This conference call transcript was computer generated and almost certianly contains errors. This transcript is provided for information purposes only.EarningsCall, LLC makes no representation about the accuracy of the aforementioned transcript, and you are cautioned not to place undue reliance on the information provided by the transcript.
Operator
Before we start, I'd like to remind participants that during this call, management may make forward-looking statements including, without limitations, statements regarding our future performance, growth strategy, and financial outlook. Forward-looking statements are based on our current expectations and assumptions regarding our business, the industry, and other conditions. These forward-looking statements are subject to inherent risks, uncertainties and changes in the circumstances that are difficult or impossible to predict. Our actual results may differ materially from those contemplated by these forward-looking statements. We caution you, therefore, against placing undue reliance on any of these forward-looking statements. You can find more information about the risks, uncertainties, and other factors in our reports filed from time to time with the Securities and Exchange Commission, including in our quarterly report on Form 10-Q for the period ending March 31, 2022. All information discussed today is as of May 13, 2022, and we do not intend and undertake no obligation to update any forward-looking statements, whether as a result of new information, future developments, or otherwise, except as may be required by law. In addition, today's discussion will include references to certain non-GAAP financial measures, These non-GAAP measures are presented for supplemental information purposes only and should not be considered as a substitute for financial information presented in accordance with GAAP. A reconciliation of these measures to the most directly comparable GAAP measures is available in our press release and you should refer to our reconciliations of non-GAAP financial measures to the most directly comparable GAAP measures in our earnings release. With that, I'll pass it over to Blair.
Blair
Thank you, Clyde. And thank you all for being here today and investing your time to participate in our quarterly update. As you have seen in our earnings release today, we finished our first quarter solidly, meeting both our financial and operating expectations. In addition, we remain on track to achieve our full year plan. While we continue to follow external global events closely and monitor market volatility, our main investor themes and objectives for 2022 remain consistent, and our focus on execution remains paramount. In our year-end earnings call, we outlined our go-forward strategy and our progress to date building product partnerships and infrastructure to meet our key objectives. We also spent time differentiating our unique business model and disruptive technology platform versus peers. In our last call, you also had a chance to hear from several customers in the automotive and industrial markets directly. They shared with us the value the AI intelligence sensing platform brings to their solutions. We would like to first emphasize the importance of 2022 as we intend to both begin shipping the Foresight product for industrial markets with our partner, Samina, as well as transferring the B sample of our first joint automotive ADAS product to our partner, Continental. In today's call, we intend to do a quick review of the market dynamics, the differentiation of our disruptive intelligence sensing platform, and illustrate why we and our partners believe AI's sensor-based operating system is uniquely positioned to enable the evolution of smart vehicles, infrastructure, and assets. We will use the majority of our time today to focus on our execution with an update on the fourth key investment theme, commercialization, industrialization, and capital light manufacturing. We will touch on both the foresight product line, as well as our joint continental ADAS product. We believe we will be the only company in our peer group to bring up volume production capabilities with multiple manufacturers. The market headline is, sensors are a highly desired addition to many vehicles, infrastructure, and other assets. Cameras and radars are interpretive sensors with unique strengths and weaknesses, but have one attribute in common. They collect information and intelligently guess. LiDAR is a deterministic sensor which can provide definitive data for many decisions, enabling new value-added features that can be standalone, like hub-to-hub trucking or highway autopilot for consumer vehicles. Or LiDAR can also complement radar and cameras to increase reliability or accuracy for existing features, such as in slower-speed traffic jam assist. What is clear is that LIDAR's commercial performance has continued to increase substantially over the last several years. Concurrently, its manufacturability is maturing, and therefore size, weight, power, and cost continue to be optimized as LIDAR is being applied across numerous industries. Many of us already have a LiDAR sensor in our smartphones, advanced driver assistance systems in our cars, and we experience traffic flow optimization on toll roads and other parts of our infrastructure. We believe LiDAR has a wide range of applications well beyond what most people have imagined. That said, all LiDARs are not the same. With many traditional LiDAR systems, data is collected in a fixed and limited manner and then passed along to a perception engine. This is a one-way flow from the sensor into an application software layer. AI software on the edge is different. First, we can control hardware components individually using a software-based operating system located on the sensor with two-way communication to change the way the sensor works, depending on different environments. In addition, the Foresight operating system does not silo itself from other sensors. Customers can create unique systems that can use maps, cameras, radars, and IMUs to trigger the LiDAR so they can be more intelligent and efficient when collecting critical information, as recently demonstrated with Continental's integration of its current ADAS suite, including radar and camera, with our joint LiDAR product. Finally, and most importantly, this software-defined architecture is natively compatible to manage data over its local sensor network and to be enabled for over-the-air updates. So we can change the way the hardware performs through software, allowing our customers in the future the ability to upgrade and to add new features and functionality. While this seems too good to be true, you only have to look to your smartphone to see the path that is already being taken by many durable goods manufacturers and infrastructure providers. In automotive specifically, the acceleration of EVs provides a natural greenfield opportunity to create software-definable platforms for cars. The future is now. One powerful example of this software definability is adaptive placement. The Foresight platform enables automotive OEMs to embed the same LiDAR sensor in various integrated locations using AI's proprietary sensing software. This optimizes performance for the vehicle-specific packaging and integration without detracting from design or limiting performance. AI's operating system provides OEMs with the ability to transform the sensor performance and enhance data capture across various mounting locations and vehicles. This is in contrast to most traditional sensors today, which cannot be optimized for placement, tolerances, and applications, making them suboptimal across a platform with multiple brands and models. At the end of the day, the ability to change the mounting locations and the height, as well as correct for curvature and transmissivity of external surfaces, allows us to increase platform adoption, optimize feature implementations, and reduce cost and complexity. This same adaptive placement capability and software definability conversely allows AI to customize across markets, allowing the use of the same hardware on a roof mount at 4 meters and a negative 40 degree angle on a Class A truck as a grill mount at 65 centimeters and a negative 15 degree angle on a trendy sports car. Up until this point, we have been talking about how our adaptive systems can add intelligence into current vehicles, infrastructure, and assets. So let's take a step back and discuss the future and what differentiates the software-defined vehicle from a traditional vehicle today that has intelligence siloed in many subsystems. On the left, you see a vehicle with all of its technology and functionality set when you purchase it. In many cases, you would need to physically change or alter a component to adjust the hardware functionality of the vehicle. On the right, you see a vehicle with a more streamlined platform reference design, reducing complexity and allowing for the flexibility to control the hardware more efficiently as part of an overall system. As we continue to advance cars with software, you will see systems begin to consolidate into software-definable platforms with more connectivity both within and outside the vehicle. With this added connectivity and distributed intelligence within the vehicle, the opportunity to add value and increase revenue from software expands. The AI operating system model is architected to complement this migration. Focus not on hardware alone, but on collecting the best data for decision-making, adding features to add safety and performance for the consumer and driving profitability for the OEM. For example, in the future, a rain sensor may trigger a rain performance mode, or a camera may trigger a lidar to confirm an object. This distributed intelligence is key for what we consider a software-enabled vehicle. In a recent report, it was estimated that Tesla today makes 67% of its profits from these types of software-enabled features. While our products already have the adaptability to be definable across multiple applications using the same hardware, the real power in the future, where cars may be driven for 10 years, may be the ability to continue to adapt over time and update remotely using OTA, an acronym for over-the-air updates. As an example, as new vehicles increase software content, OEMs will be able to update software over the life of the vehicle, similar to how your phone gets updates today. Vehicles will be able to send and receive data, enabling them to continuously increase in value. These updates will allow new features and functionality, translating to improved safety and performance. As vehicles and infrastructure head towards over-the-air evolution, we believe our software defined sensor will be a key enabler of these new business models. In summary, we believe the power of AI's unique sensor platform is that it is intended to be a set of hardware components that can be manufactured, then configured for any high-value use case in the software. For instance, OEMs or Tier 1s could use the sensor's operating system to enable ADAS features that can be bundled for a range of consumer vehicles. The same operating system could be used by system integrators in the ITS or intelligent traffic systems market who are able to optimize the sensor for a pedestrian safety at intersections or forecasting traffic flow on toll roads. Trucking can leverage high performance, high reliability sensors designed for first mile, last mile or hub to hub applications. In the high demand rail and aviation markets, each sensor can be optimized for the extreme range and the resolution they require. So let's talk about execution and our progress around commercialization and scalability. There's no better place to start than our latest product, the Foresight M, which we intend to transfer to volume production later this year. I would now like to introduce Tom Fallon, Executive Vice President of Strategic Business Development at Samina, our Foresight manufacturing partner. Take it away, Tom.
Clyde
Thanks, Blair. Sanmina is one of the world's leading integrated manufacturing solutions provider. Headquartered in Silicon Valley with a global footprint, we have earned a reputation for innovation, reliability, and quality with a passion for customer success. It is important to understand that we only win when our partners win. So we are very selective in where and when we invest in new processes and emerging companies. Each year, with our customers, we bring about 3,000 new products to market, so we are approached by a lot of companies. We proactively choose to partner with the companies where we see a mutual alignment around ideas and processes. We also look for a well-defined market opportunity that is large and rapidly approaching. With AI, we found that alignment. We also found that AI has a compelling vision and business model. we believe AI's smart software-definable sensors will be a driving force in the automation of cars, infrastructure, and assets across many industries. At the core of our relationship with AI, there are three fundamental pillars we have found important to increasing the probability of success. First, AI decided early not to build a factory, but rather invest their time and resources in designing their systems for outsourced manufacturability with an eye toward optimizing efficiency and cost without compromising on industry-leading performance and reliability. Second, AI's innovative approach of aligning component suppliers with their reference system design Utilizing modular components sourced from proven automotive grade suppliers not only allows accelerated innovation, but also is a tremendous advantage in helping us to scale and harden our global supply chain. We believe this approach creates a strategic differentiation from others by optimizing time to market, volume, quality, and cost. Third, AI has transferred much of the system complexity from hardware to the software layer and its unique sensor-based operating system. We don't usually see companies make that leap until four or five generations of product release cycles. This allows one manufacturing line to produce the sensor hardware at scale, and software is used to customize the sensor per market or partner and to enable continuous enhancements in functionality over time. Most importantly, AI and Sanmina have worked as one team. From the beginning, we have leveraged each other's strengths to develop integrated design, manufacturing, and testing processes that will bring the AI Foresight LiDAR system to the market faster and with greater reliability and performance. Sanmina believes that what we make makes a difference. We are very proud of our partnership with AI. Back to you, Blair.
Blair
Thanks, Tom. Earlier this year, we mentioned the convergence of our components across markets and the focus on shared volumes and cost reduction to drive adoption. Tom also referenced our joint efforts to design for manufacturing, reliability, and the power of converged supply chains. One example of this collaboration we would like to share for the first time publicly is how this effort drove advancements in our MEMS components. custom designed and built around standard industry processes for manufacturability. The small dot in the center of the chip on this picture is our micro MEMS, significantly smaller, faster, and more adaptable than any we have seen in commercial production, proving that record-breaking performance indeed does come in small packages. Another concrete example of how AI and Semina are innovating together is our new joint calibration and testing facilities located on Semina's San Jose campus. It is a perfect complement to AI's indoor range in Dublin. Industrialization and reliability are at the core of any successful path to scale and highly regulated and mission-critical systems. This large, dedicated, state-of-the-art facility not only allows us to do environmental and performance testing, but also to bring customers and integration partners into an immersive and flexible testing environment. This jointly developed facility gives us tremendous flexibility in validating the performance of our Foresight sensors. Working with Semina and our end user customers, we have developed rigorous testing methodologies that help us fine tune the performance of our sensors in a wide variety of use cases and applications. You can see in our video our ability to quickly reconfigure the operation to run customer-driven tests this week, from small object detection at speed, a ride or down motorcycle scenario, intersection pedestrian safety, to much larger applications for acquisition and countermeasures in the aerospace and defense markets. In addition to our extensive in-house testing with Semina, we have extended domain-specific testing resources by partnering with some of the largest Tier 1 automotive suppliers in the world. In this process, we are exposed to their world-class processes, including environmental standards, product validation, functional safety, and performance benchmarking. This has led us to collaboratively working with some of the most influential and respected third-party testing groups in the world. We have also taken the unique step of releasing these results when appropriate to the public. As an example, we work closely with VSI, a leading independent researcher of active safety and automated vehicle technologies, to validate the performance of our LiDAR for ADAS applications. We do this at locations such as the American Center for Mobility, where we were able to test and independently verify the ability of our products to perform. In this VSI designed, produced, and verified testing scenario, we were able to detect very small objects, such as bricks at long range in inclement weather, while inside a tunnel. On top of that, we are demonstrating these capabilities with our lighter looking through windshield glass, opening yet another unique placement opportunity not available to most traditional time of flight lighters. We have progressed on plan and are executing on track and on time with our development, testing, and transition to production later this year. I will now turn things over to Bob Brown, our CFO, to discuss our financial update.
Tom
Thanks Blair, and good afternoon everyone. I'd like to discuss our financial performance for Q1 and our near-term outlook. Revenue in the first quarter of $1.1 million was up 229% over the first quarter of 2021. The top-line growth largely reflects an increase in development contract revenues as we complete work with key partners, as well as higher prototype sales relative to the prior year. As we've discussed previously, the sizable percentage of our revenue is driven by one of the largest Tier 1 automotive suppliers in the world, which is a strong validation of our technology and strategy. GAAP operating expenses of $24.5 million in the first quarter rose $14.1 million from the first quarter of last year. As we've continued down the path to commercial production over the last year, we've scaled our team and spending to support that progress, as well as to support the infrastructure required as a public company. Our non-GAAP operating expenses were $19.2 million in the first quarter, which excludes $5.3 million in stock-based compensation expense. Net loss was $24.9 million on a GAAP basis, and GAAP EPS was a loss of 16 cents. Net loss on a non-GAAP basis was $19.5 million in Q1, and non-GAAP EPS was a loss of 13 cents. Net cash used in operating activities for the quarter was $16 million, and our CapEx was less than $1 million. We'll continue to manage our cash carefully going forward, and our team is managing to a strict budget. The vast majority of our spending is focused on R&D, operations, and sales and marketing, with the goal of scaling our business as efficiently as possible. We exited the quarter with $144 million of cash, cash equivalents, and marketable securities on our balance sheet. When we include up to $125 million of potential proceeds from our common stock purchase agreement, we believe our total available liquidity of $269 million provides us with a sound financial base to execute on our strategy. We anticipate that we'll begin accessing the common stock purchase agreement this year. While we're on the balance sheet, I wanted to note that we adopted the new lease accounting standard ASC 842 in Q1. As a result, you'll notice increases in right of use assets and operating lease liabilities. These amounts are primarily related to our office lease obligations. It's exciting to see how we've grown over the last few years from an R&D focused entity into a commercial operation. We're starting to reap the benefits of our capital light strategy by focusing our time, effort and money on our core competencies and the activities that will extend our technological lead while getting our products to market faster. We're executing our plan to develop products for both the automotive and industrial segments based on the same revolutionary architecture. This is key because unlike most of our competitors, we don't need to develop different products for different applications. We will use one software-defined architecture for all applications across all end markets. We expect that this strategy will provide us with economies of scale and improve our margins as we grow the business. Relative to our near-term outlook, we expect revenues in the second quarter to be about $700,000 as we wind down prototype sales in preparation for the ramp of the commercial version of our Foresight M product in Q3. So combined with the Q1 performance, our revenue for the first half of the year in total should be slightly ahead of expectations. As we mentioned on our call last quarter, we expect to see revenue growth in the second half of this year as manufacturing of our commercial product starts to ramp at Sanmina. We expect that growth in the second half will enable us to deliver on our revenue goal of $4 million to $6 million for the full year. We continue to expect a non-GAAP net loss of approximately $100 million for 2022. I'm pleased with our team's performance in Q1, and we're tracking to our plan. We continue to execute well against our strategic milestones, and we look forward to sharing further progress against our financial, commercial, and technical objectives in the coming quarters. With that, I'll pass it back to Blair to wrap things up before we open the line for questions.
Blair
Thank you, Bob. Want to close as we always do with our talent and culture. We are fortunate that we continue to attract the brightest minds in the industry. This includes our global advisory board. We started our advisory board very early in our history, and it has been a valuable resource for us as we have built our business. We expect the latest additions to continue to be a vital part of AI. Let me introduce you to a few new members, Marcus Lipinski, Art Blanchford, and Dr. Erlrich Weinman. Marcus was most recently the managing director at Aptiv, a global automotive tier one supplier. Previously, he had been a leading executive at VW, MAN Trucks, and Daimler. He has a history of delivering digital innovation in both the automotive and software industries. Art has been an executive at VNIR, now Qualcomm, and Autoliv, two leading tier one suppliers, where he led global teams focusing on active safety solutions. Art brings a breadth of high-performance business, operations, and sales strategy to the team. And finally, Dr. Wyman has extensive knowledge in the OEM space, predominantly as global SVP at Harman and COO at Alpine Electronics, as well as a senior executive at BMW. We welcome you all to AI and look forward to working together. Culture is a powerful force, and I would like to share with you today an employee-driven initiative that we are very proud of. Over the last two years, we have partnered with Richard Branson and Virgin Galactic, Virgin Orbit, and Virgin Hyperloop to bring together our technologists and engineers to explore the future of transportation. Part of this exploration has been to look at how we invest in the future and share this knowledge with the next generation. One key element of this has been the development of the BLAST program. BLAST stands for Black Leaders in Aerospace Scholarship and Training. By providing mentoring and internships, BLAST aspires to change the funnel by creating a village with a network of support that helps Black students find connections and opportunities. This program has also changed AI. In the process of mentoring, we learn and we become inspired. I can honestly say this program has been a significant value add to our culture and to our effectiveness. We hope BLAST is also an example of how AI serves as a role model within our industry and as a leader in providing opportunities to talented minorities pursuing a career in engineering and technology. Looking forward, I want to first thank the team for all their hard work as we've been busy this quarter setting ourselves up for the rest of 2022 and setting the stage to scale in 2023. As we mentioned at the beginning of the call, the world is changing quickly. As a company, we are staying focused on the things we can control and leveraging tight relationships in the community of our employees and our partners. During this call, we discussed our software platform and how we are implementing disruptive intelligence sensing. We also talked about how we have made significant progress in delivering cost and scalability efficiencies by utilizing a capital light business model. Finally, I'd like to reiterate what Bob stated in our financial review. We are on track to deliver our 2022 guidance. I want to again thank everyone who joined the call today. Operator, let's move over to Q&A.
Bob
Thank you. We will now begin the question and answer session. To ask a question, you may press star, then 1 on your touchtone phone. If you are using a speakerphone, please pick up your handset before pressing the keys. If at any time your question has been addressed and you would like to withdraw your question, please press star, then 2. At this time, we will pause momentarily to assemble our roster. The first question comes from Suji De Silva with Roth Capital. Please go ahead.
Suji De Silva
Hi, Blair. Hi, Bob. Congratulations on the progress here. So you talked about, Blair, in the prepared remarks, the B sample transferring to Conti. I'm wondering what that entails in terms of it moving to Conti for you guys. And also, has Conti been addressing OEM RFPs and RFQs ahead of the B sample transfer, or is that kind of one of the things that's happened before that, those RFPs get addressed?
Blair
Sure. Thanks, Suji. Yes, the transfer of the B sample is a little bit misleading in the sense that, as we've said on earlier calls and you've heard from Conti directly, we've actually integrated our teams completely. We actually have engineers in residence at both units. So we've been working jointly from the beginning of the HRL design to now. When we use the term transfer, we talk about the transfer to Ingolstadt, where we have sample lines – in the U.S., and we're kicking up the sample lines in Germany, which, again, puts more of the emphasis on the testing and functional safety concerns that Conti is more responsible for. But we work on the product jointly today, and we'll continue until the product ships and beyond in support. And as far as customers go, yes, we've been working, I think we've said in other calls, and I don't know what the exact number is, so I'll give a range, but there's somewhere between 15 and 17 RFPs, RFQs that we've been jointly working on. As you know from your experience in the industry, people want to see engineering samples. They want to see B samples. They want to input in what they want to see before you close out the C sample. In addition, We spend a lot of time in R&D projects with the same, you know, say 25 to 27, both automotive and trucking OEMs, looking at the next generation and getting them attuned to what is possible out there. So our sales teams are tightly integrated. And in fact, we have functional twins all over the world, as well as an individual set of priorities by region. Okay, great. Both of the questions.
Suji De Silva
No, it did, Blair. Great job. And then so the – just so I understand, as we see auto OEMs commit to the LiDAR, to the AI LiDAR, how would that – how would we become aware of those wins? And what do you think the timeframe is from this point forward as to those sort of announcements, just awareness of those wins being secured?
Blair
Sure. I know there's been, you know, a lot of – I talk, at least there was last year, about, you know, FLOB, the forward-looking order book. And we talk about that in every executive meeting that we have. But for the automotive market, that is our partners, the tier ones we work with. Specifically, the one that is moving to manufacturing first is Continental. So they will make the decision on when they announce, as you know, the tradition. in the automotive industry for radar and lidar the other lighters that have been before this as well as cameras have been to wait closer to the sop i think that we'll be talking about it much sooner just because the uh the tradition has started to to morph as there's been more uh more talk out there um so continental will be in charge of uh of talking about those things for automotive in the industrial markets however Um, in the second half of the year, we will start as, uh, we are the, um, the direct sales arm. We will start to talk about those, um, um, wins much sooner.
Suji De Silva
Great. And I look forward to the industrial announcements because that's going to help drive second half revenue. As you said, um, one last question on the industrial side. I mean, great to have the Samina executive on to help with the, get some insight there. What, um, You know, what was it about the way you approached the technology, software, and hardware that allowed you to leverage high-volume proven components, which obviously is something Sandmeat will like to buy you guys a lot. Other LiDAR guys we know are using some more exotic components. So if you could talk about how the software is allowing you to kind of pull that off, that would help us understand some of your differentiations.
Blair
I think having spent enough time with them, I'll try to tell you what we've heard from them, was that we made a decision very early that we would move from being an R&D company to a high volume production company because we believe, although this has taken a while for LIDAR to come to commercialization, that the curve will be much faster. I've said in the past that It took 15 years really for extreme penetration for radar into almost everything. And it took about 11 years for cameras. We think that that curve is on track to be, you know, less than five or six years, which is at light speed. If you believe that, then you would want to get out of what R&D companies do, which is vertical integration, exotic technology. design and building your own factories where you have to spend time on tooling that you then would have to get rid of if there's an advancement in the technology. We decided early on that we would design for manufacturability and that we would use custom designs, but standard component processes. So one of the things we've been doing over the last three years is working with the largest component suppliers, tier two suppliers, as they say in automotive in the world and convincing them that they can build our designs using their standard components. which then brings down costs and increases volume across the different types of components. So that's when we talk about this, when we talk about our relationship with Semina, they really appreciate this as the final manufacturer because they're dealing with not only tier twos who already have standard processes, but tier twos that are already automotive grade compatible.
Suji De Silva
Okay, great. Thanks, Blair. Appreciate it. Thanks, guys.
Tom
Thanks, Suju.
Bob
The next question comes from Joseph Osha with Guggenheim Partners. Please go ahead.
Joseph Osha
Hello, everybody. Happy Friday. Thanks, Joe. You too. Yeah. A couple questions. First, you know, Bob, I'm just wondering if you can just talk us through a little bit of, you know, the cadence of the burn as you kind of work through the rest of the year here. And kind of where I'm headed with this is that given your relatively modest burn so far in the amount of cash and equivalents you've got, I'm surprised that you're tapping the equity facility this year. It seems like you've got some breathing room there that would not require you to do that. So I'm wondering if that's just kind of belt and suspenders being careful, Or the message there is that the burn is going to go up as the year progresses. Thanks. And I have a couple other ones.
Tom
Yeah, you bet. Yeah, we do expect the OPEX to go up a bit this year. So, you know, for modeling purposes, I'd assume, you know, the non-GAAP OPEX will probably increase on the order of about 15% per quarter sequentially. So I would probably build it out that way in terms of modeling. But to your point, yeah, the burden has been fairly modest. You know, we don't have to tap the common stock purchase agreement. We don't want to. You know, that said, you know, there may be some advantage to doing some dollar cost averaging over time and just using a fairly moderate basis. So, We've got 11 quarters left to utilize that facility. So if you do the math on that, we've got $125 million facility. So that's a little over $11 million per quarter. So it's not a huge amount by any means if you were to average it out over those 11 quarters. So we're expecting to obviously look at the volumes, look at market conditions, and then decide how much we'll actually do each quarter. But the idea is to be fairly moderate about it over time.
Blair
Yeah, and I think, Joe, you also hit on one of our cultural imperatives. which is there's a lot of ambiguity in the world. And if you take a look back to last year, we raised the largest pipe because we wanted to offset the possibility that redemptions would go up. And that actually did happen. We also actually executed the ELOC on the heels of our IPO probably a year before most people would have done it. And that was, again, I think credit to Bob's conservatism, we got a great partner and we got tremendously good terms on that. So I would say that this is more us taking a look at how to be prudent and how to be pragmatic. We will not, we don't intend to at least do anything that would be radical, surprising, or would be detrimental to investor value.
Joseph Osha
Okay, thank you. That's interesting. The second question relates to the ramp of foresight, right? And so obviously it's an early-stage product, but as you point out, you know, this is, you know, being done in a high-volume facility with commercially available products or inputs. So, you know, I would think that on that basis we begin to get it. fairly early on kind of line of sight as to what gross margin for that part of the business might look like. Can you lend us any insight into how we might think about that?
Tom
Yeah, we'll probably have more to say about that next quarter's call, Joe. So we're expecting to get some early production here in Q2. These will be more samples in Q2. So we'll get some initial volume, we think, coming out of Samena later this quarter. But then the rampage that really starts to happen in Q3. So we'll probably provide a little bit more color on that when we have our next earnings call.
Joseph Osha
which I believe is in August. If I ask that in a different way, I mean, understanding that perhaps you can't discuss the numbers per se now, but would it be logical to assume that this gets to some kind of acceptable gross margin reasonably quickly because of this combination of the capital light approach and focus on off-the-shelf inputs?
Tom
Yeah, we'll start making progress on gross margins, certainly, we think, because we've been selling prototypes up till now. So, you know, we're expecting some improvement over time. You know exactly what that's going to look like. You know, we'll, as I said, share more in the coming quarters. But I think for the year, we're probably still looking at negative gross margins overall for 2022 is probably the way to think about it. And then as we scale it up, you know, it'll improve in certainly the future years as we start to get more scale with it. It'll still be relatively modest volumes for 2022 as we ramp it up. And we're doing, you know, proof of concept deployments with customers. So we've got to get product in their hands and get them testing it before we get widespread deployment. So the volumes will still be fairly modest for 2022. So you don't get the full value out of the leverage yet. But we think we'll start to see that in 23 and beyond.
Blair
And I know you've done a lot of research on this, you know, dug into the models before. The way that I think about it is in the automotive markets, where we're getting a standard licensing fee, we know from the beginning of the contract without any risk, what our margins are going to be. In the industrial markets where we're selling direct, we have two variables. One is a positive variable where you don't have to wait for SOP, functional safety testing, and then SOP. We can go from a pilot in three to six months to more of a production rollout in those markets. So that's one thing is we've got to get through the pilots to figure out how it's going to be deployed. The second piece, as we've talked about before, is that we believe that in the ADAS market, there's not a lot of room for add-on software, as I think many people have alluded they may be able to get to. Our opinion is that once these contracts are set, there's not a lot of room for add-on. We're happy to get our licensing fee, which is a different model. But in the industrial market, we're also looking at building on top of that operating system. It's much easier for us now that we have an operating system on the sensor to build out custom software applications, or you'd look at it in your iPhone as apps. So we don't know how that's going to play out and how fast that will play out. that could as well impact product margins, you know, over time. But I would, you know, as we've been conservative and, again, pragmatic in the past, you know, this is the early stages of rollout in this year. And I would agree with Bob, you know, we're not expecting to optimize on gross margins. We're out there to get customers to use the product, show value, and build opportunities to do production rollouts.
Joseph Osha
Okay. Yeah, thanks. I understand that's a useful color. And then the third and final question would be, you know, as regards to this kind of platform vision you articulated. I can't remember which slide it was showing the two different sort of vehicle platform visions, which is something that I hardly agree with and, you know, you and I have spoken about. But it does imply that at some point there needs to be a fairly high level level of engagement with, you know, whatever large, you know, semiconductor company you think is providing that big, powerful compute platform. How do you think about that? You know, could we see you make an announcement or form some kind of venture with, you know, an Intel or Broadcom or Qualcomm or something like that?
Blair
I think you're 100% right on. I think that at the end of the day, as we've talked about in the past, this is ultimately value. Hardware margins are tightened over time. As we get more efficient, software verticalizes. And ultimately, the data into the network is where a lot of value is added in decision-making. In autonomy, it's even more specific. You can't get to autonomy without good data to make decisions, and I believe that anyone in this space will have to have a data platform strategy with these compute platforms because I think they're going to enable us. Now, EVs are going to help them enable us, but they're already – there's over – In some cars, over 100 ECUs today. Even if we move to the platform, we'll simplify it, but there's still going to be a lot that has to do with data optimization and compute power. And you can expect us and I think any of the other leading companies in our space who are providing data into that system to have to work closely with them to make the system work. So we think we have an advantage in that we have an operating system on The sensor, which has two way communication, you know, we've, it's been gratifying and talking to these types of people that they appreciate that they see it as a, you know, as a network addition. But at the end of the day, there'll be multiple types of implementations, whether it's point sensors passively sending information. or active sensors that are pre-processing or integrating information across it. But I think your thesis is correct, is that over the next, you know, two to five years, we're building, you know, new compute platforms. They just happen to, in some cases, you know, drive around and drop off the kids at soccer.
Joseph Osha
Right. Right. Thanks. I'll jump back in queue.
Bob
The next question comes from Hans Chang with DA Davidson. Please go ahead.
Hans Chang
Hi, Blair. Thank you for taking my question. So I have a few questions. So first, just from your perspective, are there any potential technology hurdles or capacity constraints on the supply chain? For example, maybe it's laser, receiver, scanner, et cetera, just Just want to get a sense, like, if you have to identify maybe one or two factors, like, potentially you think maybe you see more challenges, like, regarding, like, we want to, I mean, proceed to that volume production on the auto side. Yeah, just wanted to hear any comments.
Blair
And I appreciate the question. You actually brought this up last time, and I thought it was appropriate. It's even more appropriate today. I'll answer two ways. One is because we have the same thought you do. We've spent a lot of time working the key components we talk about the fact that we have you know really in our system is a you know you can drive cost down in three ways right one is simplicity of system design the second is in adding new materials innovations and the third is in volume production right so when you look at our system design there's really only four components and some people would say three i mean you have a laser that that sends out the energy you have a scanner that um and in fact you know in our case we have an adaptive scanner that actually interrogates an environment. And then you have a receiver which receives information. We've spent a tremendous amount of time trying to secure those supply chains and make sure that we wouldn't have any disruption in the short term. But what I would say is that... It's amazing. When you're building a system, sometimes it's only one small component, which you didn't consider a large piece that can trip you up. So I'd say most of our, well, we've spent a lot of time on the big components. It's amazing. Every couple of weeks, you'll find out that some small piece of the puzzle is built in Shanghai and then it's closed down and we have to go to our secondary source. So today we feel good about, about what's happening with our supply. Actually, good is a relative term. We feel confident that we've done what we can do for our supply chain. But I'd say that this isn't going away over the next couple of years. And it's forced us to take a much wider lens of what is inventory management. And when you get to the full assembly, you know, everything has to be there at the same time. So there's the short version of that. If I wasn't getting paid by the word, Bob, Bob pays me by the word, but the short answer is we think we have a good focus on the main components. But I tell you, we're still shaking out every, you know, while we haven't moved into the next phase, we're still finding things that we hadn't expected and we're, and we're moving stuff around to, to fix it. And I think that's going to be indicative of many of our peers and just many of the people, you know, in tech in general. But it's a great question.
Hans Chang
Got it, got it. Yeah, yeah, that's fair. So next question, just, yeah, interestingly, just you point out, you highlight your men's technology in the slide. So actually, just kind of curiously, can you elaborate more on these? I think it is men's scanner. And what's the difference the differentiators for your technology and maybe you can address like the point of cloud density or the scan rate, something like that. And it's just curious like how these, how your technology. All right.
Blair
So you got me on one of my passionate subjects. So I will try to actually, you know, it's difficult being Italian to be succinct, but I'm going to try. So, Louie's original design came from the top down, he was designing a network software to pull in data, and it just happened that in a lot of cases you need hardware to actually acquire that data. So in Louie's original model, he wanted to be have modular hardware so as innovation happen you can plug and play the wavelength. can change the, anything in the model can change because hardware changes over time. And that's something that he and I both saw over the years in telecommunications and in the military. But just as important as that insight, I think, was Louie's insight for adaptability. So the bi-static design of our product where you separate the send and receive versus having it be coaxial and they're hardwired together is actually a legacy of this same design, which is let's design from the network information model down. So since we separate versus putting the two together, we actually have an extremely flexible send component. So our MEMS are, in some cases, 250 times smaller than many of what other people call MEMS. We call them micro MOEMs or micro MEMS. So they're very, very small. And they are not actually hardwired to the received, which means that in most cases, when someone says we can see longer distances than anyone in the world, and how could that be? The laws of physics haven't changed. The laws of physics may not have changed, but the laws of mechanical orientation have are still open. And so we don't have to wait for the light to come back with our receiver when we send it out. So we can move very quickly. That's how we can go longer distances. And we can also get greater density because if you take a look, I am not an engineer, although I've spent most of my career working with engineers. So my apologies up front for the simplicity of this answer is, The size of the MEMS is so small that the inertia allows us to move in ways that no one else could move their MEMS. A single pair of MEMS, we actually, I'm not allowed to say exactly how fast we move them, but as you saw us track a bullet, which was thought to be impossible, Everywhere in the world, we're well over 20,000 hertz, right? So we can go anywhere from 10 hertz to over 20,000 hertz in speed. We can go, and we're going to be showing in August, I think we'll be showing some implementations of this operating system. We've already announced that we can go over a kilometer, and we've, I think in some cases, alluded much further than that. And in density, density is determined by how close the actually points hit. and how you actually segment and acquire an object so that you know what the object is. With the MEMS that we have, we can change the pattern on the fly inside a frame. So it all depends on what kind of density we want. The trade-off is always speed, distance, and density. And what we do with our adaptive system is when we place it on a truck that's higher and we have a certain use case, we will optimize every shot for that unique packaging placement application, right? And so I can tell you the extremes of what we can do. We think we've set the world record in every, in speed, in density, and in distance. But again, what really matters in a network is do you get the information that a computer needs to make a decision that's going to be better than a human? And that's the definition of automation. So the micro MEMS is an absolutely important part, but so is the fact that our receiver is separated and that actually has a capability to track in a very, very unique way. In some ways, a lot like a CMOS camera. So I hope that answered your question. It wasn't too wandering in it, but the ability to have this kind of MEMS, this small kind of MEMS that we don't believe anyone else in the world has, is a huge advantage in adaptability and intelligence.
Hans Chang
Yeah, yeah, that's definitely very helpful and A quick follow-up. So is the manufacturing, the process, technology on the microfinance mature or is it kind of also new?
Blair
So as we've said, we may have custom designs, but we actually try to stay within standard processes to help our suppliers and to drive down costs. So in any way we can, we always drive to existing standard processes for manufacturing.
Hans Chang
Got it, got it. And last question, just regarding the equity purchase agreement. So I know you just mentioned this probably doesn't moderately do these over time, but do you have any kind of guideline? Do you have any price flow like you want to do these and then Or because I think for this year, you probably don't need the liquidity from this. But just kind of curious how you think about the strategy or tactic here regarding how you want to do it. And do you have any, do you have a cap like, oh, I won't do more than X amount or something like that? Yeah, just kind of any comments.
Tom
Yeah, we do internally have some of those metrics, nothing that we're going to share at this point. But we're going to be very thoughtful, as Blair said, about how we use it. So we're not going to want to put undue pressure on the stock as we use it, of course. So we're going to be very, very thoughtful about how we approach it. So As we said, it's about $11 million per quarter, over 11 quarters, if you average it out. And some quarters will certainly be below that, and some quarters will might be above that. And the differentiation there is really going to come from both the trading volume of our stock and the market overall, as well as general market conditions. So that's how we're going to approach it. So we're not going to have a specific table that we're going to lay out for folks on how we're going to use it. So we expect to be nimble with it and thoughtful about how we use it.
Blair
Right. And, you know, I have to invoke my father every time I get a chance to. But look, at the end of the day, there is some ambiguity. But the answer to ambiguity is probably not going to be certainty. It will be trust. And what Bob's saying to you is. Our, you know, philosophy is do no harm, but also be pragmatic and conservative and smart about how to run a business that we, you know, believe will be here for a long time. So we, you know, that's our commitment is we're going to be smart and you, I hope you won't be surprised by anything you do. I hope you'll look back and say that was thoughtful. Got it. Got it. Okay. Thank you guys.
Tom
Great. Thank you, Hans.
Bob
The next question comes from Andre Shepard with Cantor Fitzgerald. Please go ahead.
Cantor Fitzgerald
Good afternoon, guys, and congrats on the quarter, and thanks for squeezing me in here. I know we're about time. Most of the good questions have been asked already, but maybe just to take a step back, I was curious if you could remind us again on the strategy in terms of both the short-term and the long-term, and by that I mean In terms of your target markets, do you anticipate to kind of prioritize the automobile sector, which will ramp up over the next few years, or is the strategy maybe in the short term to pursue and target some non-automobile markets while the automobile ramps up?
Blair
Thank you. The answer is yes. So look, I know we spent a little bit of time talking the other day, and I appreciate the question because I think a lot of people want to actually – many of our peers are focused in one place or another. If you think about – how we're focused we're focused on network information where we can actually optimize our product using software using the same manufacturing lines and the same hardware components so what we believe is something that we had to do and why this year is so critical for us is we believe that we had to have the components and the manufacturing capabilities ready in both markets because in the automotive market Well, your point is well taken. You will not see the SOPs for a few years. Up to this point, there's been... pilots, but you will see over the next two years, most of the OEMs actually committed, right? So we have to have a product that they can look at with Conti and that they can trust and say, this is an automotive grade, this has reliability, this has the right cost profile, you have manufacturing set up. I mean, most people don't realize even to get to a million units, it's about $150 million line with tooling and $250 million in working capital, and then warranty and liability. So for us, our model is we don't have any of those costs, right? We basically get a royalty on every unit that goes off, but we have very tight partners in tier ones and continental who will handle that. So we are in the automotive market today. I think someone earlier asked how many RFPs and RFQs are out there. Almost every company is actually investigating over the next two years to committing to programs. Now, they're spread out over highway autopilot, hub-dub trucking, and maybe some traffic control, but they are all engaged that LIDAR is a way for them to actually add value and therefore make some more money. Now, on the other side, it's a tale of a very, very different market. When you take a look at some of the industrial markets, they have actually used LIDAR. It's a little bit bimodal. They've used it in mapping in the past, and they've also used it for highly specialized applications. So, for instance, in a mine in the dark, trying to see through dust and trying to be able to be more efficient and push throughput. Now, in those markets, what's interesting is the turn cycles are faster. You don't have to wait for functional safety. They truly appreciate that you're using automotive-grade components because the industrial market shock vibe reliability has been the major issue with the LiDAR systems in the past. But they're actually trying to focus on how to actually deploy systems that make money within the next year. So if you look at ITS, the largest amount of money that's ever been put in a transportation bill is now embedded in smart cities and ITS. So we believe those markets are going to happen and they're going to happen sooner. And I think as Suji maybe referenced, We need to be ready with Samina so that we can roll full, complete products off the line that could be implemented and be in use for three to five years. Because in many of the ITS applications, the installation is just as expensive as the actual sensor itself. So what they appreciate about us is that our automotive focus has led us to high-quality reliability. And the ability to have an operating system on the sensor allows you to upgrade infrastructure over time. And almost every intersection out there today in a major city actually already has some camera and radar. And the ability to actually look at data holistically and how we can use that is appreciated. So when I say yes, it's good news, bad news, as the Chinese curse says, who's to say? We believe that both markets are engaging right now. We're just going to market a different way. In automotive, we're going through large tier ones like Continental, and we're in the processes. And in industrial, we're working with systems integrators, and we're helping, and we're selling side-by-side direct, a fully manufactured product through Sanino. Did that make sense?
Cantor Fitzgerald
Yeah, no, that's wonderful, Ashley. I appreciate all of that insight. Very helpful. Maybe one last quick follow-up for me, and this is maybe more directed towards Bob, but looking at the liquidity, so $144 million in cash plus the $125 million in the CSPA, is the thought now that that should be sufficient to Again, to go through that ramp up period in terms of being fully funded or do you maybe anticipate additional capital raises in the next few years? Thank you.
Tom
Yeah, I think for now we feel good about where we are in liquidity and where we are with our plan. We're not going to give long-term projections on the call today. So we're going to stick with just our annual guidance and updating that. But we feel good about where we are, as you said, from a liquidity position. We've got quite a bit of cash on the balance sheet and access to the common stock purchase agreement. So we feel that puts us in a very good position today. And as we said, we're going to be thoughtful and careful about how we deploy our OpEx going forward and also how we use that common stock purchase agreement to access that additional liquidity as we need it. But we feel very good that we've got a sound liquidity base to execute the strategy from. So for now, that's what we feel like we need.
Blair
And Bob's not letting me spend any money, and he keeps cutting me out from being able to – pay for lunch. So he's got a very, very focused strategy on liquidity.
Cantor Fitzgerald
That's excellent. Well, thank you so much. Congrats on the quarter. I'll pass it on. Thank you.
Tom
Great. Thanks, Andres.
Bob
Our next question comes from John Roy with Water Tower Research. Please go ahead.
John Roy
Thank you. So, Blair, obviously, you've been talking a lot about two different markets. And you've been building this product, you know, you really want to make it reliable, scalable, industrialized. And it seems like you expect some cross-pollination from the two manufacturers. Can you go into how you expect those two to work together or not? And is this part of your business model differentiation? Will you be able to leverage that?
Blair
Sure. Thanks. We touched on this very slightly the last call, but what we are doing right now, we had originally had a sequential product rollout where we were rolling out the industrial products, and then we were rolling out in 2022 the automotive products, and then we were coming with a refresh where we converged both products in 2023. the last call we announced that we would be accelerating that and beginning to merge the product. So to your question, what that means is The software, the operating system is being built out and hardened so that it can handle both markets and it can be triggered by individual sensors in each market. Each market has their own individual sensors that they depend on. The second piece is that I think we're, I don't want to misquote it because I don't have exact number, but I think we're over 80% of the components are the same. in this set of releases, which again, reduces our complexity, increases our reliability, simplifies our design, and ultimately we believe should actually reduce costs because you're using the same components and increasing volume on those components. It may even be higher than that, but I'd say conservatively, it's about 80%. When you looked at the difference in our presentation, the size of the products, and then you harken back to looking at that tiny little MEMS and a tiny little receiver on a chip, the difference in size is really how we've optimized the boards, right? There's a lot of air in those products because in the industrial space, they actually don't mind having a little bit more size, right? Whereas in the automotive product, when you're trying to package it in the grill or on the roof or behind the windshield, size does matter. And so that's really what we believe is the key to our business model is that it was designed from the beginning to be software operating system focused so that we can, I think Joe brought it up, so we can work better with the compute platforms and that we can actually use the same hardware components across multiple markets. And as we said in the Samina piece, use one manufacturing line and then optimize and customize per market in the software.
John Roy
Now, that's really helpful. The 80% number is interesting. Now, also, you've certainly talked a lot about software dependability, the on-sensor OS, etc., We're starting to hear others use the software-definable LIDAR term. Maybe you could just give us a little bit of differentiation between what you mean by that and what maybe others might mean by that. How is your software going to really be that different? I understand it controls the hardware. I understand you're bi-static. But maybe if you can give us more of a layman's explanation as to, okay, this is what it really means to the end user.
Blair
Sure. And, you know, again, again, It would be hubris for me to get inside other people's minds and pretend that I understand exactly what they're saying. What I would proposition is that if you're a hardware-focused product, you use software-defined ability sometimes to do configurations. And so you can change some small things in the hardware at design or implementation. There's a few levers, not a lot. What we mean by software definability and why we use the term operating system on a sensor is that we actually built from the software down so that every single component is individually controlled. So I can change the way the laser works without touching the receiver. Now, why would that be important? Well, I can change the field of view. If in a certain application or a certain mounting situation, I don't have to put in a different piece of hardware to get a larger field of view or a smaller field of view. I can change it in the software. Or I may be able to actually change the way that I do density. So while I'm actually scanning across, I may actually decide that we'll actually acquire objects, which means putting more points on them within the same frame. to articulate them and pass that on to the external perception engines. Or I may decide to take an input from an outside sensor, which if it's raining, maybe I'm going to push it from two returns to four returns. Maybe I'll be at six returns, which means I'll push through obscurance and take the returns after they pass through. All of those are attributes of a software-defined ability that you need an operating system to do. They're not configuration. They're actually customization and optimization with the ability to have two-way communication between different systems. So I think that we all realize in our industry that humans are very good at intelligence scanning when they're moving. That's why 92% of accidents are caused by distraction, not because humans are not good at scanning their environment. But what we have to do, I believe, to take humans out, which is what really automation is, is we need to be 10 times better than a human. And that doesn't just mean not getting distracted. It means being to intelligently scan better than they can. And that means trading off temporal scanning and spatial scanning in the same frame. If you can do that in a functionally safe way, you have added a tremendous amount of value. Consumers will love it. OEMs will love it. And if you've already bought an asset in the industrial space, your safety and ROI go up overnight. So we always have to look through to the end. We're not building technology for technology's sake. We're building technology that can acquire data to make decisions. And that's why, again, we are software-definable, but we also are software-definable companies. with an operating system. And, you know, that's, you know, that's the track we're taking. There will be, and I'll finish with where I started again. There will be multiple types of LiDAR systems in the world, just like there's multiple cameras and there's multiple radar systems. You know, our goal is to build intelligent LiDAR systems and they have their niche and we believe they'll have a great value. And the key for us is getting through this year so that we can start to scale and get it in customers' hands so that next year you'll be asking us very, very customer-specific questions because you'll have the feedback.
John Roy
Great. Thank you so much.
Blair
Thanks.
Tom
Thanks, John. All right. I think that wraps up our Q&A session. So, Operator, I think we're going to end the call at this point. Thank you all for joining us, and we hope you all have a great weekend. Thanks so much.
Bob
Thank you. The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.
Disclaimer