And so I’d say from a customer standpoint, they need this to build their and train their AI systems there. And then when it comes to, our systems accordingly, is that, data is critical to say, feed the beast of this. We have our Luminar AI engine. We have some of our own initial vehicles that, have sort of gotten that started. And we’re very excited to be able to have the opportunity – start getting in customer data. So, for some of our customers, we actually specifically have data clauses, that allows us to be able to get access to data from customers. So when that happens, that would be very accretive to the overall software efforts and scaling. And we’re going to have more to talk about on the software front over the coming couple of months.
Itay Michaeli: Perfect. Very helpful. Thank you.
Aileen Smith: Our next question is going to come from John Babcock from Bank of America.
Austin Russell: Hi, John.
John Babcock: Hi. How are you guys doing? Thanks for giving me the opportunity to ask a couple of questions here. I guess just starting out, you did mention, in passing that Tesla is using your products and buying sensors. I was just wondering if you could talk a bit more about, the extent to which they’re doing that. Like, so are they just buying part of the sensors, or are they installing it for LiDAR? And also, are they growing business with you? And then if you could also just generally talk about, how their relationship compares with that of other OEMs, that’d be useful?
Austin Russell: Yes. So, what I would say is I’m not going to, I don’t think we’re in the best position to talk about, what they’re doing with our LiDAR. This isn’t the first time that they’ve ordered LiDAR from us, but I would say it’s been more, lumpy than recurring. The reason we’re talking about this is, because if they’re greater than 10% in a quarter, we disclose who those customers are. But, look, they, they’re buying the LiDARs for us and, what exactly they’re doing them, we can only speculate.
Tom Fenimore: They made us sign an NDA.
John Babcock: Yes. I’m not surprised at all, but appreciate the color. And then just next on the cost savings program, I think you mentioned a bit more than half is going to be cash cost savings. Could you just confirm that? And then also, what are the total costs that it’s ultimately going to take to get to that $80 million in annual run rate savings?
Tom Fenimore: Confirm the statement that you said, we kind of mentioned that in the letter. We disclosed what the cash costs are going to be there. It’s on the order of magnitude of $6 million to $8 million.
John Babcock: $6 million to $8 million.
Tom Fenimore: So call it, less than 10 million.
John Babcock: Got you. Okay. Great. And then might you be able to detail the contractor reduction piece associated with that, and then provide any additional color on the other category that’s driving that?
Tom Fenimore: Yes, I would say, you know, the vast majority of that $80 million is coming from a headcount, both our employees and then also contractors ramping down. As you ramp up and we industrialize our first product, reach SOP, there’s a lot of, I would say, non-recurring work that needs to get done, whether that’s setting up the plant, working with our suppliers to ramp up, testing of the products that is required to do, to meet our automotive customer standards. And what we try to do, if the work is not recurring, to fulfill those needs with, contractors, if it’s going to be over a finite period of time. And then as that work is done, and as the SOP and industrialization process for that product draws to a close, you can start ramping down those contractors.
And then as we, rely more on TPK and LTEC for our Halo, which is our next generation product, you don’t need those resources to recur. And so that is what’s I would say driving, the vast majority of the restructuring actions that we took last week.
Austin Russell: And I’d say in particular, when it comes to the contracting partners, we had what, over a 100 contracting partners, in aggregate to help us, in one way or another, the majority of, which were assigned to help us reach SOP. Now that that’s happened, we’re able to roll off the majority of those hundred contracting partners to be, which, helps reduce costs, but it’s something that we have planned anyway as part of this. And frankly, overall, the majority of the Luminar cost structure there, is to be able to help advance, the future of what we’re doing, it’s a relatively dynamic structure that we’re able to have. So of course now post SOP, that’s where I think that dynamic changes and the needs change and models can evolve accordingly as well.
John Babcock: Okay. Thank you. And then if you don’t mind, just one quick question. There was a seating supplier that recently commented about the EX90 being delayed due to software issues. Can you just talk about whether, or not that’s an incremental delay, relative to what you’ve discussed, or if that’s something that had been announced previously?
Austin Russell: I think that’s old news, John, from, a year or so ago.
John Babcock: Yes, that’s correct.
Tom Fenimore: And on top of that, I think there was some misinformation also that there was some delay that was caused by Luminar as well, which was also not the case there either. So, we’re very excited for the EX90 ahead, and that’s going to be a huge driver of growth here for us over the course of the second half of the year that’s going to take us to new heights.
John Babcock: All right, thanks, John. Appreciate all the detail.
Aileen Smith: Our next question is going to come from Mark Delaney from Goldman Sachs.
Austin Russell: Hi, Mark.
Mark Delaney: Good afternoon, guys. Thank you very much for taking the question. One on gross margins, your gross margin came in better than expected in the first quarter, although in the letter you spoke about some production kinks and lower ASPs, as potential headwinds over the next few quarters. I’m hoping you can help investors, to better understand the magnitude of those headwinds, and perhaps more importantly, what might be needed to reach a positive gross profit?
Tom Fenimore: Yes, so, Mark, I’d seen the improvement we saw during Q1 that was largely driven by the industrialization costs, coming out of, it’s starting to come out of the system in good chunks. We were hoping that happening Q4, you know, it happened in steading Q1, as I said, the ability to kind of predict when you kind of do what needs to be done to launch your first product, there’s some variability there. And so that kind of drove the process there. What’s happening now is there is a, once we start selling Volvo series production sensors instead of prototypes, there’s a step function and immediate decline in the ASP. And we can’t wave a magic wand and have our sensor costs decline at the same rate. It’s going to take us a few quarters to start, witnessing the benefits of the actions, we have taken and are going to accelerate taking now to get those costs lower.
We also need economies of scale, things aren’t going to go smoothly. And as you start increasingly ramping up, we’re expecting some unexpected surprises. And so, I would expect, the gross loss to get a little worse before it starts to get better. I don’t want to predict exactly, what that curve is going to look like and when exactly we’re going to get there, but reiterating what I said earlier on the call, it’s going to take us, a full year of serious production to, get close to some of the targets that we talked about a year ago for Iris.
Austin Russell: I would say overall, when it comes to the cost structure here, you have the industrialization costs, sort of launch costs, that you have there, that’s bucketed separately from the actual product costs, in terms of what it costs to be able to deliver each thing. The thing that we’ve done well on is that we’ve now been really starting to aggressively roll off those industrialization costs. So, that’s what’s driven that improvement. And actually we would have been positive this quarter, barring a couple of like, things that are unrelated to the actual product itself. But when it comes to, that next wave, now the focus is going to be as that scales up to get, realize those economies of scale. And the key thing here, is that from a supply chain perspective and supply chain basis, is being able to now, that we have that clear visibility into a volume perspective.
There’s a very big difference between, when you’re ordering components in the, thousands versus hundreds of thousands. And that’s the key distinction from a supply base that we see, we believe we’ll be able to see those benefits of when you’re not talking prototype pricing, when you’re talking scaled series production pricing. So that’s kind of that next wave that drives that.
Tom Fenimore: If you look at the overall direct costs, for example, even for this quarter, it was only around like $16 million, off of the $21 million in revenue and aggregate. So, there’s some things there that, have shown that we’ve seen those realizations, but now that we’re doing that, we’re not stopping. We’re not resting on our laurels by any means. And now we’re focused on this next wave that will drive that. And we’re going to give some more insight as well into specific parts of our business in terms of the profitability aspects, our semiconductor business as a preview over the coming months.
Mark Delaney: That’s all very helpful, thanks. My other question was on the TPK agreement. You spoke around an expanded relationship there. I think in the blog post you put out last week, Austin, you called it an exclusive relationship. So I was hoping to better understand what exactly that might entail, how the partnership and work with TPK may differ compared to your current arrangement with I believe, Celestica. And then also when you think you may go into production with TPK? Thanks.
Austin Russell: Yes, so Celestica, the relationship we have with them, that’s more of a pure contract manufacturer, which is as we make products in a series production, Celestica is making them. They made some of our late stage prototypes just to make sure that their, manufacturing system worked the way it should have. But they’re basically a series production manufacturing partner with us. What we’re doing, and TPK, the deal we announced with them last year was the equivalent of what Celestica is for us at our Mexico plant in the China plant. This new relationship with TPK expands that to more of an industrialization, particularly related to Halo, our next generation product. So, all the prototype manufacturing is going to be done by them.
I would say a lot of the design validation and production validation testing, most of that, which we did ourselves is, we’re expecting them to kind of do, most of that going forward, we’re going to be sitting there verifying and doing some of the results. A lot of the supply chain management, a lot of kind of working out the manufacturing kinks, a lot of that, the inventory management is going to be done by them as opposed to us. That’s going to allow us to move faster, more efficiently. I shared with you, some of the inefficiency costs that we experienced industrializing Iris. I don’t know how much savings we’re going to have for our next generation product relative to Iris, but I’m expecting to be substantial. And none of that is in, the $80 million number that we kind of talked about with, the direct actions that we took last week.
