Oh, okay. I left. Okay, so I have no Oh, here it is. This, right? Yeah, there you go. Okay. All right. Good morning. It sounds like the clock is striking 8:00 A.M., so it is a good time to start. Thank you. My name is Ron Glibbery, I am the CEO of Peraso Technologies. Thank you for attending this morning bright and early. Actually, I prefer this because it gives me the full day. Peraso at a glance, we are a fabless semiconductor company. That is, we make our money from selling computer chips. Oh, so I get this off here. There you go. Okay. We have been in business for a long time. We were incorporated in 2008. We are listed on Nasdaq. Our headquarters are actually in San Jose, California, but really our operating company is here in Toronto, just down the street at Front and University. We actually started the company with a professor from University of Toronto, acquired the intellectual property from the University of Toronto. We do high-speed wireless, specifically it is called millimeter wave, and I will get into that a little bit. We sold over two million chips worldwide, have over 60 patents, and really focused on some of the leading applications, fixed wireless access, which I will talk about, tactical communications, and edge AI or physical AI are the main markets that we go after. I am going to start off with a little bit of a technical lesson. Basically what differentiates us, one aspect is that we work in very high frequency range. You can see, traditional wireless is lower frequencies, in the 1 GHz- 3 GHz range. Something like Wi-Fi, Bluetooth, 5G, and so on. But we work in the 60 GHz range, so that is very high frequency. That confers some advantages to us, but it also presents some challenges that we have overcome over the years. Really this is actually an important slide because it is the last technical slide I will show, but it just really highlights the difference between our wireless and traditional wireless in the marketplace. Traditional wireless is what we call omni. It goes out in all directions, like a sphere. Basically, you can receive signals really in Oh, I see I should have that. It is okay. You look good. Devices in all directions can receive signals. In our technology, we do beamforming. That is really our special sauce, if you will. The idea here is that we concentrate the radio energy into a very narrow beam. Again, presents some advantages, and creates some challenges. Basically, one of the main benefits it provides is that it is stealth. It is very difficult to detect this signal. It is at a frequency, actually, I know firsthand that at our frequency range that I described in the last slide, there are no jammers in the world that can jam our signals. So it is what we call a stealthy unjammable technology. With the beamforming, we get a very long range, so typically we are about 1 km, but with good antenna technology, we can get this up to 25 km- 30 km of range. We have very good efficiency, so we have increased battery life. This is actually quite important, what we call reduced interference. So it is the idea that if you look at this slide or this picture, basically, the advantage of this is that it goes in all directions. The disadvantage is that it interferes with itself, so you get a lot of self-interference. This, you can have very high density of nodes, and I will talk about why that is important later. Actually, one of our main advantages is we get very high bandwidth, what we call basically fiber data rates, so in the order of 3 Gb per second, so very high data rates. Again, those are all the technical aspects of our technology that I will touch on today because they are very high level, but very important to our customers. From a market perspective, we see in the standard market, millimeter wave at a 25% growth rate. This is actually a very important slide just in the sense that this technology is coming on now. Actually, you can check this, it was a press release from Verizon yesterday. We have been predicting this for some time. When Verizon and all the U.S. carriers, for example, and frankly, the Canadian carriers went to market, they were very successful with what they call their mid-band frequency, so the 3 GHz. But one of the unexpected surprises of that deployment was that people were using their fixed wireless services or their internet broadband services, like what we use with fiber here in Toronto. But the problem with that is that actually, that market was consuming the very important spectrum. What Verizon announced yesterday is they are actually deploying now, and converting all of those customers over to millimeter wave because they have got that separate spectrum, and they can use the lower bands for mobility services. Basically, you are really starting to see actually a trigger point for millimeter wave right now, and you can start to see this. It is documented in the press. We are seeing traction, and I am going to talk about where a bit of our focus is really in a military environment. Frankly, defense applications these days are very important. Our ability to work in a battlefield and be undetected and unjammable has obviously become a very important feature in that marketplace. So we are entering markets that have very strong growth opportunities. This is actually the product that we sell. These two chips, this chip is built in Crolles, France. This chip is built at TSMC in Taiwan. It is the same factory that they build the NVIDIA chips, for example. This device here is what we call a phased array antenna. You cannot really see it that well on the screen, but actually it is an array. It is not just one antenna, it is an array of antennas. What we do is we actually combine those signals into that beam that I showed you on the last slide. That is the key to our technology, is that beamforming capability. So we put all of that onto these modules, and these modules are actually what we sell into the marketplace. One of the reasons we did that was because with our first customer, which really 2017, 2018, it took two years to get that customer into production because the technology is so finicky that if you make one small error on this design, it does not work. So what we said was, "We need to fix the design so we sell this working black box product to our customers." That is really the idea that we had in terms of our go-to-market strategy. But really, we control the entire signal path. We have a very strong software team with real-time embedded software. We have antenna technology. We have the radio frequency circuits, the baseband frequency circuits, and even the PCB is a sophisticated PCB. Actually, one of the challenges we have today is that the company that makes these PCBs is actually in China. And 10 years ago, this is very high-speed PCB technology, we found one company in the world that could build the PCBs that would operate fast enough. But it turns out NVIDIA actually found the same company. There's one company in the world that can do this, so it makes it a little more challenging operationally to get this now because AI has come along and said, "Well, this is great technology. We need that for our stuff, too." All of that is part of the value proposition that we've developed for over 10 years now. If you look at our stock symbol, we actually did a financing on Monday, and we raised $10 million. The reason the investor actually put $10 million in was because he actually financed a competitor to us 15 years ago, and 15 years later, they are not in production. What I would say about this technology is that it's very, very difficult to get into production. There's a lot of people that get close to production, but getting into production and selling millions of chips like we have is a very, very large challenge. The moat, what I would call the moat from a competitive perspective, is very, very strong for us because this is just from an execution perspective is a real challenge. That's really the product that we sell into the marketplace. Then we have several instantiations of this, really based on the different antenna configurations, which gives you the more antennas you have, the longer the range. Actually, that's just a quick snapshot. I'm going to not go through this in detail. It's on our website, so there's a lot of parameters for each one. But you can see there's an array of antennas, and this is actually a little out of date. This has actually been updated with several more of these products. This should be up there. We're up to 60 patents now. We have patents really on all aspects of that module that I showed you, either software, antenna architectures. We actually even have developed in the IEEE standard what we call essential claims. So we've got patents in essential claims, and that basically means that if anybody wants to try to compete with us, they have to come to us for a license. That, again, that's very powerful obviously from a competitive perspective. And really, the software, I've become a convert in terms of I'm a semiconductor guy, but our real-time embedded software gives us extremely differentiated capabilities from any potential competition. The markets that I really want to highlight today are three markets. The first market is what we call fixed wireless access, and that's providing high-speed internet. The second market, and really the market that we're— This market really— Actually, the quick summary there, and I will get to it, is that there is about, let us say a dozen OEMs. We sell into manufacturers, by the way, as a fabless semiconductor company, we sell into manufacturers that make equipment for service providers. We have got every equipment maker in the world as a customer. So it is what I call run the table in this regard. So for fixed wireless access on a global basis, we really control that market. So we really saturate that market. The market now that we are going after is secure military communications. Actually, really specifically drones. The idea there is, if you have been watching any of the Ukraine war, peel back the onion, you know that drones has become a big part of that war. Electronic warfare is really a fundamental issue in that war. The idea there is, I have got a drone, I need to control that wirelessly, obviously. But then the enemy has counter-drone measures, and then we develop counter-counter-drone measures and so on and so on. I find now we have got what we call an anti-jammable solution, but the term is a bit diluted because everybody says they have got an anti-jammable solution. But I can tell you unequivocally, there are no jammers at our frequency right off the bat. Second of all, you have to detect this, which is more or less impossible because it is like a needle in a haystack compared to traditional wireless technologies. Third, they have not— The power that they need would be astronomical to get the signal to where it needs to be to jam it. All of these factors make us essentially unjammable. The market that we are actually starting to do proof of concepts with, and this is actually— we did a press release last week on this, is actually what we call physical AI. The idea there is controlling, specifically either humanoid robots or self-driving cars. These are devices in the field that are moving, that need real-time model updates, and the only way that they can get their information into the data centers is through our wireless technology. So those are the markets that really we are going after today. So fixed wireless access. You know what I wish? Is there a timer that I can look at? I just do not want to— How am I doing? How am I doing for time? 8:15 A.M.. Pardon me? 8:15 A.M.. 8:15 A.M.. So I have got about five minutes to— You still have six minutes. Six minutes. Okay. Fixed wireless access basically is the idea that we supply people with high-speed internet. Actually, a market that we have been going after is, for example— Let me go like this. Actually, we have been very successful even in townships in Africa, but actually, our lead customer is a company called Ubiquiti. Ubiquiti is an OEM that is about a $50 billion OEM now. But basically, their entire product line, 60 GHz product line, is based on our silicon. You can see this is at a recent trade show. All of these antennas actually are for our equipment. So basically, this was our lead customer and have been a very good customer for us and really allowed us to pipe clean our process, if you will, and get into the marketplace. I always find this slide interesting. This is a list of, I had mentioned we ran the table. This is basically most of the OEMs in the space. You can see all, this is just a sample, but there's about 60 products in the marketplace based on our silicon. We have been really successful in this marketplace. We're not happy with that in terms of just the market opportunity. We have stars in our eyes in terms of the potential market. From a military perspective, again, we think the drone market is a very important market. The idea that you can control a drone and not be jammed. Now the issue is that, again, unlike traditional technology, it's a challenging problem. We are dealing with some of the top drone companies, certainly in the world and also specifically in Ukraine. We've got very strong contacts in Ukraine now and good relationships in Ukraine as well as the Middle East. Basically, we're really in the hot spots from a kind of drone perspective. Again, the idea there is that it's just virtually impossible to jam our systems using traditional jamming equipment. Again, this is just a little bit more technical. That's basically the actual simulation of what our beam looks like. This is the size of our product that goes into the drone. So very small size, weight, and power. Now we're kind of evolving our product from a drone perspective. This is actually the future of drones. five minutes left. This is really the future of drones, what they call drone swarms. Again, these devices all need to be linked really in a dynamic network configuration, if you will. We're working on our software to enable that. That's actually extremely interesting because that basically allows, let's say a fixed-wing drone that, let's say, is connected to some LEO satellite, could be Starlink, could be some other device. That device actually has baby drones, and the baby drones all connect with each other. Basically, the idea is to have a swarm configuration. Now when you go into the battlefield environment, they may get some of your devices, but they won't get all of your devices. This is really what our end game is from a drone perspective. That's really what we're working on at the moment. This is public information. This is actually quite an interesting application, and we can talk in more detail about that. Basically, the idea here, and this is in conjunction with, again, with the IDF in Israel. The idea is, one of the problems they're finding in these military environments is what they call blue-on-blue or friendly fire and casualties from friendly fire. They're estimating in Israel, for example, about 30% of casualties are actually friendly fire, which is really quite crazy, right? This is actually getting worse with drones now because basically a drone, for example, a high-altitude drone that's 5 km in the air, is going to have a really tough time making sure whether that target on the ground is friend or foe. What we're testing now with the IDF is, you send an initiator signal to the ground. If our chips receive that, if they feel that signal or sense that signal has come, they respond. Now the drone knows, "Well, I can't shoot at that target because I know that it's my own guy." That's called friendly fire avoidance. That's a kind of an application that we're actually working on right now, and we're in testing with the IDF. Frankly, it works the other way, too. What I mean by that is if you have eight drones flying around a stadium and you send a signal up to all those drones and one guy doesn't respond, you know that he's not a friend, and you take him down immediately. This is becoming a real problem. As you can imagine, for large stadiums where there's a lot of people and you get a drone up there very quickly, if he doesn't have an identifying signature, you want to get that guy out of there very quickly. It works both ways, not only for the drones to detect targets on the ground, but also to determine whether the drone itself is friend or foe. That's another application using our technology. Of course, the IDF insists that this has to be stealth. You cannot detect it. If the drone actually says and does a target interrogation, you don't want to give up your position at the same time either, because then the enemy will take that drone down. The last market opportunity and the one we're starting to do, or what I would call proof of concept now, is physical AI. If you recall at the start of the presentation, I showed that concept of beamforming. Basically what that does is it allows us to control the signal in a factory environment. We've actually done testing with this. Now you can have a dozen robots in a small area, let's say like the size of this room, because each one will essentially have its own access point for control. This is in contrast to a Wi-Fi based system, which kind of interferes with each other. The problem is when you interfere with each other, it creates a limited bandwidth. You have limited bandwidth from an interference perspective, whereas here we reuse the bandwidth. This ability to beamform is actually a very important concept in terms of what we call AI automation. We're actually, believe it or not, testing with some of the largest AI companies in the world on this, and also even within China. The idea there is that they use our technology to offload terabytes of data, send it to data centers, have the models developed and sent back to the robots. It's very important for us because of our ability to control the RF signal and make sure it doesn't bleed into the other access points in the factory. That's pretty much the wrap for this morning. Oh, we could give it to the thank. Thank you. My name is Ron Glibbery. If you have any questions, our IR group is the Shelton Group, but please go to our website. There's a lot more information there. Thank you very much for your time today. I appreciate it very much. Thank you. Any questions or comments? Go on. You mentioned the recent press release about raising, I think, $10 million. Yes 19 million shares, almost 20 million shares. That has implications for the valuation that this private investor has effectively bought them again, because they're also talking about him not wanting to go above 20%. Right. Well, yeah. There's some financial gymnastics there that allow him to stay at 20%, I think eventually. I think what we want to have happen there is obviously, we'd like to use that money to get to profitability. But eventually, on an acquisition, for example, he would convert everything he has to the balance of the shares that he bought, right? He actually, even though he paid $10 million, right now, he only owns 20% of the company, but he has the ability to convert that, like a convertible. It is not a convertible debenture, but it is kind of along those lines in terms of what the future looks like. Having said that particular investor was extremely important. I think there is a quote from him in the press release. His name is Patrick Soon-Shiong. He made several U.S. billion dollars from his pharmaceutical startups, but he has access to every single world leader that we need to be engaged with. I think what we foresee is that with our core technology and his global connections, we really feel that the business is going to really take off now because of having a guy like him. Basically, from a financial perspective, he is limited to 19.9% right now, but on an acquisition, he would actually exercise the balance of his convertible options and have more percentage of the company. I am not sure if that is what you were asking, but Well, if he has invested $10 million and effectively got 20% ownership that implies that the entire market cap of the company Oh, yeah. is $50 million. Yeah, that is correct. That is the correct. That is correct, yeah. It does not show up on Yahoo like that right now, but it would. It will when we update the share count. Okay by the way. Yeah. We only update our share count once a quarter, so the next time we update the share count, that valuation will be reflected properly. Yeah. Your annual revenue runway currently is? In 2025, we did about $13 million-$14 million. This year is going to be. We ran into some headwinds this year because actually, the whole AI thing has been a challenge for us because of just getting supply. I am not sure what it is going to end up being this year, but it is going to be down for sure. Hopefully next year we are back up around $17 million-$18 million. That is kind of the thinking right now. Any other questions? Sure. Given the $10 million infusion from that. Yes Do you still intend to draw down the draw facility, or is there a specific trigger that you will choose to use either or? That's a very good question. The answer is probably not. We put that in place as a fail-safe. I'd say likely right now, not really. Right now, we actually feel that $10 million is a forever number. I think in 2025, we kind of spent net $3 million. So the $10 million is a very good number for us. We have to do a little bit of development, and what that specifically refers to is in the drone market specifically right now, we can go about 5 km, and that's not bad, especially for some of the swarm configurations we talked about. The swarm is normally just maybe about 1 km. However, with about a $5 million R&D spend on top of our regular run rate, we can get that drone connection to 30 km, and that's a game changer. A 30-km chip that's non-jammable is going to change some battlefield dynamics. Right now, we don't anticipate using much of that ELOC from Roth, actually. We'll see, but I don't think so. There's no trigger, actually. Can you give us some basic ideas around what the $5 million is going to deliver? In other words, is it you need higher power, you need a bigger antenna? What are the elements, the dimensions that need to change to get to 30 km? The elements that have to change is the output power. So basically, it's very specifically. Today we have 16 elements in our phased array that I showed you that picture of. So that needs to get to between 16 and 256. Phased array technology is really becoming a For example, in Starlink, they actually use phased array technology to go from their satellite to the ground station. They're at 17 gig, but basically they use 1,000 elements. So the more elements you use, the further you go. With the $5 million, that allows us to go from where we're at today, which is 16 elements, up to 256 elements, and that gets us the power we need to go to that 30 km. To be quite frank, there's zero risk that we'll be able to achieve that. We know for sure it's going to work, actually, because the math works. Right? We'll probably be starting that chip next week, actually. So we have a very, very aggressive schedule to get that into the marketplace. Any other questions? I'm totally unfamiliar with warfare and drones. Why is 30 km? Well, it's because it's the limit of most conventional kind of wireless today, because basically you can go 30 km. Maybe it's not a term I love, but there's a term that's very prominent now in the Ukraine war called the kill zone. You need to be autonomous throughout the entire kill zone. Basically, now you know for sure that enemy's going to be 30 km away because you can have drones out there. If you see a tank, you're going to blow that tank up. If you see a battalion, you're going to attack that battalion. Now you've got a 30-km buffer. The longer that, if it was 100 km, and by the way, we could do 100 km, but that's a little more sophisticated. The further your kill zone, the further the enemy has to be away from your front line. That's where the number comes from, actually. It's actually the most important number. Yeah. Any other questions, comments? All right. Thank you very much. Thank you.