Luminar Technologies Inc (LAZR) 2021 Q1 法說會逐字稿

(technical difficulty)

I'm Trey Campbell, Luminar's Vice President of Investor Relations. With me today are Austin Russell, Founder and Chief Executive Officer; and Tom Fennimore, Chief Financial Officer. As a quick reminder, this call is being recorded and you can find the earnings release and slides that accompany this call at luminartech.com/quarterlyreview. In a moment, you'll hear brief remarks followed by Q&A.

Before we begin, let me remind everyone that during the call, we may refer to GAAP and non-GAAP measures in our remarks. Today's discussions also contain forward-looking statements based on the environment as we currently see it, and as such, does include risks and uncertainties. Please refer to our press release and business update presentation for more information on the specific risk factors that could cause actual results to differ materially.

Before handing it over to Austin to dive into this quarter's business update, I'd like to share a video detailing the next major milestone for our company with our new manufacturing partners, Celestica and Fabrinet. Let's roll.

(presentation)

Today is a defining moment for Luminar and the broader industry. We've launched the inaugural Iris off the line in Mexico with our newly-unveiled series production manufacturing partners, Celestica and Fabrinet. There's a lot of implications here and I'll speak more to this later. I wanted to take a quick moment for a thank you to all of our employees and partners in this heroic effort to get the facility tooled up and online and continue to iterate.

Zooming out as a recap, last quarter, we made the transition from a lidar company to a lidar-led autonomous vehicle company, combining our software with Volvo's autonomous software subsidiary, Zenseact, as well as partnering with the largest automaker in China, SAIC. We've continued our winning streak across the board and look forward to sharing more in key areas of business for this update call, including team, product, production and commercial wins, and then jump to our CFO, Tom, who will share some of the company-level annual milestones and where we stand on those that -- from what we set forth at the beginning of the year.

Let's start with the team. You can't overemphasize how critical this is. This is a technology, a product in an industry that's never existed before and something that we've been having to pioneer at the tip of the spear. Because even if you have the perfect vision technology or product, you need a world-class team to be able to execute, scale and go to market with commercial wins and build the critical infrastructure for the business.

Two leadership hiring wins that you've probably seen, one of which kicked off the call, is Trey, who is now leading our Investor Relations as an executive coming from Intel; and Alan Prescott as our Chief Legal Officer, who is a safety engineer and legal industry leader who led that team and function at Tesla.

What isn't as visible is what we've done in parallel across all the teams to accelerate our ability to execute. So far this year, we've already hired nearly 100 additional top-notch hardware and software engineers, supply chain experts, employees and contractors to execute the back half of product industrialization and preparation for series production. This includes everyone from the chief assisted driving engineer from a top Tier 1 all the way to bringing on a top industry deal executive to lead our corporate development efforts.

Next up, product. So first off is with Iris. We've continued to hit our key milestones and deliverables, with now multiple key partners already running live with Iris as we optimize it for series production. We've also continued to quantitatively hit our key lidar performance metrics we've promised, not just as numbers on a slide. Our target milestone for the end of the year remains delivering the first C-sample, which we remain on track for.

Second, from a software standpoint, the Sentinel. Following its introduction for the last quarter, we've kicked off the next phase of our software development in partnership with Zenseact at full speed ahead, leading up to the alpha release before year-end. We made 2 key achievements on this front already since the announcement a couple of months ago: one, we successfully collected enough Iris lidar data to train and optimize the performance of our perception software, which we've since executed; and two, we received the green light from Germany to proceed with Sentinel development and testing on public German roads where our Munich operation is based. This is all only reinforcing our transition to a system-level autonomous vehicle company and partner, moving up and beyond the foundation of our lidar.

Next up, production. This area is a key focus for the call, with regards to supply chain execution and manufacturing execution on the path to series production. Today, we've achieved the most significant milestone we've had to date when it comes to advancement towards series production. Two leaders in advanced electronics and optics production have partnered with us to scale our product for us to deliver to leading OEMs and beyond. And in fact, we've already been working closely together for months to get the initial line up and running in the Celestica facility in Monterrey, Mexico. As was mentioned in the video, we've pulled it off.

Celestica is responsible for product-level assembly and fulfillment while Fabrinet assembles some of the core optical components for us. They'll both continue to accelerate and derisk our plans. Just as we pioneered this breakthrough technology as well as advanced manufacturing process, we pioneered a new manufacturing business model that's a first of its kind. The norm is that OEMs only work directly with legacy Tier 1s as they've been working with for decades, and to have a shot, you have to license the technology to those Tier 1s.

Early on, we knew that while that model was easy and appropriate for less complex technologies or products Tier 1s had experience with in a prior capacity, we knew that we needed to have ownership over the product, process and customer end-to-end if this is going to be successful. And for the past 5 years, we've executed exactly to that, retaining 100% ownership over the product manufacturing process and, of course, the economics and relationship with the automaker.

Lastly, when it comes to a commercial standpoint and major commercial wins, we've been crushing it beyond what we could have imagined at the beginning of the year just a handful of months ago. In Q1, we announced a big leap forward with 2 major partnerships. First, from a product standpoint, was Zenseact and second, from a commercial standpoint with SAIC.

So far in Q2, we've continued to diversify beyond our core OEM business, working with leading players in verticals from aircraft to robo-taxis. The partnership with Airbus' UpNext division marked a significant milestone for Luminar as our first foray into aviation, a nearly $1 trillion industry on its own and a step in the right direction for the broader automation economy.

Earlier this week, we also announced another major commercial win with the largest autonomous car company in China, Pony.ai. Since launching their pilot service not long ago, they've driven more than 5 million kilometers across global operational coverage area in 5 major cities, providing more than 220,000 robo-taxi rides. Our business in China is certainly off to an incredible start, following our kickoff in the region with SAIC just a couple of months ago.

So in conclusion, at this stage in our company, execution is everything. And as you can see, we remain diligently focused on delivering our goals and milestones and have been hitting them accordingly. From a business and execution standpoint, I'm very proud in terms of what we've been able to pull off this past quarter and look forward to taking the industry a step closer to making safe autonomy a reality.

Now I'll hand it over to Tom to review our annual goals and milestones as well as current financials.

Thank you, Austin. Our focus for this year remains on achieving milestones that will deliver long-term shareholder value for Luminar in this emerging industry. At the beginning of the year, we introduced 5 critical milestones to help do exactly that and measure success for this year. Thanks to the great work of the team, our progression so far has us either on track or ahead of schedule to hit each of these milestones.

Let's review them one by one. Our first milestone relates to the Iris industrialization and production plan. This is a multistep plan and as Austin outlined, we took a huge step forward with the process transfer to our contract manufacturing partners in bringing them online. The specific milestone is to reach the C-sample stage for Iris by the end of the year, which remains on track.

The next annual milestone relates to software and specifically, the alpha release of Sentinel by year-end. We are continuing to build out perception, proactive safety and highway autonomy functionality internally and with our partner, Zenseact. We are on track for this milestone and actually plan to show off some of our capabilities before the end of the year.

The third milestone was to win 3 major commercial programs this year. As Austin mentioned, we've been winning more deals from major players and at a greater pace than anticipated. We expect to increase this guidance during our next earnings call.

The fourth milestone was to grow our forward-looking order book by 40% this year. Once again, we anticipate increasing this guidance on our next earnings call. Our final milestone relates to maintaining a strong liquidity profile and cash position, with the target to end 2021 with a greater cash balance than where we started the year. In the first quarter, we raised $154 million from redeeming our warrants and had a net cash burn of $29 million, putting us well on track to achieve this milestone.

Executing on these 5 milestones is how we will measure success this year, and the team is off to a great start to achieve or beat each of them. Before we progress to Q&A, I'd like to share a few updates with regards to our Q1 financials. While our quarterly financials in the near term are not a good indicator of our longer-term profitability potential, we nevertheless continue to execute accordingly and have some good updates.

Revenue for the first quarter was $5.3 million, up 120% from the prior quarter. One thing worth noting is that while we recorded a gross loss for the quarter, this is not reflective of our unit economics, our long-term profitability as there are additional and onetime COGS from the initial ramp-up of Iris production, absorption costs from Hydra production ramp-down and development-related expenses related to launching our initial series production program.

We've maintained a strong liquidity at the end of the quarter with $610 million in cash and equivalents, and we had approximately 340 million shares outstanding at the end of the quarter. Lastly, we are on track to achieve our 2021 revenue guidance of $25 million to $30 million.

In closing, I'd love to reiterate my thanks once again for the entire team at Luminar doing a great job during this past quarter. We're off to a great start in 2021. And with that, I'll hand it over to Trey for Q&A.

And our first question will be from Emmanuel Rosner with Deutsche Bank.

So it seems you've already got 3 new commercial programs this year between SAIC, Pony.ai and Airbus. Am I correct in my understanding, this was the guide for the full year? So you're saying it will get raised next quarter? How should we understand this? Are you expecting additional new wins for the year?

And in particular, can you provide an update on your pipeline or funnel of opportunity? I think last quarter, you were saying you have maybe 14 program opportunities. Any update to this number?

Sure. So first, with regards to your latter -- to your latter question here. So the pipeline still remains at 14. As we kind of moved those 2 additional ones now into the win category, we've replenished them with 2 additional ones. Quite frankly, with just how much the team is focused on execution and the level of intensity that it takes with each of our customers to convert them from kind of that development category into the win category, I wouldn't expect us to make that number higher than 14 or materially higher in the near term here, just based upon the level of resource constraints that we've had.

The second thing I would say is, yes, you're right. So far this year, we've won 3 major wins. We are -- a lot of them just happened recently, so we're taking a look at our current pipeline, having conversations with our customers not only about potential new business, but the business we've already won and the exact programs that they're going to include in them. And we plan to provide updated guidance, higher guidance at our next quarterly earnings call.

And I guess as part of this, when would you expect traditional volume automakers to make sourcing decisions about lidar? It feels like so far, not specifically just for Luminar but in general, a lot of the sourcing decision seems to have been either the high-end automakers or robo-taxi commercial vehicle. What's your view around time lines for traditional OEMs?

Yes. So when it comes to some of the traditional OEMs, the businesses only continue to accelerate, as Tom mentioned, for the existing programs that we've been working on. When it comes to getting designed into the major platforms here, oftentimes, it starts with kind of the lead luxury models and then can expand there on out. And those are the exact kinds of conversations, dialogues that we're having in terms of the scope of how the rest of this plays out. And the time line to launch of the different models and programs.

But when it comes down to it, the OEM side of this, this is the core and the lifeblood of the business that's been driving this forward and such, the schedule for all of this at the end of the day. And there's no slowing down on that, so for both existing and for ultimately what we'll have ahead. So that's why all those things considered, as Tom mentioned, we'd be increasing guidance next quarter as we continue to get these new data points as -- from our recent wins and things happening later this year.

And then, maybe just final one. So you now have 2 major wins in China between SAIC and Pony.ai. Can you describe the competitive dynamics over there, maybe relative to the U.S. and Europe? Are you mainly competing against local suppliers or again, some of the U.S. ones? And I don't know if you're able to comment on this, but for a program like Pony.ai, would the pricing be higher, given low volume or -- generally on robo-taxis versus traditional automakers?

Yes. So I'll say, generally, as kind of a rule of thumb for these programs, people come to us when they're really ready to be able to reach that series production phase and when they're starting to make that transition out of development. We generally avoid working with kind of earlier-stage programs that may or may not materialize or have a lower likelihood of that, and that's why we tried to stick to the major players that have the resources to really be able to put behind it and see it through at the end of the day.

And yes, when you talk about leaders in their respective fields like SAIC and even Pony here, that's certainly the case for China and beyond. When it comes down to it, that's where, again, there's a lot of different approaches that you can take when it comes to the development and testing, but that's the whole point of why we're here and what we're changing.

So taking a holistic step back, yes, I think we continue to certainly see that accelerate and, as I said, are kind of pedal to the metal, and this is really the start of our operations in China.

Question is from Dustin Scaringe with Baird. He's on audio only.

This is Dustin on the line for Tristan. I was wondering if you could talk a little bit more about Sentinel. You received the green light in Germany to develop and test on public roads. I was wondering, first, if this was a relatively easy process. And then what are the implications for testing and development in other countries? Do you think it will be easy to test elsewhere? And then I have a follow-up.

Yes. So when it comes to Sentinel, there's been a lot of work that we've done and put into place. Just a second here. Sorry, just a technical adjustment. So when it comes to Sentinel, this is where we continue to be able to focus on the core development of this. There's a lot of work that's left to be done to really be able to make this happen and bring this out in the broader industry.

The whole point of what we've been doing in Germany and in Munich, we actually did an acquihire of this team from Samsung previously to be able to accelerate the development of this full stack solution and, of course, recently partnered with Volvo's subsidiary, Zenseact, to be able to accelerate this. We have now really kicked off the actual testing and development on roads. And the approval from Germany was certainly an important step in that since a lot of the development that's happening there is really being realized and tested in Germany itself.

And that's where we are, of course, having a global scope of the overall deployment. But testing and validation is a critical part of all of these processes. And for the -- until it's actually production-ready, you need to be able to have permission to do that.

Got it, okay. And as a follow-up, I was wondering if you guys can talk a little bit more about Airbus UpNext and your partnership there in more detail. What vehicles do you think your lidar sensors will be placed on? What problems are they solving? And then, finally, given that solving the autonomy problem in cars and trucks already is hard enough, how feasible do you think lidar is in flight, given that they travel at similar speeds in completely undefined environments?

Yes. Yes. So in some ways, the autonomous flying problem is easier in the sense that there are a lot fewer obstacles that you don't necessarily encounter in flight. At the same time, it's more difficult in the sense that you need to have a critical level of reliability to be achieved to be able to make this safe and autonomous. And that's really the whole point of what we're enabling is to be able to better enable autonomous flight on everything from -- starting on helicopters and moving all the way to VTOLs and ultimately things like fixed-wing aircraft in terms of the kinds of capabilities it can have.

What we're specifically designed into is their Vertex platform, as they call it, which is basically to be able to demonstrate and prove out and validate these safety improvements and capabilities they have on these flights. And what they have, of course, a certain set of program milestones and everything that they've been delivering that on. But basically, we are designed into that platform and solution that they can then branch off into the different areas.

So of course, aviation is kind of its own world and has its own aspects to it. That's a little bit different than the vehicle platform side, but this is kind of the key first step, and we thought couldn't think of it and find a better player to be able to work with than the largest aviation company.

Our next question is from Gus Richard with Northland.

I was hoping you could give a little bit of an update on your work with Mobileye at this point.

Yes. So that's certainly continuing. I think when it comes down to it, one of the things that we mentioned is that we now have actually been deploying Iris to multiple programs and customers at this stage. So I'm going to be careful not to comment on anything related to customer confidentiality. But things are absolutely continuing full speed ahead with the program. There's no otherwise significant update other than we're now deploying this in a more widespread capacity across the board with the key programs, to launch programs with customers and are going to continue to deliver against the execution plan, leading up to the start of production dates of our various programs that we've been designed into.

Got it. And then in terms of production, you've got units coming off the line in Celestica in Mexico. Sort of what run rate are you at now? And are these considered B-samples? Where are they in the maturation of the product?

Right, right. So as of today as we speak, it's a little -- we've actually just gotten things up and running with those guys and online. So part of what we're building out is for a capacity kind of in the -- that would be capable of in the tens of thousands of units annually per line in terms of what we're building out in terms of that. And then part of the advantage of working with folks like Celestica is that it's very scalable, so you can basically adjust the amount of lines that you need in a modular capacity to be able to fit the dynamic nature of what this industry is and how this scales.

So that's the plan that we've been executing to. And this is really all leading up to the start of production before the end of next year. And that's where with these programs, we're working hand in hand. They're coming in, auditing the facility, our process and what we have to do in terms of this process transfer, but we have that well underway. We're off to a great start and we've proven that it actually works and the model works.

And then what has been the customer interest in Sentinel? How widespread is that? Is that half of the engagements? Can you give a little bit of a sense of how many customers are interested in just the sensor and how much on just perception and then the full stack?

Yes, yes. So I'll say this, when it comes to the full stack, I would say 80% of the broader automotive ecosystem has to have some -- either Sentinel or some equivalent of this if they actually want to be successful in deploying pass-through vehicle autonomy and these kind of proactive safety capabilities. Because when it comes down to it, there really isn't -- most of them don't actually have the internal development capabilities and expertise and know-how to be able to really make this happen in the first place, and that was part of the inspiration.

So we've been continuing through the development process and engaging in a number of conversations with all of these guys, some of which we've already seen start to be adopted. And we -- it's not actually an all-for-one either. Some automakers have certain software components that they've developed in-house and then are taking it in a more a la carte offering. Our ultimate goal is to be able to make sure -- ensure widespread penetration of Iris. But of course, there's also great economics with the software as well.

So yes, I'd say, for -- I would say for some components, at least some components of Sentinel, the majority of customers do have interest in this. And that's where this is going to be leading up to our alpha release at the end of the year.

The next question is from Aileen Smith with Bank of America.

I wanted to dig in a little bit on some of the recent series production wins, particularly with Airbus. At the time upon going public, you guys were pretty focused on your product and your addressable market opportunity being at Level 3, 4, highway autonomy, less so on the Level 4, 5 robo-taxi market and then even less so on nonautomotive end markets like aerospace and defense and industrial.

So what of your discussions with new and potential customers over the past few months informed you about future product direction and the relevancy of lidar and really where you want to focus. And is the emphasis still on Level 3+ systems for production vehicles? Or are the recent wins, in any way, indicative of the planned traction for highway autonomy being pushed out a little bit?

Yes. So when it comes to our core capabilities and what we're enabling, nothing has changed there. We're 100% focused on delivering highway autonomy, delivering proactive safety capabilities into these core passenger vehicle production OEM programs. That hasn't been slowing down. But what has been accelerating as an opportunity for us to leverage is actually some of these other companies in addition to the major OEMs that we see as major players that opportunistically can come up that we can be able to work with to leverage the exact same products and all of the hundreds of millions of investment we've put into this technology and this business for our OEMs, but for these other industries as well.

And I think the expectation is still absolutely the same that the only -- the way that you can build your business, you have to have series production OEMs that are going to be driving that volume curve along with this. So it's really icing on the cake for these other programs in terms of what we have and what we're winning. For the right opportunity though, we're certainly not going to say no to it. And vice versa, we're actually going to work with these companies to be able to ensure that it's seen through.

But we'd maintain the time horizon for each of these different areas is different. I still actually firmly believe the passenger vehicle OEM market is actually going to be realizing the volume a lot more significantly and sooner than the robo-taxi market that's out there. But the nice thing is that it is still an additional opportunity.

And I don't want to make this sound like there's no potential for robo-taxis over the long term. I actually -- I'm still bullish on that for like -- just let's hope for the next handful of years in terms of volumes, I would think it's going to ramp up more towards the end of the decade and really through the 2030s.

One other thing, and I realized I wanted to make sure we address one other part also that also relates to a previous question, is that, the question is, if something is lower volume, do you get higher ASPs? And the answer is yes, there is a price volume curve associated with each of these things. So it does make a difference.

Great, that's very helpful. And then I wanted to understand a bit better the outsourcing strategy and the announcement around Celestica and Fabrinet. For the automotive layperson, can you describe for us what each of those partners are really bringing to the table and so we understand it correctly? And apologies if this is a rudimentary question. Is the manufacturing process separated at all between the 2? Or is it rather a setup of dual sourcing as would be pretty standard in the automotive industry?

Yes. So when it comes down to this, I mean, the reality is this is an incredibly intensive process that we have to be able to streamline as much as possible. And we've been building out and focus on building up the supply chain for these newly developed products. And this isn't a commodity, this isn't something that has existed for some time. So our whole plan has been to work with the best possible players in each of these respective areas in terms of what we can focus on.

For example, with Fabrinet, these guys are historically known for building complex optical assemblies and doing so with precision optics incredibly well, in doing so, taking it from low volume into high volume, helping with manufacturability and working hand-in-hand with that. And that's been great. They've been working with and collaborating with our team in Orlando from an advanced manufacturing perspective.

And then equivalently with Celestica, they're actually in charge of the overall assembly of the lidar and the device, final assembly packed out and shipped. And that's what we've been able to leverage and outsource with those guys. So it all kind of fits to the broader picture and puzzle.

Could we take an approach where, for example, we also build it ourselves? Part of the plan in terms of what we have internally is we're actually defining the process, the capability, all the stuff that's really, really unique and stuff that relates to us. But what we don't plan to do is also build the whole system and everything internally. Actually, these partners exist for a reason and are partnered with us and do a great job out of this and in a very cost effective capacity.

So it just makes the most sense and it's very modular, it's scalable. They already have a lot of CapEx there, too. So -- but it's all about working with the right folks, getting the right arrangements in place and setting up the supply chain right, and that's exactly what we've done.

Great. And one last question, if I may. We touched upon this a little bit in terms of the hiring efforts of the company over the past few months. As you make a shift from a lidar hardware supplier to a lidar-led autonomous company, by your definition, are there any technology gaps that you see within the business that you would look to augment through acquisition? And how much of the talent and engineers that you've been hiring would you estimate are more responsible for industrialization and commercialization of your existing products versus responsible for new technology and product development?

Yes. Yes. So I think we'll probably have more developments to speak to next quarter when it comes to all of those things, but I think you asked a very good question on that. And the answer is, yes, there are other aspects that aren't maybe as obvious when it comes to the full stack solution in terms of things that we may decide we may not want to specifically do every part of it in-house.

I mean, part of what we've done, for example, for the strategy in working with folks like Zenseact is that they're actually building a core module when it comes to the planning controls, actuation, everything associated with that part of the full stack solution. And working with folks like that, that can spec out the designs of also certain hardware components as well to go along with the holistic solution.

But we are continuing to onboard a significant number and folks with engineering talent, as you noted, not just from a hardware standpoint but also significantly from a software standpoint as well. And that's really driving the solution. But yes, we're going to be making our own set of -- for certain aspects of it of make versus buy decision, so to say when it comes to this.

But the important part, as you start with the actual, the foundation that can enable all of these guys because again, part of the whole point of this and kind of as it relates to your earlier question, is that we're doing this to help enable automakers that can't -- that don't have a path forward, that don't have a solution unless they can actually source something more holistically. And we're really the only ones that can actually do that at this point. So that's the focus then make sure that they can get enabled as well and not get left behind in the autonomy race.

Our next question is from Itay Michaeli with Citi.

Just I had 4 questions on the Airbus agreement. First, can you talk about the number of lidars per aircraft that you expect? Second, maybe talk about the competitive set and whether your architecture of 1550 maybe is more suitable for aviation applications? And then is there any software component with that? And should we expect perhaps more R&D to support these types of programs?

Yes, a few things there that are packed into it. So the first one, when it comes down to it for a number of the lidars, it really depends on the actual vehicle and everything. Is it a helicopter where you're trying to be able to enable safe landing? Is it a VTOL where you're trying to be able to get 360 surroundings for urban air mobility? Is it a fixed-wing aircraft? And yes, you can see it scenarios anywhere from 1- to 4-sensor configurations for each of those different modes of air transport. But depending on forward-facing versus 360.

But when it comes down to it, I think from a technology standpoint, the answer is, yes, this is well suited to really -- to things that require high-performance systems, there's going to be a lot of different adjacent market applications that can benefit from lower performance lidar systems and other things that are there. I mean, part of the whole point of what we've done is we've built this incredibly stringent specification that -- of what's needed for automotive in all of these different key areas that you have to get simultaneously.

I mean, the obvious ones, some other people will talk about it, is range and resolution, but there's like 10 other things that you have to do, and you have to do them all very well simultaneously. And that's where we've kind of built our claim to fame. That's why we've seen such adoption with the major OEMs and for series production programs.

But those benefits do apply into other industries. But folks like aviation require the utmost level of performance in safety, and that's exactly why it's well suited. 1550 is certainly a significant part of that and kind of going back to the core architecture. The same principle applies to cars, of course, as well. But yes.

And then lastly, from a software standpoint in addressing that question, the answer is yes, some of our software is portable into other industries. The full scope and set of software is not necessarily relevant, for example, at least not yet. There aren't highways in the sky. So it's really just kind of more general autonomy. And the most important things are like takeoff and landing.

And you do need actually some level of perception on that too, so like you can't actually have applicability of the perception part of Sentinel when it comes to that. But I would say the software, more than anything, is focused on enabling OEMs and the majority of which that don't actually otherwise have a solution. So kind of thinking a number of moves ahead in this overall game. I mean, there's the guys that have that set things up that need a lidar today with this, but then there's the guys that don't actually have a plan, and that's what we're trying to optimize for.

Perfect. That's very helpful. And just as a follow-up going back to Sentinel. You might have actually covered this, but are proactive safety and highway autonomy kind of 2 separate offerings? I mean, you can get an OEM, just go with one. Maybe talk about the progress on both and maybe even the cost range of how you see proactive versus highway autonomy.

Yes, yes. And so there are 2 different offerings that are out there. So when it comes to the proactive safety capability, this is the kind of capability that we can see that can be -- and what automakers are asking from us, that they'd want to ultimately have a vision of standardizing it across the vehicles they make, of enabling a completely different level and capability from a safety standard perspective such that we can go from starting to try and mitigating accidents and how they happen to actually just be preventing them in the first place.

And this is a capability that we can offer at a very low cost as well to be able to make a standardization even more capable. At the same time, we don't want to give up on the margin for highway autonomy capability of why we have like, for example, these dynamic pricing models and other things that are set up to enable those kinds of things, and effectively as upgrade options to the consumer.

So that's really the vision in terms of what we put forth. What we think proactive safety is going to be the thing when talking about by 2030 or even 2040 of seeing this ultimately be a requirement for vehicles and actually changing -- and when we did the modeling, it can change the safety factor of the vehicle. People fight for like single-digit percentage changes in safety, but this can be up to 7x in terms of those kinds of capabilities.

So that's why we're really focused on it. It's a really overlooked area of the overall autonomy landscape, and it's something that really nobody has been doing or paying attention to. It's just the legacy ADAS suppliers that do like camera radar stuff, but that's where we have to take it to the next level. And that's also an important aspect of it is the same hardware that enables proactive safety is the same stuff that enables highway autonomy. So you can do over-the-air upgrades, even after the fact, after you buy the car to have those kinds of capabilities. And you already see folks like, obviously, different set of assisted driving capabilities that you see the folks like the Teslas of this world that are already starting to do certain things like that, which is smart.

And then I'll let Tom comment on some of the economics and how we're seeing that in terms of the bigger picture.

Sure. And what we've said historically when you actually look at the content per vehicle, to borrow a word from the auto supply landscape, for highway autonomy, and this is both the hardware and software, we think it's going to be on the order of magnitude of about $2,500 per vehicle for the passenger vehicle space. And then for proactive safety, probably somewhere on the order of magnitude of about $1,000 per vehicle.

And so from that perspective, that's how kind of we see the pricing opportunity between the hardware and software. And then clearly, as you move into the commercial truck as well as into the robo-taxi space, there, you're talking about multiple sensors per vehicle. But on the passenger vehicle side, initially, we're seeing one sensor per vehicle.

Still leaves some good room for profit for the OEM, too.

Our next question is from Richard Shannon with Craig-Hallum.

I wanted to follow up on one, Austin. I'm not sure if I understood your response on the Fabrinet and Celestica dynamics here, whether they're doing exactly the same things or they're doing -- passing 1 up to the other subset. If you could just describe that more specifically, I just want to make sure I understand that.

Yes, absolutely. We'll dive into that next layer of detail. So Fabrinet, they're basically doing the transceiver assembly, so basically, some of the core optical components that go into it, particular focus on the receiver. So they basically take the ASIC, the chip that we make, combine that with our indium gallium arsenide photodiode, package that in automotive grade and sort of by processing capacity. Then that's shipped off to Celestica in Mexico, where that's integrated as part of the overall final assembly of the device and then ultimately shipped off to the end OEM customer.

So it's all part of the value stream that kind of fits into play. And there's ultimately more than Celestica and Fabrinet, but those are kind of the 2 central players, with Celestica being the company that actually does the final assembly. And we worked with them to be able to set up a test process, qualification and validation procedures for the actual lidar as it comes off the line.

So that's part of what you saw in the video there, too, with some of the different targets being set up, calibration modes and all of those other things to be able to do so in an automated capacity. Does that make sense, kind of what you're looking for?

It does. I missed -- I got in a little bit late, so I missed the video early on. So it's probably what I was missing there.

My second question was regarding the comment about increasing the order book as you get into the next call. Is this specifically related to your 2 most recent deals, SAIC and Pony? And not sure what that adds to it, and so that takes a little bit of time to understand it. Or is this also maybe some deals that might be announced between now and the next call? Or even customers and partnerships have been announced even before those 2 that had just been announced this year?

Yes. And specifically, what we're going to be planning to do at our next earnings call is to update the guidance not only on the forward-looking order book that you just mentioned, but also the number of major commercial wins that we expect for the year. And the latter drives the former. And so we've had, as we mentioned earlier, during the prepared remarks, a better pace of winning than we were expecting.

And some of those wins have just happened recently. And so we're going to develop a -- sit down and look at what we expect to happen between now and the rest of the year, not only in terms of how many additional wins that we expect but what that translates into with regards to the order book growth. And we'll provide that updated guidance at our next quarterly call. So I think it encompasses a lot of what you just referenced.

Okay. Maybe one last question for me, Austin. This kind of big picture long term, you kind of talked about 2 to 3 different markets here, passenger cars, trucking and then robo-taxis. You kind of made a comment earlier to someone's question about the dynamics within robo-taxis.

But just kind of broadly speaking in the last quarter since your last update, I guess, about 2 months ago. What dynamics are you seeing here in terms of either acceleration or slowdown in the passenger car and trucking space, particularly those things that are outside of your control, outside of the perception stack? Are you seeing them pull forward, pushed out? And if so, can you comment on those drivers?

Yes, yes. So when it comes down to it, I think this is where each OEM, of course, has their own program time line. Generally, there's a lot of time in terms of the lead-up to the actual start of production with each of these guys. But I don't think we've seen any kind of monumental wave or shift there when it comes to a couple of months, would absolutely stand by the plans.

And what we are seeing is that OEMs becoming even more and more invested into seeing this through and potentially actually expanding it to more models of vehicles than even we were initially thinking beyond just some of the immediate top-of-the-line luxury vehicles. So that's kind of where we're in active review with all of these guys.

When it comes down to it, I think still bullish on autonomous trucking as well. I think that's where the highway autonomy capability and having that constrained use case is really important. Of course, various players are at different stages. And I'd say the overall autonomous trucking space is actually a little -- it's a little bit behind the pass car space when it comes to just overall readiness to really get into production. So that's going to be a bit of a different time line there. But I think over the long term, that's bullish.

And then over a little bit of a longer time horizon as well, like I said, I just want to make sure that we're realistic on it so that it's not a surprise. And I'm talking just at the industry level, not just as it relates to us a few years from now when our robo-taxi overlords are driving around everywhere, taking everybody between every city. It's going to take time for this to proliferate. But again, it is one of those incredibly valuable spaces that is -- it does make a lot of sense.

And then when it comes to software adoption, everything, that's only accelerated for us. Ever since we announced Sentinel, we've seen a huge amount of interest. And as was we've mentioned, not just from -- for both our customers, but also again, engaging folks that aren't our customers that can now become our customers because we have this holistic offering. So that's where we're continuing to do that even before the full software suite is ready.

And our last caller is David Kelley from Jefferies.

Just a couple from my end and maybe starting with your recent announcements, SAIC, Pony.ai. Just hoping to get your thoughts on the China autonomous opportunity, maybe even high level as a region. I guess from our vantage point, it feels like the market is perhaps more focused on streamlining autonomous rollout. So curious as to how do you think about adoption time lines in China versus the U.S. and Europe. And then if the regional opportunity there and your strategy there has changed at all in the last few months, it would be great.

Yes. So our strategy has definitely gotten more aggressive over the past few months. I mean, we knew that we wanted to be able to launch big and bold, and that was really to kick off with SAIC in terms of what we're able to do and pull off there. That's -- I think working with some of the key leading players in China is going to be key. I think at the same time, what's tough is that, just as with a lot of opportunity, there's a lot of noise, and I think there's actually like a surprising amount of noise even within China. We'll see how the rest of that landscape plays out over the next couple of years there, I think.

People are moving very quickly, which is great. This stuff takes time for development though. There really are no shortcuts. And I think the important part is taking the high road and not trying to cut the corners when it comes to performance, when it comes to safety, when it comes to all of these things. And that's really what the -- a lot of the top players are doing, and that's the whole point of why they've tapped us to work with them, to solve exactly that, just like in Europe and the U.S. here.

So that's continuing. We're building out the Shanghai office and that's something that's really just starting here. All this has kind of come together in the past few months. So we're having all this play out in real time. But this is as good of a kickoff as we could hope for, for our operations in China.

Okay. Got it. That's super helpful. And then maybe just as a last one here, Austin. You mentioned the recent employee additions. I think you might have mentioned 100-plus or so, a number of roles there. Just hoping you could maybe frame that for us, how that's tracking versus your prior expectations. And if we have some visibility to the order book ramping here. Do we expect some more aggressive uptick here in the future?

Yes. So we are absolutely aggressively investing into the overall team, into these programs. Of course, each additional program that you take on provides huge opportunity, huge growth, huge everything. It does take kind of incremental resources, of course, that you invest into it to be able to see that through.

When it comes down to it with all of our team, I think if you ask anyone within Luminar, you'll probably get an answer of it's kind of a miracle how we've pulled this off so efficiently in terms of going through the execution path, and this is where everyone's been cranking day in, day out with a relentless approach and this maniacal focus at delivery and execution and all these different milestones that we have to hit for our core product and the customer milestones associated with each of this. So that's been good.

The 100 folks is plus or minus in terms of according to the plan. We remain on track with regards to the overall net spend for over the course of the year. But importantly, more than anything, and the team will always reiterate this, I'm always a huge fan. And it really -- I've been fortunate enough to learn this earlier on of just the quality of people and quality of leadership is just so much more important than the absolute quantity in terms of what you can do.

Now of course, you still need that to be able to drive, like if you want to build an auto grade product that actually meets all these specifications, even in a perfect world, it's still like a $0.5 billion to $1 billion worth of investment and years to actually do this and execute. But if you want to actually have a shot of seeing it through, you need to have those right team members in place. And it really makes all the difference with us.

So we did a recent highlight. We actually have a couple of the folks here in the room today, with Al as our Chief Legal Officer and Trey leading our Investor Relations. But there's also a host of other new great hires that we've had that are -- adds incredible skill sets and levels that I'm sure we'll have more to talk about to later for next quarter.

Okay. And that was our last question from our analysts.

Great. Well, thanks, everyone. This has been a, hopefully, helpful overview of everything and kind of status update in terms of where things stand. I really couldn't be more excited in terms of where we are and what we've been able to do. I think the whole Luminar team would have been blown away if we had heard just a year ago what we would be able to pull off and accomplish. And I think that a big thank you to the entire Luminar team in that respect of working day in, day out to be able to make all of this stuff happen.

And we're only just getting started. So thanks, everyone, for joining. We'll see you next time.

Thanks, Matt, if you can close out the call.