Ladies and gentlemen, thank you for standing by, and welcome to Ambarella's Q1 Fiscal Year 2023 Earnings Conference Call. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question and answer session. To ask a question during this session, you will need to press star, then one on your telephone. Please be advised that today's conference is being recorded. If you require any further assistance, please press star, then zero. I would now like to turn the conference over to your speaker for today, Louis Gerhardy. You may begin. Thank you, Tawanda. Good afternoon and thank you for joining our First Quarter Fiscal Year 2023 financial Results conference call. On the call with me today is Dr. Fermi Wang, President and CEO, and Brian White, CFO. The purpose of today's call is to provide you with information regarding the results for our first quarter for Fiscal Year 2023. The discussion today and the responses to your questions will contain forward-looking statements regarding our projected financial results, financial prospects, market growth, and demand for our solutions, among other things. These statements are subject to risks, uncertainties, and assumptions. Should any of these risks, or uncertainties materialize, or should our assumptions prove to be incorrect, our actual results could differ materially from these forward-looking statements. We're under no obligation to update these statements. These risks, uncertainties, and assumptions, as well as other information on potential risk factors that could affect our financial results are more fully described in the documents we file with the SEC, including the annual report on Form 10-K that we filed on April 1, 2022 for Fiscal Year 2022, ending January 31st, 2022. Access to our first quarter Fiscal 2023 results, press release, transcripts, historical results, SEC filings, and a replay of today's call is on the investor relations portion of our website. Fermi will now provide a business update for the quarter, and Brian will review the financial results and outlook, and will be available for your questions. Okay. Thank you, Louis, and good afternoon. Thank you for joining our call today. During our first quarter, we announced the passing of Casey Eichler who had been our CFO since 2018. Casey's positive demeanor was an inspiration to many of us and we are thankful to have experienced his leadership. On March 28th, we announced the appointment of Brian White as Ambarella's CFO. After two months, we're excited about the leadership and experience Brian brings as we scale the company for the significant opportunities ahead. Our Q1 revenue was essentially as expected, f lat sequentially and up 29% year- over- year. CV revenue grew significantly, both sequentially and year- over- year, representing about 40% of total revenue, and our video processor business declined about 20% sequentially. Our blended average selling price continued to rise. During Q1, the pandemic flare-up in China and the resulting lockdown disrupted customer production schedules and orders placed on us, as well as logistics in the greater Asia supply chain. A majority of our customers' products are manufactured in this region and are subject to impacts from China-related supply chain disruptions. We are seeing a similar degree of impact across both our automotive and IoT businesses further complicating the pre-existing kitting issues we have discussed before. Last quarter, we've reported a supply constraint for our 40-nanometer video processor SoCs. At this time, we continue to expect the Q2 impact to remain at about $5 million. We are now confident we will see a Q3 and Q4 improvement in supply of a 40-nm video processor SoCs. As previously noted, the supply of our computer vision SoCs has not been impacted. We are very excited to announce in late May, we received first silicon for CV3, our first central domain controller processor, and we have successfully brought up the key functional blocks on this 10+ billion transistor SoC. There's more bring up work to be completed, but we are confident that this first rev will be sampled over the summer to key customers, and it will demonstrate significant performance and power leadership. As a reminder, CV3 will be a family of SoCs that we expect to command a selling price between 5x to 20x our current corporate ASP. We remain focused on capturing the significant revenue opportunities in front of us. New 5-nm products are on schedule with the CV3 close to sampling, and the CV5, our first 5-nm SoC, is expected to commence mass production in the second half of the year. We continue to expect our CV revenue to be about 45% of revenue this year and this richer mix is expected to drive our blended average selling price higher. Interest and activity around our products and technology remains very strong. I will now provide some examples of our market development activity. In the automotive market during the last quarter, Ambarella announced its success in sensing applications, an entirely new market for a company. In applications such as ADAS and driver monitoring, as well as new viewing markets like electronic mirrors. At the Japanese OEM Honda, two of their joint venture companies introduced EV models using our H32AQ SoC for multifunction electronic mirrors plus drive recorders. Dongfeng Honda's e:NS1 EV and the GAC Honda's e:NP1 EV both use this solution. Also during the quarter, Isuzu, a Japanese OEM, introduced through a joint venture company, its new MU-X SUV. Based on our CV22AQ SoC, the SUV enables an intelligent driving assistance system utilizing a front-facing camera supporting lane departure warning. We remain optimistic about the vehicle in-cabin monitor system opportunity for us. In June, we expect Geely to announce a new passenger vehicle utilizing our CV28AQ for driver monitoring. Dongfeng, one of the largest commercial truck OEMs in the world shipped its new DF760 truck utilizing a single CV22AQ SoC to support the multiple cameras and the multiple functions for driver monitoring around vehicle monitoring, and blind spot detection. In AIoT market, our smart home business, which we have referred to as CV wave two, we are pleased to announce Vivint as another major CV customer. In May, Vivint announced four new products, including outdoor, indoor, and doorbell cameras, and a spotlight. All of the cameras implement our CV SoC to execute Vivint's intelligent AI algorithms for a variety of people and package detection and the tracking functions. The spotlight uses the best detection algorithm of the outdoor camera to illuminate intruders and follow them as they move around the property. In the IoT market, the largest portion of CV revenue has so far been realized from new product cycles in our enterprise and the public security camera business where the trend continues with the vast majority of customers' design activities involving our CV SoCs. During the ISC West Security Exhibition in March, almost every major security camera company was demonstrating new products based on our CV SoCs. There were also many public demonstration at ISC West of companies entering the access control market with system based on our reference designs. The access control market is a great example of how our CV portfolio allow us to reach entirely new markets we did not serve before and a ccess control is one of the areas where we are showing encouraging early customer wins. In access control, Motorola Solutions announced the new Openpath Pro series video intercom readers based on our CV25 SoC. The reader combines video, audio, and enterprise call routing. Motorola has made a number of acquisitions in the last few years to leverage its IoT camera expertise into new verticals, and Openpath is one example. We are proud to report that most of the camera companies acquired by Motorola are using Ambarella SoCs and we're eager to help them grow their business. In access control based on our CV22, RealNetworks announced SAFR SCAN, a touchless biometric system with anti-spoofing based on the fusion of a structured light and the cameras to ensure the most accurate results. During the show, targeting the enterprise and public markets, Hanwha Techwin announced two new AI-powered multi-sensor camera based on our CV2 SoCs and featuring deep learning-based object detection and classification. In the enterprise and public IoT market, i-PRO, a full part of Panasonic, introduced its new multi-sensor S-Series camera, which is based on our CV2 SoC, offering deep learning intelligence at the edge with pre-installed AI applications and the ability to add third-party applications. In the public safety fleet market, i-PRO also launched a 4K panoramic front camera based on CV22 with the form factors and thermal budget to be mounted behind the rear view mirrors on the windshields. Factory automation is another greenfield market opportunity, and during the quarter, iRAYPLE introduced two products based on our CV2 AI SoC, a stereo 3D factory automation camera with precise measurement capability, and a code reader with high resolution and deep learning abilities. In other IoT market, Insta360 introduced the ONE RS action camera. The innovative consumer camera is based on our H22 SoC and can shoot 4K 60 video, take 48-MP photos, and includes AI-driven editing to make stitching 360-degree footage simple. Recapping this announcement, a majority of the projects I just described are CV-based. In this product, our state-of-the-art video processing expertise is leveraging into new applications where the camera, instead of just enabling great human viewing experience, is sensing, collecting data, and then making decisions. A lot of incremental and specialized processing is needed to do this most efficiently, and our high bandwidth CV4 AI SoC provide this for an increasingly diverse set of IoT sensing applications. About half of these products I described represent the new product cycles in existing markets, and the other half represent entirely new markets for us, such as automotive sensing, factory automation, and access control. We are in the midst of a very strong shift in design activity to our CV SoCs. Long-term, we expect to earn a higher ASP with CV products and the combinations of higher ASP and unit volume is expected to drive a premium growth. We are confident that our strategy addresses the mega trends for security, safety, and automation enabling our customers to innovate and transform their own business. Our rapidly expanding AI product portfolio serves the diverse secular growth opportunities emanating from these trends. Now, Brian will review our financials. Thanks, Fermi. I'll review the financial highlights for our Fiscal Q1 and provide a financial outlook for our second quarter of Fiscal Year 2023, ending on July 31st, 2022. I'll be discussing non-GAAP results and ask that you refer to today's press release for a detailed reconciliation of GAAP to non-GAAP results. For non-GAAP reporting, we have eliminated stock-based compensation expense and acquisition-related costs adjusted for the impact of taxes. Revenue for Fiscal Q1 was $90.3 million, in line with the midpoint of our prior guidance, flat to the prior quarter, and up 29% year-over-year. As expected, a sequential increase in IoT revenue offset a decline in automotive. Both IoT and automotive were up strongly on a year-over-year basis. Non-GAAP gross margin for Q1 was 63.8%, slightly ahead of the midpoint of our prior guidance range. Non-GAAP operating expense for Fiscal Q1 was $39.8 million down $500,000 from the prior quarter. Non-GAAP operating expense was $2.2 million below the midpoint of our prior guidance driven by the timing of new product development activities. Our non-GAAP tax provision was $900,000 or 4.8% of pre-tax income, and we reported non-GAAP net income of $17.1 million or $0.44 per diluted share. Now I'll turn to our balance sheet and cash flow. Cash increased $30 million to $201 million driven by strong operating cash flow of $34 million. Fiscal Q1 cash flow was aided by decreases in both inventory and accounts receivable. Inventory decreased $4 million from 128 days to 117 days, and accounts receivable decreased $16 million from 45 days to 28 days. The substantial decrease in accounts receivable was attributable to a front-end skewed revenue profile in the quarter. We had three logistics and ODM companies represent 10% or more of our revenue in Q1. WT Microelectronics, a fulfillment partner in Taiwan that ships to multiple customers in Asia came in at 57% of revenue. Chicony, an ODM who manufactures for multiple IoT customers, was 11%. Hakuto, a logistics partner who primarily supplies multiple automotive customers in Japan, was about 10% of revenue. I'll now discuss the outlook for the second quarter of Fiscal Year 2023. As Fermi described, the external environment remains complex and dynamic. The supply chain already stressed with persistent COVID issues is now also facing the rolling pandemic impacts in China. Our guidance, to the best of our knowledge, contemplates these challenges. We estimate our Q2 revenue to be in the $78 million to $82 million range. We're down approximately 11% sequentially at the midpoint. We estimate Q2 non-GAAP gross margin to be between 63% and 64% relatively flat to the prior quarter. We expect non-GAAP OpEx in the second quarter to be in the range of $42 million to $45 million. The sequential increase in OpEx is driven by the beginning of an advanced 5-nm automotive grade CV SoC development project. Our Fiscal Q2 forecast for Ambarella's non-GAAP tax rate is 4% to 6%, and we estimate our diluted share count to be approximately 38.7 million shares. Ambarella will be participating on June 1st in Craig-Hallum Virtual Investor Conference, June 2nd at Cowen's TMT Conference, June 8 at Bank of America's TMT Conference, and on June 9th at Rosenblatt's Virtual AI Scaling Conference. Please contact us for more details. Thank you for joining our call today. With that, I'll turn the call over to the operator for questions. Thank you. Ladies and gentlemen, as a reminder, to ask the question, you will need to press star one on your telephone. To withdraw your question, press the pound key. Again, that's star one to ask the question. We ask that you limit yourself to one question and one follow-up. Please stand by while we compile the Q&A process. Our first question comes from the line of Joe Moore with Morgan Stanley. Your line is open. Great. Thank you. I wonder if you guys could talk about the issues in China in the coming quarter. Ca n you tell how much of it is, you know, is there any demand-side issues in there? Is it all supply side? Y ou mentioned it's broad across multiple customers. Is it sort of, you know, Chinese customers manufacturing in China and multinationals? Just any kind of more color you could give us on what the dynamics are of the challenges there. Hi, Joe, this is Fermi. What we saw was that at the end of March and early April, when we start seeing China's lockdown in Shenzhen and Shanghai, almost at the same time we start seeing our global customer base start pushing out their Q2 demand. We believe the reason that were given was that with the kitting issue persists and also that lockdown make the kitting issue a lot worse for our global customer base, particularly for those who manufacture in China or in greater Asia locations, and they were impacted the most. That's the scale that we are seeing. In terms of demand, you know, at this point, particularly outside China, I would say that most of our customers are still telling us they can sell almost everything they can build. I think the demand at least outside China is pretty solid in my opinion. We believe we will continue to monitor this progress through this lockdown situation. We all heard that the Shanghai lockdown will probably expire on June 1st. We are looking forward to see any updates on the market and from our customer point of view. Okay. That's helpful. Thank you. In terms of the incomplete kitting issue, is that an issue that people aren't ordering Ambarella parts because they're waiting to get other things? Is that an issue that they have inventory of Ambarella parts because they can't get those other things? Can you tell how much of this is going to be an inventory burn off that's required versus just, you know, a pent-up demand? Well, I think a portion of that is definitely there's some kind of inventory sitting in the channel. I do believe this lockdown is really when the customer find out that they really cannot allocate enough other parts, particularly like power, PMIC, the Wi-Fi and in particular microcontroller for automotive. I think that they decide to push out their Q2 demand to future quarters. I think that inventory situation is like what we discussed before, but this time, I think that a lot of pushouts is based on the kitting issue and particularly the new kitting issue, which is amplified by this lockdown situation. Great. Thank you. Thank you. Our next question comes from the line of Ross Seymore with Deutsche Bank. Your line is open. Hi, guys. Thanks for letting me ask a question. Brian, welcome to Ambarella. I just wanted to ask about the duration side of this equation. I know lockdowns and what COVID is going to do and China's going to do is kind of beyond my pay grade, and maybe you guys don't have great visibility into that. What's your assumption beyond the July quarter, whether it be the Samsung side getting better, as I think you alluded to, Fermi, or the persistence of the lockdown headwinds? Well, I think the major impact from the last quarter is the lockdown headwinds. Until even today, we don't have visibility, you know, how fast this recovery will look like. You mentioned the Samsung situation. Samsung 40 nm, we still expect the $5 million impact to us in July quarter. We expect that as Q3, Q4, the 40 nm supply will improve. I think this time we are really talking about, you know, this China lockdown causing a lot of kitting issues for our customers. As a follow-up question, to get into the model, you guys are now just having the two segments, the IoT and the automotive side of things. Can you give us any color about the size that they were in the quarter, either as percentage of sales, what they did sequentially? I know you said up and down, but any numbers around that? Is there a difference in the guidance for the July quarter between the two segments? In Q1, automotive represented about 25% of our revenue with IoT at 75%. As we look into Q2, we expect both of those segments to be down approximately equally in percentage terms. Both down in the double- digits sequentially. Got it. Thank you. Thank you. Our next question comes from the line of Quinn Bolton with Needham & Company. Your line is open. Welcome, Brian. For me or Brian, just, you know, it seems like it's a dynamic environment out there with the rolling lockdowns in China. My question is, it sounds like customers have been delaying orders here in the near term because of these kitting issues. Do you guys have any sense based on your customer conversations, whether you expect that to extend beyond the July quarter o r do you think that as the rolling lockdowns hopefully come to an end, that orders, you know, start to pick back up? I think, you know, when we talk to a customer, we got a lot of uncertainty from them because they still don't know how fast this lockdown situation will improve. We believe that if the lockdown and this kitting issue not getting any worse than right now, we believe Q3 revenue should be better than Q2, and m aybe even have better recovery in Q4. That's under a lot of assumption that how this kitting issue and the lockdown issue will continue or not. Got it. That's helpful. M aybe for me, a longer-term question. I know you guys have been talking about the opportunity around Level 2+ design wins. I know you haven't announced anything to date, but do you feel like you're making progress? Do you think you're getting closer to potentially a large Level 2+ win that I believe in the past you've said you expect to hopefully secure at least one, if not multiple Level 2+ wins this year? Right. I think that throughout all the discussion, a lot of people keep asking about our CV3. We believe that by announcing a CV3 that is up and running, and we have brought up multiple functions in there, we have a high confidence that we will ship it. We'll sample the CV3 with software, you know, over the summer. I think this will really give our customer a sense how real this project is and what kind of a, you know, performance and the power number that we can prove in a real silicon than just show up in a presentation. I think that with the CV3, we are very optimistic about that we continue to make progress with Tier-One OEMs on this design wins. Thank you. Thank you. Our next comes from the line of Matt Ramsay with Cowen. Your line is open. Yes, good afternoon. Thank you very much. For me, I wanted to follow up on what you were just talking about there with CV3, and maybe you could give us a sense of what the steps are from here. It sounds like you have first silicon back, you have some bring-up being done in the lab. There's obviously more software and BIOS and other work that you'll need to do to get everything fully up to where you can sample the customers in the summer. What I'm interested in is just the next steps beyond that, where you guys could start to engage with customers with CV3, demonstrate stuff live in labs, and what the timeline could look like to potentially start to convert some of these, the part of the automotive funnel that you're going after into one business. What kind of timelines are we looking at to where we could see something announced with CV3? What are the steps that are kind of in front of you to get there? Thanks. Right. I think, you know, maybe just talk about how much we have done with CV3. CV3 silicon came back to our lab seven days ago. In a short period of one week time, we basically brought up all the fundamental blocks and making sure that the chip is not only alive, but we can start doing software integration. We start verifying all the performance number and power number we quoted to our customer in the past. We are getting our confidence level getting higher and higher every day. After the sampling to our customer, the near-term goal for us is to bring up our VisLab software as well as Oculii software running on CV3, so we can do a system level demo. It's not only just demo to show the competency as well as ability of this silicon delivery, the overall system delivery to our customer. I think that's a major milestone because if you talk to all the OEMs today the biggest question for them is not only just whether your silicon can perform, but also whether you have software that can match, you can really leverage and to deliver the best performance. That we have both a VisLab, a complete software stack for autonomous driving all the way to Level 4, and as well as Oculii, you know, centralized radar processing capability. I think that's our next important milestone in the near future to demo it, and hopefully that we're still targeting, you know, year 2025, 2026, SOP with our customers. Got it. No, thanks for that, Fermi. That's helpful. Just a real quick sort of follow-up to that one, before my second question. T he quick follow-up is through this summer, any sense of how many customers you anticipate sampling CV3 to? I'm just trying to understand a little bit more about the breadth of the engagement. Then I guess my follow-up question for Brian, any way that you can try to size the impact of all the logistics issues in China on the guidance for July? I assume it's just $5 million for the Samsung impact. Then if there's any quantification of what you guys think that impact is in dollar terms, that'd be helpful. I know that's a hard one to do, but thank you, guys. Appreciate it. Yeah. In terms of CV3 sample, I will believe it's a double-digit customers. Yeah. In terms of the impact associated with the supply chain disruptions in China, you know, I think, you know, our guide is down about 11% at the midpoint, and that's really all attributable to that factor. Got it. Thank you very much. Thank you. Our next question comes from the line of David Kelley with Jefferies. Your line is open. Hey, good afternoon, and thanks for taking my questions. I believe you noted CV mix for the year is still on track for 45% of sales. Can you talk about the lockdown and supply chain disruptions specific to CV, if you're seeing any less or more impact there and maybe demand trends, you know, throughout the last couple of months? Yeah. I think that this lockdown and the kitting issue impact both IoT and the automotive equally, but also, apply to a video processor and computer vision chip equally. We don't see particularly which product line got impact more than the others. But we believe that the kitting issue right now is really, you know, it can happen to any product line because it's really, for example, both video processor and as well as computer vision need a power PMIC or Wi-Fi or, you know, microcontroller to complete a product. Okay. Got it. That's helpful. I might have missed this, apologies in advance. Can you provide a bit more color on kind of the drivers of the OpEx cadence here into the second quarter guide? Yeah. We said that OpEx would increase coming into second quarter, which is really driven by a new CV 5-nm project that's beginning, and that's starting to kick into the OpEx. Okay. Perfect. Thank you. Thank you. Our next question comes from Tristan Gerra with Baird. Your line is open. Hi. Good afternoon. I know there's been some questions obviously already on the impact of the July quarter guidance and you've reiterated the mix of CV at 45% of revenue. Previously, we thought that CV revenue will double this year. Assuming that the lockdowns go away on June 1st, notably in Shanghai, how much of your second half concern about, you know, the top line rebounding sequentially will be impacted by weak consumer demand in China, as opposed to component shortages like MCU and Wi-Fi? What do you think is the biggest bottleneck or uncertainty after the lockdowns impact goes away? I think that when I talk to our customer beyond Q2 going into Q3 and Q4, I think that their answer is really about uncertainty, y ou know, even they say they relax the lockdown on June 1st, but how fast the production line can go back to normal stage, how quickly that the production can start working to full speed. I think all of that need to be answered. That's not only just the component side, but also our customer side. There is still uncertainty in Q3 and Q4. In terms of China demand, I really think that the China economy that as we see is definitely weak, but we have roughly 15% of our product are consumed in China. I think we are right now more focused on to understand the demand situation outside China, l ike I said in the previous answer, I think that outside China, I think our demand still are strong. Okay, great. As my follow-up question, how should we quantify the ramp of CV5 in the second half, which I think you've said previously initially is going to go into consumer? Is that as well impacted by the current lockdowns o r is this going to be material in terms of lifting your second half versus first half? Right now, because also most of CV5 products are still in design phase and the revenue ramp-up will be happening in Q4, so I think that the lockdown or the development cycle are not severely impacted yet. We are watching it. We're still confident that several of our customers, AIoT customers will take CV5 into production, particularly on the security camera side, and as well as, like you say, consumer. We don't give consumer. We put all the past consumer products into the AIoT. So I think multiple AIoT customers will take CV5 into production in Q4 and start ramping up from there. Great. Thank you very much. Thank you. Our next question comes from Andrew Buscaglia with Berenberg. Your line is open. Hey, guys. I wanted to talk through maybe a dynamic you might be seeing in that, you know, if you have all these supply chain constraints your customers are seeing. I wonder if customers that have yet to really move forward with implementing AI in their products sort of speed that process up and move kind of to the next gen AI stuff, you know, rather than kind of continuing to purchase and develop or maintain non-AI, like more so human viewing applications. Yeah. I think independent of this lockdown and also the kitting issue, we did see that most of our customers' new products are CV-based. Of course, there are still new projects based on video processor, new product, I mean, that kick off in the last several months. In terms of percentage, we see higher and higher still CV-based product kick off in the last several months. Yeah. Okay. Okay, interesting, and m aybe one for Brian. Y ou guys have in your press release a little statement about stock repurchase that you approved an extension. I wonder if, you know, with where the stock is and where valuation is relative to all the CV development if share repurchase becomes more of an area of capital allocation going forward? Yeah, I don't think that there's really a change to the strategy or thought process around repurchases. As you mentioned, the board did extend the existing $49 million authorization for another year. You know, we have had strong cash flow and we believe we have sufficient liquidity. W e did consume around $300 million for the purchase of Oculii and the radar technology late of last year. At this point, we still view the repurchase option as being an opportunistic alternative rather than something that has changed from a strategic perspective at this time. Okay, thank you. Thank you. Our next question comes online with Kevin Cassidy with Rosenblatt Securities. Your line is open. Well, thank you for taking my question. This quarter you had Hakuto, the Japanese automotive distributor, was a 10% customer. Last quarter, they were a 12% customer. Do you expect them to stay as a large customer and m aybe what kind of service do they provide? Maybe if you could just give us a little more description of your relationship with them. Yeah. I think that for Hakuto, they are basically our Japanese automotive distributor. Basically almost all of our Japanese automotive business is going through their service. Their service is just providing two things. One is logistic, basically making sure that chip is delivered properly and we collect cash for us. On top of that, another important function they perform for us is really providing inventory for their automotive customer. Because I think all of our Japanese customer, because they want to make sure there's no supply issue, so they require their distributor, in this case, Hakuto, to keep anywhere between six weeks to eight weeks of inventory for their production. They are definitely planned to push through this at the end of a project. That's the main function. Because the Japanese auto business continue to grow for us, we believe Hakuto will continue to grow. From quarter-to-quarter point of view, the up and down cannot be really managed, because there's some project ramp up, some project ramp down. If you look at a straight line, I believe our Japanese automotive business will continue to grow. Okay, great. Thanks for that answer. Also, on Oculii, could you give us a little more description of what, you know, what progress you made in the past quarter? We didn't separate out the Oculii disclosure but it's very consistent with what we said when we acquired them. We believe this year we're going to do roughly $3 million to $4 million of business with Oculii. Most of them are still at the module level and NRE level. The first major production will happen next year and in terms of auto multi-customer. We haven't announced that product yet. That's all we are doing. However, at the same time, I think all the things, everything I just said is based on Oculii's business before they got acquired. After acquisition, one of the major goal is to integrate Oculii software into our CV family chip, particularly in CV3, so that we can provide integrated radar and video solution to our customer. That has to be a main focus for us and that will also be a long-term revenue source for us. Okay, great. Thank you. Thank you. Our next question comes from the line of Tore Svanberg with Stifel. Your line is open. Yes, this is Jeremy calling for Tore. First, just a quick clarification. I think you mentioned a percentage of products of your revenue that is consumed in China, w as that 50% or 15%? 15%. 15%, o kay, that's what I thought. Secondly, in terms of the inventories, you know, it was down sequentially both dollars and days. Is this something that was intentional? Was this, you know, part of the impact of the $5 million constraint o r is this something that you're kind of managing a little bit more on the long-term basis? Just help us understand your inventory strategy here. Sure. It was really driven by our ability to get materials during the quarter as opposed to an active effort to reduce inventory levels. I mean, longer term, we'll look at targeted inventory levels at, you know, potentially lower levels. I n the current environment, we're trying to have sufficient inventory on hand to, you know, satisfy customers in a very dynamic environment. That decrease was not driven intentionally. It was driven by our ability to get materials. Great. Thank you. Maybe a little bit of a longer term question, looking at some of the new markets that you're, you know, just entering, whether it's automotive, factory automation, access control. Maybe just looking at the latter two, since automotive seems to be its own category. Can you maybe weigh which one you see ramping sooner and which, you know, which one might have the larger TAM? Well, in terms of larger TAM, I think the mobile robotics obviously is the largest one in my opinion. In terms of happening faster, in terms of revenue, I think access control is probably the candidate. However, I think the point we try to make is that we continue to see new verticals that we can address with our CV SoCs. Hopefully, of course, we are focusing on all of them, but I think that we have a high hope on the mobile robotics because of the revenue potentials. Thank you. Our next question comes from the line of Suji Desilva with Roth Capital Partners. Your line is open. Hi, Fermi. Hi, Brian. Nice to be talking to you again. Best of luck in the new role. Brian, on the OpEx, the R&D, you talked about the 5-nm project. I'm wondering, is there any kind of sizable tapeouts involved in the Fiscal 2023 timeframe which may not recur, or is that kind of too aggressive assumption about a year-over-year trend? Yeah. Wh en I look at street expectations for OpEx for the fiscal year, I think the consensus is somewhere around $171 million to $172 million in total for the year. That's probably in a reasonable range to think about. One for Fermi. On the CV3, can you help us understand the design cycle there? Presumably, it's longer given the level of integration and effort on the part of the customer with, you know, radar and other things. Can you give a sense of how much longer that design cycle might be versus the other CV products as you start having customer sampling in the middle of the year? Right. CV3 is designed for automotive markets just by definition. It take longer, and CV3 particularly complicated because not only because of huge die, but also very complicated software integration. I think we think that the four years design cycle is probably right. The first two years probably just on the software and the integration, and the last two year is really qualification and the verification. I think that design cycle is probably standard for automotive market. I think that will be CV3's design cycle. Thank you. Our next question comes from the line of Gary Mobley with Wells Fargo Securities. Your line is open. Hey, guys. Let me start out by saying Casey will be missed, but welcome to the call, Brian. I have just one multi-part question and that relates to you know your Samsung relationship. Can you give us some color on what was the issue with respect to 14-nm constraints? The reason I'm asking is just to try to assess the probability of future reoccurrence of the issue. A s it relates to perhaps some of your purchase obligations with Samsung, you know, given the weakness in revenue, is this going to be a situation where you're going to be forced to maybe carry a little bit extra inventory than you hope for or is that even an issue? Thank you. Yeah. Thank you. For the 14 nm, although we didn't get a clear answer about exactly what happened, my gut feeling is that somebody took that inventory away from us by paying a lot of money. I think that's the reason. Obviously, that's a problem that I need to talk to Samsung about and definitely setting up meeting with their CEO, trying to talk about not only the short-term problem, but also the long-term collaboration. Particularly, for example, if we need to go into this automotive business negotiation with customers, that our foundry partner, obviously, play a major role in there. In terms of inventory level, I don't believe that we need to take on a lot more inventory than we have today. We don't plan that. You know, I don't think that's an issue that we need to worry about in the near future. Thank you. Thank you. Our next question comes from the line of David O'Connor, BNP Paribas. Your line is open. Great. Good afternoon. Thanks for taking my questions. Maybe one for Fermi and one for Brian. Maybe firstly on the edge AI partnership with Lumentum, c an you talk, Fermi, around the traction you're seeing on that and the strategy kind of going forward on the edge AI? Do you have all the building blocks and software internally to kind of support all those different end markets or are we going to see more partnerships like this Lumentum partnership from you going forward? That's my first question. Maybe one for Brian on the ASP. I think at the start of the call you mentioned the CV3 family of SoCs with an ASP range of 5x to 20x the corporate average. Can you give us just an example of the functionality type for that 5x versus the 20x? Is that kind of redundancy or is it an L3 versus L5? Just conceptually, how we should think about that, the range of ASPs. Thank you. Right. I think let me answer that, the ASP problem. When we talk about CV3 5x-20x ASP, I think that it's really referring to CV3 family of chips. In fact, CV3 is a chip that we tape out and we are giving samples, but we are planning to build a family of CV3 from the low-end to high-end. Lo w-end can be level two plus market. Middle-end can be low level three and high-end for level four. You can imagine that ASP for each different market can be different based on the current price quote given to customer. We're seeing 5x to 20x of a selling price versus our current corporate ASP. It's really about function performance, not about the redundancy yet. The other problem is you asked about Lumentum and our partnership with other sensor company. When we introduced CV3. We mentioned that we want to be the domain controller, which means we have to integrate multiple sensor modality into our chip for different applications. Our partnership with Lumentum is really about integrate structured light into our solution for access control market. We've done very well in terms of building a reference design and winning design win with them. We are very happy with the partnership with them. They have been very helpful to help us to solve a lot of, you know, design and production issues. From the software point of view, we've already identified all the software partner we need to have to provide this reference design. Moving forward, I think there will be more and more this kind of cases we need to address, meaning more sensor modality we need to integrate, f or example, thermal sensors, you know, lidar sensors, other sensor. We believe instead of acquiring all of them, but I think our approach will be working with the partners like Lumentum, and to interface our CV3 chip to their sensor modules. We can add value by providing software solution on CV3. That's our strategy and we definitely continue to look for other sensor modality partners. Thank you. Our next question comes from the line of Brian Ruttenbur with Imperial Capital. Your line is open. Yes, thank you very much. You mentioned in your commentary about the traction you're getting on the security side with Vivint specifically. Can you talk about other traction you're getting on the security side in the quarter? A re you able to more easily procure chips for the residential security industry versus other verticals? Yeah. In, we call consumer security camera or security camera for home, w e announced Ring and Vivint as the two major customers. We expect more in the future. In terms of supply chain issues, we see a similar impact for all the customers. For example, I think although Ring is a huge company, they really have a power to buy components. We also see them have some limitation on the supply side. I think that the kitting issue and the supply chain issue impacting everybody, just with different scale. Is there a second part of the question? Thank you. Our next question comes from the line of Martin Yang with Oppenheimer. Hi, good afternoon. Thank you for taking my question. A question on CV3 potential customers that we'll sample, c an you talk about the geographic breakdown and how that compares to your current automotive customer base? I think it's going to be very similar. We're going to sample probably everywhere, y ou know, U.S., Europe, Japan, Korea, China. We all have some identified customers that we probably will sample to. Thanks, Fermi. That's all for me. Thank you. Next question comes from the line of Richard Shannon with Craig-Hallum. Your line is open. Well, thanks guys for taking my question. Fermi, first one for you, m aybe you could talk about the competitive dynamics that you're seeing so far and expect to see with the CV3 chip, both in terms of some of the, you know, new sensor modalities like radar it's generating. Then also, maybe compare that with the competitive dynamics with past chips going into, you know, less complex use cases. I'd love to hear your perspective there. Thanks. Right. I think, like I said, CV3 is really competing in many different set of verticals and have all kinds of different performance and price. That's why we believe CV3, we need to have a family of CV3 chip to address all the opportunities. You mentioned about radar side. I think, you know, there are a lot of radar company out there doing 4D imaging radar, like NXP, Infineon, and also there are some newcomers coming up with the solutions. I think what really differentiates is our approach. Oculii has this algorithm-first approach to really try to minimize the number of antenna and the RF chip you need and to achieve similar radar performance. With that, that really allow us not only building a high-quality, cost-effective solution, but also make the centralized radar solution possible in the future, w Hich means that also allows that we can really have a sensor fusion between video and the radar happen on the same chip. This is a unique offering that we think we will differentiate against everybody else. In terms of the CV3 domain controller competitors, I think that our biggest competitor is NVIDIA and Qualcomm. We believe that when we can demo CV3, we will be able to show the performance and the power advantage that we talked about in our presentation. We are working to and we are very eager to find a way to demo that to our customers. Okay, perfect. I appreciate that perspective, Fermi and m aybe one for Brian here. As we think about the goal of hitting CV 45% of sales this year, and I think it's a fair assumption to think that the IoT segment is higher relative to automotive, would it be fair to think about that IoT segment already being or close to 50/50 CV already? Sorry, can you say that, w hich segment you're talking about? Yeah. Just wondering if the IoT segment already is or close to 50% of sales coming from CV already? In fact, that for the IoT portion, professional security camera probably is close to 50%, but in the other IoT level is not yet. Think about, you know, last year we're only 25%, so we need to grow to 50% this year. By the end of the year, I will say probably the professional security segment will probably be close to 50%, maybe a little more, but other areas will be way below that. Okay, perfect. That's what I was looking for. Thank you, Fermi. That's all for me. Thank you. Thank you. I'm showing no further questions in the queue. I would now like to turn the call back over to Dr. Fermi Wang for closing remarks. First of all, thank you for joining us this afternoon. Although the Fiscal Year 2023 is bringing some unexpected external challenge to us, but I'm really glad our CV momentum remained very strong. I'm particularly excited about the technology development progress we made with CV3 and the business development with CV5. These two final leadership will provide significant opportunity for Ambarella for all of our markets. With that, I want to thank all our stakeholders and especially our global base employees who continue to support us. Thank you. I'll talk to you next time. Ladies and gentlemen, this concludes today's conference call. Thank you for your participation. You can now disconnect.