I'm Austin Moeller, the research analyst here at Canaccord Genuity. Today, I have Brian Faith and Elias Nader from QuickLogic, and I know you have a few slides you want to go through just to give an intro of the business. Yep. All right. Thank you. Good afternoon, everybody. Excited to be here, excited to talk about QuickLogic. Firstly, QuickLogic is within the semiconductor industry, a sub-segment called programmable logic, and that sub-segment is going through a renaissance right now. We're seeing lots of use cases across all different market sectors, especially with our peer companies. Where we're focused right now is on aerospace and defense, so I'm going to go a little bit more into detail on that. Aerospace and defense happens to be spending more dollars on programmable logic than any other microelectronic category. Within that, they have ASICs and FPGAs that they purchase. Our technology, be it devices or IP, fits squarely within that hashed area of FPGA and ASIC. High growth market. We're looking forward to talking about it today with you and also addressing some of the nice interesting questions from Austin. Before we get to the questions, we have two types of products. We have tangible devices called discrete FPGAs. This is sort of our founding technology, and we continue to sell these today. We're going through a big development contract right now with the U.S. government to modernize some of these devices for very specific use cases that I know we'll cover in the fireside chat. On the right-hand side, we have embedded FPGA IP. This is intellectual property that we license to some of our customers for them to include in their own custom ASICs. Again, it covers both FPGA and ASIC devices and IP. In short, the benefit of our technology is that our customers can take their own IP, program it into their devices, and we don't ever have to know about it. They get faster time to market, and they get better development cost in using our technology than if they did any custom silicon themselves. With that, I will now kick it over to you to ask questions. All right. I guess just to start off, can you differentiate between the chipset designs that QuickLogic was producing or designing with those of some of the peers in the defense market and the applications that they're going into and being used for? Yeah, definitely. If we look at the FPGA market in general, typical companies, you might see Xilinx, which is part of AMD, Altera, which was part of Intel, now it's back out independent again, Lattice and Microchip. Within those, the larger Xilinx, Altera, tend to focus on high-end compute data centers. When we look at aerospace and defense and more industrial control applications, we see Microchip quite a bit more. One of the things that we're doing right now that's really exciting, I think, on that, let me advance the slide here. With respect to aerospace and defense is a purpose-built FPGA for strategic rad-hard applications. This is something that's never been done before in our industry, and I've been in this industry for 30 years. This is very unique in the type of application space it's targeting, and again, it's new and different than any of the other peer companies that we've had for programmable logic. How important is the ability to program and reprogram FPGA architectures over a program life for a system? Yeah, that's a good question. If you think about the way system companies do design, if you look at a consumer company, they may design an architecture like Apple would do, knowing that they're going to sell hundreds of millions of units of one thing in one year, and they're going to refresh that the next year and do the same thing over and over again. So they know exactly what they're going to be doing in that technology. Forward programmability probably has less value for them in that sense. There are other ends of the spectrum, though, where you have systems that are highly specialized and have to live in airplanes or in hangars or below the sea level or in silos, and they have to last for decades. You can imagine that over the lifetime of those units, they will have to be updated with algorithms. Technology evolves, software evolves. Those are the types of applications that really benefit from programmable logic, be it a discrete device or an IP in an ASIC. Again, you could be wanting to reconfigure something decades in the future. If you have a fixed silicon component in that architecture, you will have to redo that entire chip, and now we're talking about tens if not hundreds of millions of dollars of NRE fees and the time it takes to do a custom ASIC design and validate and manufacture. However, if you have programmable logic on that same device, you can update algorithms in hours. So it's a dramatically different economic question, dramatically different risk profile when you can use programmable logic to make those changes versus tear it apart and rip it up and do the silicon from scratch. Can you comment on essentially the premise that once you are designed into a major weapon system or defense program, you're essentially locked into supporting it as a supplier for the life of the program, correct? Yeah. In most cases, that's correct. If you go back in our history, we've had a lot of devices we continue to sell today to different defense contractors for weapon systems. I agree with what you're saying. Once you're designed in, the cost of requaling a system to make a component change is very large, and most of these contractors want to avoid that kind of allocation of resource to go save pennies or maybe to replace something that's been end of lifed. So they really take great care in looking at suppliers before they do their design to know A, that the technology meets their needs now and in the future, and B, that the company is going to meet their needs in the future as well from a supply chain perspective. They typically do design in one thing and then make sure that that works, again, work for the decades that they may want to have that thing in deployment. Again, it's unlike the consumer business where you can pick people based on lowest price, and are they going to be around next year before the design refresh happens? It's very different dynamics in the aerospace and defense industry. What does the timeline look like right now for evaluation of your RadPro dev kits, and then reaching a decision on low-rate initial production, then full-rate production orders over the next couple of years? Yeah. A little bit more context before I answer the specific question. QuickLogic, in August of 2022, was awarded a contract by the Department of Defense, and that contract was to build a very purpose-built FPGA for strategic rad-hard applications. If you look at and you read about radiation-tolerant microelectronics, there's a whole spectrum from radiation-tolerant, which would be for Low Earth Orbit satellites, to rad-hard, which is for more mission-critical applications. Whenever you see the word strategic in front of rad-hard, that really means it's designed to be for weapon systems, defense systems. It's the hardest to get to that level of radiation hardness. We've been on this journey for a few years now, designing these, and last year we shared with our investor community that we have the government contract, but we also decided to self-fund a version of that so that we could get samples of those devices out into the hands of the defense industrial base sooner so they could start their technical evaluation of that technology, our devices. The goal being that once they go through their evaluation, we can start getting designed into programs of record, which is the ultimate goal here. Development contract is nice. The end system is where we really want to be. We want to enable the defense industrial base with this technology. We have development kits. We've told people publicly that we've sold these, we're supporting them now within the defense industrial base, and that we hope by the end of the year, we'll have some indication of feedback from them. Again, the goal being that next year we're starting to get designed into actual systems that could roll into what's called low-rate initial production and full-rate production in the coming years. If we're successful in doing that's the kind of dollar value that we're looking at for each one of these major systems. We're talking about hundreds of millions of dollars in revenue potential, which would be quite transformative for QuickLogic. In addition to the technology itself, there's also other things that we need to bring to the party as far as value proposition goes. A, we're U.S. domiciled, we can allocate U.S. persons only on the design, we're manufacturing this onshore, and all of these are really key, important criteria as the defense industrial base is going through their decision-making process on this. What's your take on the recent discovery in the last week that the U.K. Ministry of Defence had a bunch of Chinese firmware on their components in their unmanned surface vessels, and that was sending data back to China, and the importance of component sourcing for the U.S. defense market? Yeah. My initial reaction is low cost has a price. I think that in general, a lot of the electronics that we've been making as a country, cost has been a primary driver. I know that because we were actually in the consumer space for a long time, selling to mobile devices, where cost pennies per unit matter. I think that that's a really good example you just point out where there is a movement now to have more of a Trusted supply chain. When I say Trusted, I'm talking about with a capital T, where you know the chain of custody of these parts. You know the IP that was used to create them, the provenance of that IP. You know the design teams that are doing that IP. You know that they're onshore. You know when it is manufactured, that what you actually got back from the manufacturer matches what it was supposed to be at the design level. When you get to FPGAs, it is what you program into the chip is exactly what you wanted it to be and nothing more. Nothing phoning home when you are running your application. I think this is actually shining a bright light on the importance of onshore trusted, resilient supply chains. I like to think that we are sort of at the intersection point of the awareness that is happening now, the government funding that has been going on dating back to the beginning of CHIPS Act, where they are starting to onshore critical manufacturing capability. Now we are thinking about what designs, what chips can actually be done based on this onshore manufacturing capability that will be trusted. When I do say trusted, I go back to this slide here. Is it a U.S. company? Is it with U.S. persons? Is it manufactured onshore? Can you document the entire chain of custody of that product from design all the way to shipment and programming? That is the intent behind this, and I think that there is going to be more demand for this now, especially for what I like to call critical infrastructure. It is not just weapon systems. It is going to be our utilities. It is going to be a lot of the energy sources that are going to be deployed because of all the insatiable demand for energy for data centers. All of that has to be protected, and part of that protection is knowing that something is trusted, and you know exactly every transistor that has been designed into that chip. How do you view the U.S. government data center market for your products, just given that the D.C. beltway is the single largest data center market in the entire world? I am sure we can guess what the applications are for. I probably will not go into what the applications are for, but I think the U.S. government is going to be wanting to maximize its investment in the infrastructure especially on the manufacturing side. I think they are going to be giving, I hope, preferential treatment to people that are designing chips for these applications. Not just data center, but think about unmanned vehicles and satellites, anything that touches important information to be U.S.-made and U.S.-supported. What value do you provide to customers with eFPGAs with post-quantum cryptography IP cores on a system-on-chip or SOC? Yeah, it's another interesting topic, and I think there was even a presidential order recently related to this. I think there's a lot of excitement about quantum computing. There's also a lot of fear about quantum computing and what's going to get unlocked when that kind of compute capability is available. One of the things that people fear is that you can do brute force attacks and just break through classic cryptography algorithms, and I go back to this whole notion of critical infrastructure. If you can break into critical infrastructure quite easily with typical methods, then how safe are we? I think there's an executive order that talks about needing to examine post-quantum encryption and cryptography so that we can be putting gates into our system to protect against when quantum computing is more deployed. The challenge with that is that not all these algorithms for post-quantum are actually well-documented, specced, agreed, and no longer going to change. Think about our cell phones and the cellular networks. There's always the next generation that's going to be faster and better. The same I think is true with some of these cryptographic algorithms. If you're designing silicon today that's needing to go into systems in the next couple of years, that will then be out there for decades, think about infrastructure now, not a cell phone, but actual infrastructure, what provision can you put into that silicon now so that you can program it in the future with something that you know is going to change? This PQC, post-quantum encryption, is an example of that. We've actually been talking with companies that are experts in post-quantum cryptography about using our programmable logic as a way of vehicle, if you will, to design flexibility into these chips so that five, 10 years from now, as those standards evolve, as the algorithms evolve, it's just a matter of reprogramming the technology over the air and not needing to rip out printed circuit boards and de-solder parts and put new parts into systems that have it. I think it's a great use case, and we are working with a couple of companies to make sure that that's easy to do for people that want to go about and do that. Can you discuss how we should think about the margin ramp as some of these R&D contracts and test kits evolve into production awards? Yes. I like this chart because it shows from our standpoint how we make money as a company, and we do have three major buckets of product revenue, or I should say revenue. One is services, one is devices, and one is IP. So IP is great margin, typically in the 80% range. Devices in the 60% range. Services, as you might expect, is not high margin. We do service contracts, service designs, when we think that the ultimate product of that will land as an IP or as a device we can sell to customers. We do not do services for the sake of services. Right now, we are heavy on services because of this government contract for the Department of Defense. As that project is wrapping up to completion and that revenue is dropping as a percentage of total, we are planning on device revenue growing, from that development, IP licensing continue to diversify and grow, and both of those two categories generate higher gross margins than services. That's what the thesis is around improving gross margins as a company so that in a normalized state, we are in the high 60s, if not higher, as opposed to where we are today, which is in the 50s range, high 40s to low 60s%. How important is it to the primes to be reaching out to QuickLogic and other U.S.-based fabless semiconductor designers, given the new NDAA requirements about eliminating all China-sourced components by 2030, and preferably everything be produced in the U.S. overall? Well, it's important. We have a longstanding relationship with a lot of the prime contractors, and we have this in our investor deck that we've consistently done business with five of the top five, eight of the top 10 U.S. defense prime contractors, including some of those of our allies in Europe. That commercial relationship is very important because it shows that we are a trusted supplier. We've delivered devices, we've delivered IP, we've supported them. I'll actually go to this slide here. This is one of the slides I'm most proud of because as a company, BAE Systems, a U.S.-based defense supplier, gave us an award as a supplier of the year. This is not because we're lower price. This is because when they licensed IP from us, there was a successful deployment by us in support of them. I think these are the kinds of things that these companies look for as they're looking to have more U.S.-based from the NDAA recommendations. It's not just are you U.S.-based, but what care do you take in your IP, and how well do you support the customer for the long term. I think we can point to examples like this in our 30-year history and say, yeah, we have been there for the U.S. defense prime contractors, and we're going to continue to be. We talked a little bit about the trusted rad-hard opportunity earlier, but do you see any significant opportunities for your products within space-based systems for SkyRange? Yeah, there's I think two questions embedded in that. This is the slide that we've talked about, the strategic rad-hard program. I think from a technical perspective, if you can hit the strategic level, you can definitely hit lower levels of radiation. We all know about the big build-outs that are happening in space, especially around low Earth orbit constellations, satellite constellations. Not just people like Starlink, but other ones that maybe are new business models, different communication mediums, even own countries are looking at establishing their own sort of independence or sovereignty of their satellite communications. From a technical standpoint, this is something we're definitely trying to do a sort of land and expand. If we can land strategic, let's expand into the rad-tolerant domain, that's what they call it, for space applications. I think there's ways that we can do that with devices. There's also ways that we can do that with our intellectual property for people that are committed to doing their own ASIC. It's a good growth market for us that I think we can leverage a lot of our developments into. Can we just talk a little bit about QuickLogic's balance sheet and the capacity to support the program ramps over the next two years? Also, how much of your R&D is generally customer funded? I think we have enough cash to make it to next year and beyond. We ended Q2 with $13.5 million of cash, free cash flow. We have an LOC for $10 million that we use only to window dress the balance sheet. We don't use it to buy any R&D equipment or anything. We have no debt, per se. We don't buy capital machinery or whatever. The biggest expenditures we have is in R&D and in hiring staff, especially engineers. There's only 49 people in the company, and we're lightly spaced in the world. We don't have too many footprints. I think we're okay in terms of that. Most of the R&D expenditures these days are funded by the customer who basically gives us money to work on that. I'm not concerned. Do I have any audience questions for QuickLogic? Can you talk a little bit about Australis and what that means to you? The question is, can we tell about Australis and what that means to us? If we go back to when QuickLogic started as an FPGA company, we did everything full custom. What that means from an investor standpoint is we had our entire engineering team working on one technology for over a year. You can imagine that we were somewhat constrained in what we could do from a customer support perspective and porting our technology. It was about five years ago, we actually came across some, believe it or not, some open source automation capability that had actually been funded by DARPA. We took on board some of that technology, and we built our development flow around that now. What we have as proprietary is Australis 2.0, has open source components, but it has unlocked our engineering capacity. What was our entire team for over a year, one design, is now four teams in parallel that we can deploy on four different designs, four different customers in any given quarter, with that same number of people. Dramatic difference in terms of capacity, and that is really what has unlocked the operating leverage we have as a business model is this whole automation with Australis. It is a wonderful thing, and we are doing every design that we can with that. Oh. Can you talk about your shareholder base, what the insiders own, who are the 5% holders? Sure. Actually, now that we're in the Russell 3000, we've had a little bit of reconstitution of the shareholder base. I think still our top shareholder, individual shareholder, is a gentleman that's a filer. He's a big believer in the story, in what we're doing on the rad-hard side, on the defense side, this growth into the Storefront potential that we talked about. We have several other individuals that have actually participated in direct placements with the company over the last few years. Then I think with the Russell reconstitution, now we have BlackRock, Vanguard, the typical funds. As far as management and the board, I think we own, slightly under, 10% in total, maybe a little bit less than that. Yeah, a very supportive investor base, especially in the last, I'd say, three years when this story has really started to take its stride. All right. Well, I think we're at the end of our time, but I just wanted to thank Brian, Elias again for Of course. coming and giving us the download on QuickLogic. So Thank you for hosting. Thank you.