing for us is being able to keep up and keep launching at the same time with competition, but we're already 2x ahead of them and what we can do with 8K Vision and 60p that other folks can't do it yet, right? So that's differentiator. But me talking about it, we wanted to also have one of our customer testimonials couldn't be here. But from IQSite, we have Sabrina the CEO talking about it. [Presentation]

Hello. My name is Sabrina Stanburn, CEO of IQSite. IQSite was recently launched as the driving force behind Bosch branded intelligent video solutions. As a spin-off from Bosch, we have a global footprint and a solid foundation of proven reliability, quality and a relentless drive for innovation. By using AI, we help build a future where safety and security incidents no longer disrupt our lives. Our cutting-edge video solutions combine world-class hardware with intuitive software to protect people and assets. Bosch cameras are installed across the globe in a broad range of applications. from buildings and public spaces to critical infrastructure. At a IQSite, we will continue to lead the way in AI-enabled video. We are on a mission to advance predictive security to help our customers know what's coming next, act faster and unlock insights beyond traditional security. This is an exciting period of transformation for physical security driven by innovations and AI technology. And we at IQSite, won't leave the way forward. Today, AI-based video systems can classify and tie capture image data to streamline monitoring. As we embark on the next generation of AI, we are providing customers with technology that will help them overcome the inefficiencies of human monitoring. Our new portfolio with GenAI models allows us to automatically monitor and generate scenes, descriptions and the images recorded by our cameras. For example, our cameras understand the difference between a person on the phone walking close to a car and a person actually vandalizing a car based on the same natural language GenAI technology that has taken the world by storm. -- these visual language models help ensure that no potential threats goes unnoticed and at the same time, avoid false alarms. Over the last 2 decades, our partnership with Ambarella has been the key ingredients of our innovation and physical security. Ambarella's impressive video processing heritage and their systems on chips incredibly efficient AI performance per watt enables us to run our AI software on edge and low-power devices like our cameras. Our joint innovation continues. We are expanding our edge AI portfolio with more advanced mutual capabilities. These will allow us to generate reliable alerts. And thanks to ASP for Ambarella, we will now enhance video quality in real time. at the edge, it's important to have everything integrated into one chip. The perception, AI and general processing blocks. That's what Ambarella's single system on chip does. It makes employment of visual language model so much more efficient. And since we run Gen AI-based monitoring directly on our cameras, we reduce human error and cut down on cloud processing costs for our customers. But there's more to it. Everything happens locally, which means faster response times plus sensitive data stays on the edge, which improves privacy, security and reliability. Looking ahead, as we continue to further improve the efficiency of our Edge AI technology, we see a huge potential to unlock new revenue streams from these advancements. Our collaboration has been a key driver to our shared success. Together, we've combined cutting-edge technology with deep market expertise to deliver a meaningful impact for customers deploying Bosch video systems. Ambarella's world-class support has consistently spout no matter how complex or challenging our requirements have been. We at IQSite, look forward to continuing our joint collaboration with Ambarella in the security and AI markets.

Thank you, Sabrina. I think it was a pleasure kind of working and work with some of our top customers. I just want to highlight, if you haven't had a chance to go look at some demos and robotics, but robotics is being adopted from different chipsets all the way from just vision in Level 2 to stereo vision Level 3 as well as the brain autonomy that we just discussed with Alberto over there. We have a whole bunch of kind of demonstrations. But one of the first ones you will see flying around is the antigravity drone that's if you haven't visit, you should go check it out with the vision goggles. It's -- we call it the first type of robotics as these robotic drones that are flying around on wise commands and following item stuff. So let's quickly kind of switch over to auto side of the business as well. So that's -- auto is the second part of our large business, 30% of our business. But we're seeing, again, huge AI adoption auto safety, telematics business is growing significantly. We've had a lot of kind of design wins and announcements with key customers in that area. And we see AI utilization inferencing and a lot of adoption in fleet and aftermarket like drive recorders, e-mirrors, driver management systems, occupant monitoring systems as well as fleet and telematics. At the same time, we had this good discussion with Alberto on where we're going with our AD family and software stack from L2+ to L4, starting to get some design wins and software can be built out. So we have a whole range of chips. And this is very important is people have a lot of chips on different kind of markets. But to Fermi's point, we also have this whole range of these legacy chips, all of them ASIL compliant, all of them ECQ100 compliance. So all the way from vision, viewing, sensing as well as the RADAR family for Oculii acquisition, which is our CV3 family, which is available, and you should take that demo out. But beyond vision and perception of CV2, CV7X and 5 range of series, the CV3 AD, which also acts like the domain controller, like central domain controller for all of that. you need that brain, that aspect of that as well along with the viewing and the sensing aspects of our portfolio. And performance-wise, you can benchmark, and if you see some of the demos, we have amazing performance against competition that you can see in real time as you go check it out, right? So again, enough said, we'll have one of our customers' testimonials from Kodiak.

I'm [ Jamie Haffacker ], VP of Hardware at Kodiak Robotics. I am responsible for our hardware platforms, including our Class 8 big rig trucks. I've been fortunate enough to work with Ambarella for almost 20 years now, including developing an Ambarella ASIC basin coder, as some of you may remember, that was used by NBC to broadcast the 2008 Summer Olympics. So what do we do at Kodiak? At Kodiak, we're a leading provider of Level 4 AI-powered autonomous vehicle technology. And today, we're focused on tackling some of the biggest jobs in trucking. Kodiak has built a single integrated software platform designed for deployment across 3 main verticals: long-haul trucking, industrial trucking and defense. Take our industrial deployment with Atlas Energy in West Texas or we've been contracted to deliver 100 AI-driven trucks. These trucks are operating today completely autonomously. We designed the Kodiak AI driver to operate in challenging driving environments, and this is definitely 1 of them. We operate every day of the year and truck uptime is everything. We can't stop delivering even in heavy dust or rainstorms. So this is where we turn to Ambarella to customize the solution with Kodiak. We need to do the best camera SoC available on the market that we needed it to be robust across a wide range of conditions. Today, each of the AI-driven trucks in West Texas is running with 4 Ambarella CV2 providing best-in-class performance, especially critical in low light and high dynamic range driving conditions. Beyond our West Texas deployment, we are particularly excited about our current work with Ambarella on the C3 platform. Not only does the CB3 provide best-in-class camera performance, but we're also able to utilize the additional processing power for RADAR and LiDAR processing as well as running our time-critical neural networks at the edge. Using the CB3 for sensor processing also materially reduces complexity and the power needs of Kodiak's overall solution. Instead of running individual cabling to the central compute node, for example, we're able to do core processing next to the sensor, which also reduces central computing demands. Ambarella continues to be a great partner to Kodiak, delivering practical solutions and top support, allowing Kodiak to stay on the cutting edge.

Thanks, Jamie and the team at Kodiak. So I think I'm going to wrap up the auto section, as we discussed with Alberto, there is both that online software stack from deep sensing across images and imaging and radar but also the deep planning capability for us to do motion forecasting, maneuver and trajectory planning as well as pet control. as well as an offline data pipeline, right? So which is basically data capture autoanotation that Alberto touched on and our data selection, ODD, which is very important and how do we run future data acquisition campaigns. So that's kind of the -- our auto part of the business. Now Fermi mentioned, we're also thinking about expanding our go-to-market. And to touch on the go-to-market aspect, which Here's our current go-to-market as it stands, right? We take and have our silicon, it goes to ODMs, Tier 1s, goes through OEMs and Tier 2 kind of partners and scale. I call it a typical push motion, right? So it's just basically, hey, we take a design win, we go into and gets involved. And what you're seeing as we kind of evolve our approach in our stack and going to a full stack solution by the time if you leave -- remember one thing, you're going to remember this full stack, right? So this full stack solution then requires us to actually also create what I would call a pull motion. A pull motion happens when you start building beyond your design wins to start having software, which is built an application which are built some affinity to your silicon. So that is usually built with vertical use cases. with ISVs, software vendors, who kind of solve what's a particular problem and how do you kind of scale that and then get that deployed through system integrators in the global kind of landscape of doing that. So with that in mind, that's the kind of expansion and growth we want to see through our channels and scaling that we expect to kind of go in our next phase of growth. And I think that's definitely why today we launched in our developer one. developer zone allows for folks to come in and build applications, applications, test out our models in the cloud before they go and actually go and do this evaluation. So yes, we've talked about our edge AI software stack and the Cooper is developer platform, where we have our SoCs, our operating system, our kits and our SDK. But today, as part of that developer zone, we're also launching our capabilities of model garden. I think Malhar touched on this. It's dozens of models which are tuned for our second already. genetic blueprint. Agent gives us a level of automation before if I'm just going to go and enroll developers, have to publish my SDKs, APIs is a cumbersome process for people to learn how to code with our APIs, but fix it. Agentics allows us to provide that level of automation very quickly. and be able to -- so we're creating agentic blueprints, which are published on our website with several applications. And at launch, we announced it 2 of our ISV partners and [Cogniac and Mel CX ] we have the CTO of Cogniac, Sandeep in there. We'll have a wave, and then we have the COO for Mel CX [ Thor ] in the back there. So they're launch partners. What's amazing is within a matter of weeks and this we're able to actually bring their applications in retail market, in hospitality market in transportation, like railway network applications to be ported in our platform very quickly. Shows the programmability of the software aspect, how fast we can go and get this done. So we're kind of opening this early access program for developers to kind of come and start building this. applications have affinity to our silicon and we're enabling that because that will give a level of scaling that's kind of new to us. So when the developers -- I think Malhar was trying to articulate this. So we're able to kind of bring any type of data, rear all simulated data and take foundational models. We published a few eventually, we expect there to be any model. And then bettering it to us in a platform to a software stack to put and train that but then test those models in the cloud or eventually in the kids where they can attach any type of modality and deploy that to different types of any edge. And then, of course, you want to fine-tune that model as you kind of do this. So our software stack is really kind of providing that capability for any model, any type of data, any modality to any edge, really, what we expect is like I think Malhar said this, AI is happening super fast. So we don't know what comes in the next 6 months or a year, what type of new models and technologies. But as Ambarella, we want to be ready for any. So any type of model, any type of modality. We are ready, and our flexibility and architecture is really available for that. And that's -- we'll prove that through our kind of ecosystem and how we go to market. And I think the timing is also critical because we're seeing this explosion of as we saw that transition. Agentics provides us a level of sophistication and automation that hasn't existed in the past. So the timing is really important that we're doing this now because the onboarding of ISVs takes days and weeks, not months or years, right? So that's really the speed at which we can go. And our installed base our type of model architectures that are already solved for our ecosystem is ready to kind of go and expand on this at this right point in time because the time to market with us is going to be super fast. And to close up my section is really our growth trajectory will have a chunk of business still growing, expanding our IoT and auto customer expansion. We'll see additional revenues go through this channel scale motion because now suddenly, you have distribution resellers, ISVs, SIs will start picking and taking a whole range of customers, which we haven't tapped into in the past and in citing that of new opportunities for our AI and infrastructure customers happening. So really excited with the growth that Ambarella has in forward. With that, I want to probably pass it on to Chan to talk about -- Chan's our COO, has been here a long time and could talk about skilled operations and our VLSI sophistication. Over to you, Chan.

Hello, everybody. My son watches YouTube and the guy so impatient would play everything at 2x or 4x talk, it sounds like it's running like that. So maybe I need to speak like that, huh? Yes. So thank you for coming. Okay. I will introduce you to my team, what we have been doing and maybe just a glimpse of the future. So from the beginning, we have taped out over production associates. And Bob just mentioned, he said AI. And it just hit my head. It was our 130-nanometer first silicon. Yes. Bob looks the same except he had maybe more hair back then. So the latest as we take out just recently with Samsung, it's 2-nanometer. It took a lot of work, but it was beautiful. What's really remarkable about our silicon team is not just how many chips we take out, but it's about success rate. 97% of our take out actually went into production, which is a single spin or less. Out of those, 70% of them actually went into production with no spin, E0 production. And this was -- I used to work at Intel and there was this dream of Intel to go into production with E0. But I was reading some news about Intel. I guess they're taping out E0, stepping to go into production, which I was quite sad to see this kind of thing, but at least we are doing a much better job here, and it is truly exceptional in my mind. Okay. microphone is not working. Sorry. So recently -- most recently, we have been working on these more advanced node products, 10 nano, 5 nano, 4 nano, 2 nano. We taped out 14 production SoCs as recent as a few days ago and 2 nano and 12 of those are actually in production already, demonstrating our technology leadership in every sense. So Ambarella has been in this business for quite some time, 22 years. We have gathered some domain expertise over time. We have very mature proven vision and multimodal AI IPs and deep system-level expertise. Ambarella team is known for very efficient execution as Fermi was referring to. And at the same time, high-quality silicon. So as you can see from the number of spins we do on the silicon -- we also do rapid process node migration, moving down the node process node very quickly as a reliable semiconductor partner to our customers. And lastly, we are very proud of this $400 million -- actually more than $400 million shipped to date. This showcases our product quality. We can scale and our scale of operations, we can ship these products to our customers' product and shipping everywhere, which leads me to my next topic. So as a semiconductor company these days, supply chain and operation becomes very important. -- ever more critical now with the current environment, you are all well aware of in the backdrop. So to that end, we have maintained decades long strong cooperation with our partners. Samsung Foundry, I recognize some faces here from our partner. So I'd like to tease and call out some name just for fun. There's Peter I see. And Sean, and -- oh, there's Kelvin, yes. Samsung Foundry partners are here to support us. We worked together for 17 years. Fermi and I had a bet actually, are we the only company that's been working with Samsung exclusively for so long. And I lost the bet actually. He was right. We are the only one. So -- what this long and strong relationship gives us is we are able to get early access to the most advanced node like 2 nano process GAA technology and able to get stable wafer supply through so many supply chain shocks. And we have to grant those over and over recently. Our supply chain partners are global, geographically diverse and resilient to supply chain shocks. And Samsung Foundry, for example, have mega fabs in South Korea and Texas. And our OSAT partners like ASE, CGuard and others, we have locations across Asia, Taiwan, South Korea, Southeast Asia. So with our operations team, very flexible, and we can rapidly reallocate resources, relocate our resources to remediate any kind of surprises and shock and stress in the supply chain very quickly. So that's been our strength. Next is Samsung Foundry's video clip speaking is Margaret Han, who is our -- who is a U.S. Head of Samsung Foundry.

Hello, I am Margaret Han, Executive Vice President and Head of U.S. Foundry and Samsung semiconductor. In the AI era, Samsung Foundry delivers customized solutions across a full semiconductor value chain, from advanced process technology and design IP to full turnkey services and advanced packaging. Our state-of-the-art manufacturing fab spent from Korea to the onshore facilities here in the United States. These capabilities enable next-generation AI system to process a massive amount of data faster, more efficiently and with great preceding. One of our most valued long-term partnership is with Ambarella. We have worked together for more than 17 years. Starting the 45-nanometer node for their first video processor SoCs where we help them become leaders in HD broadcast video and sports cameras. Over the years, our strategic partnership has delivered more than 40 products. This includes Ambarella's second-generation AI SoCs built on our 10-nanometer process between 2019 to 2021. And today's third-generation AI SoCs implemented on our events 5-nanometer, 4-nanometer and 2-nanometer nodes. Most recently, Ambarella announced its new Edge AI CV7 SoC family here at CES. These devices are built on Samsung Foundry's latest 4-nanometer process. Within days of receiving alpha samples, Ambarella is true casing the CV7 at CES, highlighting its readiness and real-world impact. The CV7 delivers a major leap in AI capability. In able intelligent processing of multiple live 8K video streams with improved energy efficiency. We truly value our long-standing partnership with Ambarella. Through close collaboration, we have helped ship more than 39 million Edge AI SoCs and more than 385 million SOCs in total. The newly announced CV7 family built on this success Using our 4-nanometer process and advanced architecture, it delivers higher AI performance per watt. This enables support for Lantis VOM, VOA and agentic AI models across both aged infrastructure and physical AI applications. Looking ahead, our joint innovation continues at full speed. We are excited to have Ambarella as 1 of our lead customers to enter production on Samsung Foundry's most advanced 2-nanometer giga around process technology. Samsung Foundry will continue to empower Ambarella to exceed customer expectations, together, advancing the future of edge AI and semiconductor innovation.

Okay. Thank you. We love the phone nano actually. It's a beautiful process node. So one last thing, one more thing. So Ambarella since founding, we have done in a very limited way and selectively some semi-custom joint product SoCs with key customers. very few. More recently, we are seeing very strong interest and our customers' voices are getting louder and louder, asking for custom products, customer SCs or semi-custom products to differentiate. And it's sold out. It's ringing in my ears now. I've been hearing it all last year and recently also, we think there are 3 main thrusts, 3 reasons really driving this. First is just the cost, simple cost. So 2-nano, 4-nano SoC development, it's economics. It's so expensive. We see people estimating hundreds of millions of dollars to develop such SoCs. And in 2-nano, some people are estimating $1 billion investment. So mistake or surprise while you're doing -- while you're investing $1 billion, failure is just prohibitive. And second reason that I think is happening is there's scarcity of design talent. So pool of engineers who can actually develop these advanced node complex SoCs are just not enough worldwide, and it's -- everybody wants to design such products, but it's just not easy to find people to do it. And the third reason, I call this Apple, Apple invested for decades to get to where they are in billions of dollars maybe tens of billions of dollars. But system OEMs need this differentiation in their products. And the heart of this differentiation starts from silicon. Silicon is differentiated silicon is what you really need. And this comes from Steve Jobs at Apple in 2000. I think he started investing heavily into this. So those are the 3 reasons. Behind this -- in this environment, in Ambarella, we have unique system design expertise, years of years in this business. We have stable and mature SDK and vision and multimodal AI IP that's proven in the field. So I think Fermi was referring to this 2 nano product, semi-customer SoC. We just taped out this product working jointly with our key customer and the spec and all aspects of design was jointly defined. And it will be used by our key customer, of course. And because there's just so many -- so much interest and demand we are seriously exploring expanding this custom and semi-custom SoC product opportunities. And this is something that we can look forward to in the near future. Thank you. So Louis can come up, we are going to Q&A.

Yes. Thank you, Chan, for your presentation, and thank you to Samsung for 17 years of support. Look forward to the future together. We're going to -- we've used up our 90 minutes. We fit a lot into that. I hope you agree with us, but we'll jump right into Q&A now. And I think Casey -- Sure. I think, Casey, if you're here, if you arm yourself with a microphone. And if anyone has a question, please raise your hand in the room. Sure, go ahead. I do have a few online questions as well. So there's a question over there. Yes, Kevin? Well, let me bring you the microphone first. Yes. There you go.

Yes, thanks for the presentation and a very impressive history. And last topic caught my attention to custom SoC. ASICs were very popular for a while in the '80s and '90s and then the cost, I guess, even FPGAs or cost of doing the new design got to be too high. But now ASICs are back again because you can't get accomplished any other way, and the cost doesn't matter to some of these companies. where would your product fit in? Is it because you can't do it otherwise? Or is it a low-cost solution?

It's definitely not for low cost. It's really for a differentiated IP. All the customers who come to us who ask for this kind of service or cooperation is because they value our IPs, perception systems, particularly AI accelerators and also our 2-nanometer expertise. All of that is the reason they came to us. If they don't appreciate any of the IP, we won't take that business anyway. And also, if they want to come to us for 2 nano because of the cost, I think they come to a roundhouse. I just -- Ambarella's famous for the 60% gross margin business, and we're going probably trying to maintain our corporate gross margin on that.

Congratulations. Maybe a follow-up on Kevin's question about the semi customer custom ASIC business. I guess, just walk us through how that might impact financials how much NRE, the NRE sort of rev rec is revenue? Is it contra R&D? And then do you sort of once you get into production, sell the chip and what kind of margins? I mean, some of the bigger ASIC vendors might talk about margins below your target of 59% to 62%?

So thank you for that question. In fact, that's one of the reasons we kind of hesitant at the beginning, but I think with a lot of enough interest, we think we should need to start talking about this. Maybe let's use the first semi custom shows an example is an NRE payment that already being -- majority are being done because we take all the chip. And also that the ASP has been a green that is closer to our gross margin. I don't think there will be a significant impact. But for the future projects that you can see there for the company if they come to us, they might have the expectation. We definitely want to figure out a business model that can benefit both of us. So I would say that for the first product, the first semi-custom chip that we design for our customer, the impact to our gross margins a little. But moving forward, we need to figure out that business model, and we'll continue to report to you when we make more progress on this.

Let me do an online question and then we'll go to you, Corey. Several different ones. Let me paraphrase it. One, I think this question, it is for Muneyb and it was about the comment scaling through the channels. And if you could talk about the expectation for kind of the sequence of events or milestones that investment community should expect as we move that direction.

Yes. I think we have some clear milestones in this year to start, We started with the ISV community. And there's a sequence too, because what ISVs like Mel CX and Cogniac kind of bring is a lot of application catalogs. So people start getting visualization of type of use cases. So as more and more ISVs open up, we'll have a lot -- a catalog of different applications and use cases. then you start getting folks from -- so we have targets of by the end of the year, we'll have dozens of then you start onboarding channels and distribution and resellers that happens in a few, and you don't want to do too many. But these are value added, not what we already do. There's additional on that side. And then have system integrators, a handful of them also come in. So by end of the year, we should have some progress in building out the priority is set of applications, more and more, the better. And the faster we're seeing, so it will be accelerated. Second, system integrators who can kind of bring this together, putting system hardware software and taking it to market. and then having distribution and channel resellers. By end of the year, we should have some pretty significant kind of opportunity sign up. The revenue flows will start coming in the future years, but onboarding them, as you know, it takes time.

I did miss part of that question, which congratulations on having your First 2 ISVs. I'm here at the show. Let's see. I think, Tore, you have a question. Let me just get in one more online. And that, I think, is for a combination of firm in Alberta. And the question is the question is about one of your competitors made an acquisition today a company that's experienced in the autonomous driving area, in the robotics market. And so can you talk more about your robotic strategy and how it will play out for you? And also, given that you have the SoC and Alberto talked about the software being applicable for that market, why not -- the question is why not just do the robot yourself?

Well, maybe let me ask the -- answer the question -- the last question first. They are just 11,000 different type of robots. There's no reason, no way we can address that. That we can do autonomous driving software because that's 1 software can serve all the OEMs, but I don't think that a robotic software stack, there's no application that we can set for that. But however, I missed that acquisition news that you mentioned?

Yes. Mobile Eye acquired a robot company.

Oh, I see. I'm sorry, I missed that. It's a busy day for me. But I really think that this is a really new opportunity for all of us. we talk about aerial drones, and I can expect that there are huge consolidation on the drone company in the next year or so and tell applied to all the robotic companies. So I won't be surprised to continue to see more and more people. But for Ambarella, our strategy is very simple. First of all, as a semiconductor company, we need to focus on volume. And we focus on auto driving cars and also aerial drones first because the volume can support our R&D investment. But more importantly, the second point that Alberto made in his presentation that all our investment on time driving can be a help of the robotic software stack, that's definitely something we leverage on and we're going to continue to use that software stack doesn't -- not only just for the demo purpose, but we are going to open up the software just like we opened up for our OEM to license our software stack. We're going to open up our software stack for our customer -- our robotic customer who want to license our software. It doesn't matter what perception, the decision-making VRMs or LLM, whatever we have been putting to our system will be opened up for our customer to leverage on. So for our robotic customers, they can focus on their expertise, and we will provide other pieces of IPs that they think they can come for us. But from the technology point of view, maybe Alberto may want to have a few words.

Just perfect, Fermi.

Sorry, that's not intentional.

Yes, Tore, please go ahead.

Tore from Stifel. Fermi, could you just double-click on the timing of the strategic change to go after more of a channel strategy? I mean, is it because physical AI is an inflection? Is it because of all the IP you've developed on the auto side that cannot be leveraged to new markets? Is it the maturity of Cooper? I mean why now, I guess, is the big question for me.

This idea started, I would say, at this more than a year ago. The reason for that is when we start engaging with the robotic pain, physical application, it becomes so clear, there are many, many opportunities. Most of them data company trying to come to this market, they -- none of them has a mature solution, right? So for us, in the past 20 years, we focus on key customers in the large areas Motorola, the large companies, that has been our go-to-market strategy. But when we start engaging on this physical application, it became very clear very quickly that our traditional go-to-market strategy doesn't work. And from there, even before we engage Muneyb, we helped tremendously have joined to put this whole thing together, but we, at the time, even before we hired Muneyb, because we've become clear that this has to be a strategy for Ambarella. And really it's because we want to be the leader of Edge AI and the way Edge AI is developing is really go through this thousands of possible customers, and we need to find a way to support them. And I think Ambarella has the technology, silicon software and everything, but that go-to-market strategy has to change fundamentally change so that we can engage the customer quickly. You're going to continue to hear us, particularly MuneybMuneyb and myself and Louis start talking about how this give you updates on this progress. how we engage with ISVs, how they help our go-to-market strategy, and that has to be a fundamental change for this conference today.

We're already about 12 minutes behind. So we'll take maybe another question or 2 in the room before we take the break for refreshments in the back, yes. As a reminder, we'll have refreshments and some light snacks right outside to help you pregame for your night in Las Vegas.

Just -- Fermi, do you think there's any need as you go more into like the SoC ramp for AI, Edge AI that you'll need connectivity assets? Or do you think like you can do this just on the device and you don't need to do that? And then yes, I have a follow-up.

I think your connectivity means how to scale silicon to build -- to address trillion parameters type of application. I think eventually, we will, but the connectivity connection that is required for edge influence is quite dramatically different than having a connection for the training system in the top data center. We definitely think that our -- when we approach to this market, we are focused on most of the HA applications that can be addressed by single silicon without going to multiple silicon solution. But that having a multiple selling solutions to our customers to scale up on the Edge AI performance inference, I think it's critical. But that technology doesn't need to go expensive as data centers. So we already, in fact, Bob says with histology, he cannot talk about the details, but you can imagine that, that has been a hot topic inside Ambarella, how we want to have a next-generation a chip that can scale up for those potential influence solutions. And we will give you more updates later.

Would that be licensed? Is that organic? Or would you have to license that? .

I think it's definitely licensed. And that's definitely an area that we don't think that we have in-house capability. But also, we only need the physical connectivity, right? On the higher level there's anything we can do to help to integrate the basic or basically the link service into our system and how to integrate a service into our system most efficiently. I think from an architecture level, we know better than people outside for the inference engines.

Got it. And then maybe just sizing some of the stuff. So like to do a custom, semi-custom chip, the customer has to think that the revenue like the life -- let's say this is a 3-year chip. The lifetime revenue for this ship has to be -- I mean, it must be over $100 million be able to -- is that the right way -- like is that the right way to think about it?

We do that, we won't engage -- but all the customers we're engaging have easily can produce more than $100 million revenue per product. .

And then just lastly, I guess like I think 2 quarters ago, you announced some big edge AI wins, which was positive to see. Last quarter, we didn't really hear much of an update. But -- so I guess just -- does that indicate the momentum had changed in terms of design wins? Or is it just like lumpy and you kind of -- like just the timing of announcements is different? I guess this maybe you can talk to that a little bit.

So first of all, yes, it's lumpy because they're people are trying to figure out. So we are engaging -- the one quick update to you is we are continuing to engage multiple edge infrastructure customer talk. In fact, some of them in our space here to give some demo prototype. And that particular design win will go into production in Q2 this year and definitely that will trigger other conversation because that's brand new application and then how easy to -- the easier use of that product definitely will trigger -- give people an idea how to implement similar products in the space. So I eager to work with our first customer. And maybe first, a few customers to introduce the product, so that would definitely trigger more of the discussion in this space.

I just wanted to add -- just adding that right, so you're going from these edge infrastructure. You're going from design wins from more consumer to B2B enterprise. Those design cycles are a little longer and then more mature. So it's going to be a little bit more time. So as we see this, the end market you're going is B2B enterprise, and that's a longer design cycle, if you think there. So that's also the cadence that you're going to see differently. Unlike a consumer, we're coming next, next, next, because on hand to the consumer market. Enterprise is a little bit longer maturity and cycle, but not as long as like auto, but it will take just a little longer, right? That's it.

Had a related question today. It's not coming in online, but I saw maybe John and I saw maybe 20, 30 investors today. And this question is probably for you, Muneyb maybe Malhar that Ambarella has such a reputation of taking in data from real-time sensors, but now in some of these new applications like edge infrastructure, less time-sensitive data is needed. What has to change in the Cooper development platform or what's already changed in order to accommodate that type of workload?

Sure. I think A lot of the data at the edge, and I know as you kind of look at us and look at the computer and vision data real time is one aspect. But 80% of the data today is still time series data. They think about data that comes out of machines, PCs, in IPs, et cetera, et cetera. There's a whole kind of aspect of that that's untapped. And what you're seeing is multimodality is how do you combine this to? And what could -- what ready for already a lot of time series data is just text. And with time series and parameters, our ability to handle LLMs, our ability to handle text in a fashion is super critical to now start putting context like you put context as perception for autonomy. Now you're taking context of vision with time series from robotic arms, from PLCs and putting a scene together. So recreating a scene in a factory, recreating a seminar. So readiness because a lot of the time series data is real time, but it's also tech space but combining that with vision and putting that context gives us a huge advantage. We're already ready and already there. This is where the use cases with ISVs and all this is important because the ISVs are building such applications. Sorry, go ahead, Malhar.

On the Cooper platform, we already announced or they're using VLM Gen as a tool. So that's basically -- it's not constrained on the time like the frame per second and all that. Instead, we have this prefill time or the time to first token that's very critical. And we can use batching for various things that we already announced and VLM Gens already does that. And the tokens per second. So in the edge infrastructure, the tokens per second or the decoding rate is less important than the time to first token where we can do batching and take advantage of our architecture. So some of the tools are already added in the in the Cooper development platform.

Thanks. Any other questions in the room? All right. Well, please join us outside for a reception. And also One other thing I want to mention that Jerome, who heads our IoT marketing and Jason Hwang, are you here? Jason? If not, he'll be outside, but he's responsible for automotive marketing and system solutions as well. So they will be available for the reception also.

Thank you.