ENVX - NASDAQ

[Audio Gap] Letter and our filings with the Securities and Exchange Commission. All of our statements are made as of today, July 31, 2025, based on information currently available to us. We can give no assurance that these statements will be proven to be correct, and we do not intend and undertake no duty to update these statements, except as required by law. During this call, we will also discuss non-GAAP financial measures, which are [indiscernible] and are in accordance with generally accepted accounting principles. All of you can find a reconciliation to the GAAP financial measures to non-GAAP financial measures in our shareholder letter [indiscernible], which is posted on the Investor Relations page of our website. With that, I'll turn the call over to Raj. Raj?

Thank you, Rob, and thank you, all, for joining us for this presentation, and thank you, TJ, for joining us today. Fellow shareholders, the company made great progress in Q2 in all fronts. We launched our A1 product platform. We hit key milestones with many of our strategic customers. We launched a warrant dividend to simultaneously reward our shareholders and also fund our future growth. Fab2 in Malaysia had made the first A1 batteries now, of our high-volume manufacturing line. Now with our product sampling to 2 major smartphone OEMs and a leading eyewear company and strategic IoT customers, we are truly now moving into the commercialization phase of our journey. Now, in terms of highlights for the quarter, there's a few highlights, and I'm going to walk you through some of them using their shareholder letter in the backdrop here, and I'll point you to the things I'm talking about. Revenue, we delivered Q2 revenue of $7.5 million, above the guidance range, and it's up 98% year-on-year. And this was driven by strong demand for our products. And our non-GAAP gross margin reached 31%, reflecting the higher margin of the defense products that we get from our Korean factory. And we launched the A1 platform. The A1 platform is a leadership platform with the core battery architecture. It has some -- all the advanced materials, [ microenvironment ] design rules, and it meets all the requirements of the smartphone market. And it is, in our opinion, the highest watt hours per liter, 900 watt hours per liter battery with 3C charge rate, and now we feel it goes up to 1,000 cycles. This is the first platform that has 100% active silicon anode, and it's in commercial qualification with our customers now. And again, we have over 400 patents on top of this technology that we have built. Doing this high energy density with long cycle life and fast charging is really the key value of this platform and our customers are -- a lot of interest from our customers to qualify this product. In a small battery like this, we have 26.3 watt hours per liter. And I'm going to have TJ talk a little bit about what that number means.

Thank you. I decided, prior press release, which I asked him to put in here, do a little peek work on this thing. The battery is 1.8 cubic inches. And then you asked, well, how much energy is 26.3 watt hours in that little package. Well, the answer is if you do a calculation, it's enough to lift a 5,000-pound truck. This is the height of this -- working height of 4.7 feet, 3x in that little battery. And we're actually working right now on a demonstration to do that for real, except I'm going to use the Lamborghini instead of a truck.

Thank you, TJ. So a couple of other highlights I want to talk about. Here, we've now sampled this battery to our lead smartphone OEM, and they are doing all the different tests that they do to make sure this battery is working well in the phone that they want to launch in. And this testing takes time, 1,000 cycles, takes multiple months to test because when you -- when they put a new [indiscernible] battery in, they go through the full cycle life test. And we started sampling them in July, so in 3 months. Of course, we have weekly meetings with them to see how the tests are going, and we expect the test to be completed in 3 months' time. If everything goes well, we should be getting orders from them to get this to mass production. We've also made a lot of progress in smart eyewear. The AI-1 battery in smart eyewear is also a tremendous value because in smart eyewear what happens is there is not much space to put a big battery. So typically, as you can see here, the battery is really small. It fits in these lengths here. And sometimes people put 2 batteries, one on each side. One of the key things about the -- these smart glasses is AI is having a huge impact on the usability of these glasses because now you can talk to these ones and most of them have cameras that look at the world, so they're able to detect what's going on in the real world and then use that information to change how the glasses experience is. So which means you need tremendous amount of energy density in a small package to actually do that, and it's a real advantage for us. And we have sampled this to multiple AR customers and the amount of samples we did also increased this year. And the other important thing to remember is that I believe that the AI use cases will only continue to grow in this market. And as these use cases grow, the process have to run faster, the memories have to run faster. There's a lot more data that has to be computed. There will be more sensors there, which means we need a road map of product technology here where we continue to increase energy density in the small size. And you can see here on the right side, we have a road map of different technologies that will continue to increase that. Now a few other things. One thing I find about these markets like AR when they first start is that it's very important to be the battery of choice for the ecosystem. And what I mean by ecosystem. Ecosystem is people who make processors, people who make the waveguide optics, people who make different sensors. And we are now working with the people who actually make this process. We have sampled these batteries to them to make sure that the batteries have the performance they need in the various use cases because in these markets, the use case are a little bit different than smartphones. The discharge rate is a little bit different and so on. So it's very important to have a battery that supports the ecosystem and we be in reference designs than any number of customers that can actually take the reference design and the Enovix battery has a potential of becoming the standard for that. So we're making a lot of progress on that this quarter. The other thing I've said before is that when you make a battery that can meet the requirements of the smartphone market, high energy density, long cycle life, and fast charge, that is the toughest battery to make. That battery can then be used in many other markets. And we are seeing that now with the IoT markets, where recently we've had a customer in one of the leading customers in industrial handheld markets sign an agreement with us so that they can actually get these batteries and test them in the next generation of the products they want to launch. Now I also mentioned that with the recent -- in the last call, with the recent tariff situation, people really want -- in defense markets, people really want batteries made from countries with more advantageous tariff situation. And we have this Korean asset that we acquired, and we added to that a lot more capacity last time with the acquisition we made from SolarEdge. That is now helping us grow this market into some of the other U.S. defense companies, and we started sampling to multiple customers now. Now our factory in Penang is really coming up together well. And in Q2, we accelerated our customer qualification efforts. Multiple customers have come now and visited our factory. And as I mentioned, in this space of making lithium-ion batteries, making custom battery is very important. We have now reduced the time it takes to make a custom battery down by 50% from before. Because as I mentioned before, our factory, our machines are able to make batteries of different sizes, but we have to reconfigure them, and we are able to do it much faster now. But the factory -- this AI-1 smartphone battery has also now passed the UN 8.3 certification, which is very important for airline safety certification for our first battery, and now we are able to ship these to our customers. And as we build confidence in the customers we are sampling and the qualification that's going on, we now started placing purchase orders, initial purchase orders for adding more capacity to the factory and building some of the long lead time items for the second HVM line. We were also able to do that in the last quarter. What I actually want to do now is, I know we put a little teaser of a video before, I actually want to show you a video we just recently did of the entire factory producing the AI smartphone batteries that we are sampling to customers. So let's play the video. [Presentation]

Awesome. So I hope you guys got to see that video, and we are super excited by that factory. And really, that's what state-of-the-art battery factory exactly needs to look like, and you can see the cells being produced there. Now, in July, we also issued a special shareholder warrant dividend, rewarding all our shareholders and giving us an opportunity to raise additional capital because we want to build out the rest of that factory that I show in that video. Now the asset we acquired in Korea to add to our -- increase our defense capabilities -- defense battery capabilities also came with additional coating equipment. And it's a really important asset that we acquired because this newly acquired coating equipment is going to help us ramp the batteries in Fab2 as we start building out the factory and as we start getting higher and higher volume into production. Now I want to say a little bit about coating. Coating is a very important part of battery making. In coating what we do is we take this giant copper and aluminum rolls, and I see here like 105-millimeter wide and 500-meter long rolls of copper and aluminum foil. And we get this latest anode powders and cathode powders that they keep changing in the market as people make more advances, and we coat them on top of these rolls. Now it's very important to be able to coat this to the precision we need for the density we need and so on, because the way we make the batteries, the way we dice the rolls and put them together is very different from people who just make jelly roll batteries. So it's important that we control the coating. So this ability that we acquired is really helping us. And it's helping us in many ways. It's helping us reduce cost. It's helping us reduce -- increase the battery performance. It's also helping us prototype new materials really, really fast. It used to take us 20 weeks to prototype a new material. Now we can do it under 7 because we own our own coating. And this is the power of being vertically integrated in this battery manufacturing. Now I'm not -- this is the investor letter. I'm not going to go through all of that, but I just want to say something which I mentioned when I first joined the company. The battery is the bottleneck, and it is becoming even more of a bottleneck for people to really realize the full power of the devices they buy. And this has become even more and more evident now as the AI and AI applications have become even more prevalent in the phones and in other consumer electronic devices. Now recently, and I show a little anecdotal evidence here, we saw a smartphone AM launch like a mid-tier or maybe even a little bit lower tier phone, which is $100, $200 phone, $200, $300 phone, have like an 8,000 milliampere battery. Why is that important? And that used to be like 6,600. Of course, they made the battery bigger to get that. That's important because the [ build up ] materials in a smartphone, in $1,000 smartphone, the battery is becoming bigger and bigger percentage of that. And the space is not getting any bigger. So energy density is at a premium. And for us, having able to produce the AI-1 battery with a strong road map to even higher energy density, these are nice tailwinds for us to get higher ASP over time and access a much larger market.

Can I make a point?

Yes, please.

One quick point on the batteries bottleneck that makes it, I think, obvious. A typical AI chip is bigger than PC chips used to be. And it contains today 100 billion transistors. So you're looking at a chip that's the size of a commemorative postage stamp with 100 billion transistors on it. And that sucker, even though it draws nano-amperes or nano-watts on node-by-node basis, draws amperes of current. And that is the state-of-the-art chip that's already made. They're making now down to 3 nanometers to get it done. So to do AI requires a lot of power because you're using giant chips with a lot of transistors, and that's unavoidable right now.

Yes. Thank you, TJ. Okay. I'm going to turn it over to Ryan now to talk about the financials.

Thank you, Raj. Thanks, TJ. First, a quick housekeeping point. On July 7th, we issued preliminary results, the final print you'll see here, which includes a few positive adjustments. So first, we'll talk about the Q2 results. So revenue, $7.5 million, well above our guidance range of $4.5 million to $6.5 million, as Raj mentioned, up 98% year-over-year. Fifth straight quarter that we've beat the midpoint of the guidance that we've provided. Nice result. Also a nice result in terms of gross profit and gross margin. As you can see, gross margin year-over-year turned positive 31%, really strong quarter on the -- basically on the back of a great product mix with the defense sector. As we move to operating expenses, you can see as well, $28.8 million for the quarter, down 5% year-over-year. The company maintained discipline in terms of spend even while it's continued to execute on meeting the mission of building out Fab2 and continuing with the R&D efforts. Loss on operations, you can see there is a result of that, so an improvement year-over-year. Let's go to the next slide. A couple of points here as well. So on the -- in terms of the net loss per share, on a quarterly basis, $0.13, again, also improved from our guidance range of 15% to 21%. And a couple of points on cash. We finished Q2 with $203.4 million, and that included after making payments, completing the acquisition and additional capital expenditures related to Fab2. Another point that we have here on the bottom is that the Board authorized, which we previously announced, a $60 million share buyback program. As of today, we have not made any purchases under the program, but we stand by at the ready, especially in the face of market volatility. And finally, here, you see our promises for Q3 in terms of the outlook. Revenue on a top line basis, we do see growing sequentially and well above where we were last year. Net operating loss, we see increasing slightly. Really 2 things, less favorable product mix we see in Q3 and increased operating expenses related to the manufacturing readiness scale up. And you see in the net loss per share, we see a range of $0.14 to $0.18 compared to the $0.13 in Q2. And the midpoint of that range, right on par with the $0.16 that we posted last year. Yes. That’s it.

Thank you. Can I go for Q&A?

We will now begin the Q&A session. Please note that this call is being recorded. Before we go to live questions, we are going to read the 2 most highly voted questions submitted by shareholders ahead of this call during the call registration. The first question is, when will Enovix batteries appear in products?

Yes. Thank you for that question. I think as I mentioned, we are sampling our AI-1 battery now to multiple customers and particularly in the smartphone space is where we expect our first products to launch using our battery. And they are in cycling, they are in doing different task. So like I said that takes about at least 3 months for them to do all the test. And after that, once we pass all those test, we expect to get a purchase order from them, and expect the phones to launch. So it’s now in the hands of our customers and we expect them to come out as and all the test and all qualifications is done by them.

The second question is, can you explain the need for the warrant? What will the money be used for? Is it for the new lines or a new facility? How many lines can you fit in your facility?

Will you take that, Ryan?

Okay. Yes, no, the purchase from the warrant, if exercised, will be used to complete the built-out of Fab2. So we have one high volume line that's operational now, and the facility is sized to support 4 more lines. Of course, the funds will be used for general working capital purposes, so supporting our strategic initiatives such as continued R&D development. And I can report in terms a little progress on the warrants. So again, we're pleased with the result. As of yesterday, we had roughly $34 million worth of exercises, I believe, was the data that I've seen. So money that essentially has come into the company soon.

[Operator Instructions] Our first question comes from Colin Rusch with Oppenheimer.

Congrats on all the progress. Can you talk a little bit about where you're at with the customers in terms of specs getting finalized for some of these new products? I know you've talked about the one customer getting launched here later this year, but I assume that you're making some meaningful progress with other folks as well. And when we can start seeing some of those incremental announcements and product finalizations or product specs getting finalized over the next several quarters?

Yes. Thank you, Colin. So specs-wise, the first customer we've had the specs, and that's the specs to which we made the battery and we send them the cells, and they're testing them now. They're actually doing cycle life test, which, like I said, is charge, discharge at various times. Like we saw in the -- as I mentioned in the call, our batteries now, we feel are projecting now that it can go up to 1,000 cycles. And last time when we announced AI-1, we thought it could be like up to 800. Now we progressed further along now. And so we feel good about our technology and our customers are testing that. We've just sent -- shipped samples out of our factory to the second customer now, second smartphone OEM. And those will -- again, I've been holding them back a little bit because I wanted to make sure that we satisfied all the requirements of the first one before we sent to the second one. And we sent them now. And we got the specs from them on exactly what cycle life, what performance trade-offs that they want, and we feel our current technology can meet those. My expectation is that they will test these batteries and then they will give us the exact dimensions, just like how the first customer did on the size of the battery. That they want for the launch next year, because next year phones will have a slightly different battery based on the display they chose, based on the camera they chose and so on. And once we get that, we will make batteries to that, and we expect that to go to production sometime later part of next year. And then of course, we're getting a lot of interest from more customers, and we will start sampling them through the year. TJ, do you want to make a comment?

Colin, I'd like to make a comment from a Board perspective. So I come here for a full day Board meeting once a quarter. I also come here for 3 other 6-ish hour meetings, including manufacturing and R&D. And to me, the process by which you get the spec is more important at my level than the spec itself. Do you have 1,000 cycles? Is the new spec bigger? That's the sausage making. But what we failed at, I failed at in the old pre-Raj Enovix was, we had a set of specs and we went out and made them like engineers. And I'll give you an example. There's a video of me on the website where I'm poking a nail through a battery. And it's really true our battery doesn't light up and turn on fire when you poke a stainless steel nail through it. And it's really true. In our real lab we have a nail poker machine that's got fireproof chamber in it. It's also really true the cell phone guys don't care. And we were working on something that was kind of an engineering abstraction for real, when our customers were doing things different. So, when Raj came here, he went to customers and said -- loved him, literally, I was shocked. He would send back pictures of boardrooms in Asia, with 20 smiling people waving at the camera. There were Raj big in the middle grinning. So we got their spec and their spec wasn't the kind of spec I'm talking about, it was a book. And these guys are world-class users of batteries. And then they sent us some people. And they said, "No, you got to do this. No, you got to do that. This spec is interpreted this way. And they actually did some of the work to validate our spec." I'll just -- back in my example, the state-of-the-art way to set a battery on fire and make sure it doesn't get on fire is you crush it. You actually crush it down and grind it together to start mixing the things inside the battery, and that's a more deadly way of starting a fire than poking a nail through it. And we now do that test. We just finished that test last week in A1 battery. So, point of all this is, Raj is a world-class guy working with customers, and the semiconductor industry was different. If you got Moore's Law and you've got the fastest memory with the most bits and you can meet their price, you win. And in this case, these things are pretty complicated systems, and we're building to their specs, working with their engineers. So, my apprehension that if you build it, will they buy it, is way lower. They're going to buy it, and we're going to build it the way they want it, and I'm talking books worth of data on what they want. That's the difference in the old and new Enovix pre-Raj, post-Raj.

That's incredibly helpful. Appreciate that, TJ. Just thinking about capital equipment and the potential for expansion here, can you talk about preparedness of your capital equipment suppliers to support an expansion? And any sort of tweaks that you're thinking about on the line now that you've been able to do a fair amount of work on testing each of the systems?

Yes. So I think I have mentioned that, a couple of weeks ago I was in Asia, and I visited some of our equipment suppliers. This time around, we are actually -- as I said, I mean, compared to what we used to build in Fremont to what we're building in Malaysia. The equipment suppliers now that we are using are actually people who are used to making very complex semiconductor equipment. This is people -- I mean, I saw a supplier, they kind of figure out how to put really thin copper pins into the [ dice ] at like submicron precision. I mean it's crazy. And these are the people we're using, and there's some very exciting new technology there, and I visited a few of them. And we have some new interesting ways to actually continue to dice the electrodes and make them even faster. And so we started some proof-of-concept systems, and I saw some of the systems being built. That's what I meant when I said we are now making some initial purchase orders towards that. So yes, I think we are telling them and that what we need. And some of the suppliers have actually visited our factory in Penang, and they know where these machines are going to be put in. So in terms of readiness, yes, absolutely. We are starting the work now and very exciting to see the latest advances in semiconductor manufacturing that we are now able to use to make batteries, right? I think that's the key that you will see because we make submicron like micron precision cutting and stacking. So the equipment we use is very precise, and this is what is being done in semiconductor industry, and we're using the same suppliers.

Our next question comes from Jeff Osborne with TD Securities.

Just 2 questions on my side. If you were to win the 2 phone engagements as well as the eyewear, and it sounds like those would launch 3, 4 months after the sampling started last -- or this month. How do we think about the need for sort of working capital growth? Are you already operating as if you're going to win? What are the lead times for the materials that you need to buy to ramp up Line 1 and Fab2?

Maybe I'll jump in there. Yes. I mean I think I'll answer the last part first. We're certainly operating to win. That's our -- how we're operating. We -- obviously, the balance sheet, we finished the quarter with over $200 million at the end of Q2. And -- so we're sufficiently capitalized and we're preparing for the manufacturing ramp.

Good to hear. And then just maybe another question on the balance sheet side, just given the warrant structure was a bit creative and not familiar with most investors. I've had several people ask just in the event that only a portion of the warrants were to be exercised, is there sort of a minimum level that would avoid the need for a future capital raise? I think you mentioned $34 million as of yesterday was done. As you look at the ramp into the middle of next year, if you were to win these engagements and maybe more, how do you contemplate building out the Fab2 completely? Is half the warrant enough to do that? Or maybe just walk us through that?

I'll step in here as well. I don't think the capital warrant would be sufficient capital to completely build out all 4 lines. We've talked about and we provided publicly our estimate of how much those lines cost, which is $50 million to $60 million. But certainly, even adding one additional line, which again, our balance sheet is strong as it is, right now, and the $34 million is -- helps. That's going to get us to substantial capacity, 2 lines running in that factory. We talked about nameplate capacity of 9 million to 10 million units per year. So that's going to go a long way to helping the company meet important milestones.

Our next question comes from Mark Shooter from William Blair.

I was hoping that as you guys get close to commercialization, we could revisit some of the numbers that we were talking about in regards for ASPs. We've seen that batteries have increased so much since you guys have first started talking about pursuing the smartphone market in terms of milliamp-hours. So what are you seeing now in the market for ASPs as we approach 10,000 milliamp-hours? And how do you see the AI-1 being able to get pricing power against the incumbents?

Yes. I mean -- great question, Mark. As I mentioned, right, I mean, if 2 years ago, if we said, 3 years ago or even last year, we said, hey, the next smartphone battery is going to have a 10,000 milliamp-hour battery, I don't think most people would have believed. But today, that's true. And the other interesting thing is, it's not that the battery is getting that big. It's the battery is getting that big across the tiers. It is something I saw when I was at Qualcomm when we started launching smartphone process, and we said we're going to use a high megapixel camera, like a 50-megapixel, 60 megapixel in the premium tier smartphone. And everyone said, okay, premium tier will have a high megapixel. I mean today, $200 phone has like over like 50, 60, 70, even 100 megapixels. So the cameras have just kind of -- and we talked about multiple cameras being on the very high end. Now multiple cameras are across the board. I see the same thing happening on batteries now. What used to be a premium tier battery, which is 7,000 -- 6,000 milliamp-hours, we're now seeing that in the $200 phones and probably go down even more. And then we see even higher capacity batteries at the top. And that clearly, just like it happened with DRAMs, it's dollars per amp hours, right? So the ASPs will continue to move up, and we probably have to redo our models now, but I do believe -- and also redo the factory loading because bigger batteries take more capacity to make. I mean this is a great thing to have, a great situation. And I actually think that the ASP will continue to go up and the premium we get for producing a higher capacity battery in the same space, which is really the key because you can't just keep making phones bigger, which means you need higher energy density. And not only higher energy density, but have that higher energy density with fast charge, with 1,000 cycles and high temperature storage, we benchmark a lot of batteries out there. And the AI-1 is the only battery I know of that can meet all those requirements right now, and we have a strong road map for making that go higher. So, to your point, we do think that the ASPs will go up, and we do think that the premium we can command will go up with AI-1 as we start making this higher energy density reality in these phones.

Great. Thanks for the color, Raj. The next market that you guys are excited about is the AR/VR and the smart glass market. So, in the last quarter, we've seen a few more products come to market or at least start to ramp. And I'm wondering if there's any increased engagement going on there? I know you mentioned in the shareholder letter with the sampling. But can you go over a little bit more color of why you can command more ASP or more gross margin in those products and why the [indiscernible] is so crucial for the AR/VR smart glass opportunity?

Yes. So -- Good question. Two things. First, people have always tried to make another smart device outside smartphone, right? People had tried for a long time. People tried watches, people tried pins, people tried so many things. But none of them really -- the watches took out a little bit, but none of them really took out to a point where you don't have to pull out your smartphone to do something, and you can use this new device. Glasses are the first ones that have the potential to do that because a huge percentage of the population, including people like me, wear glasses. So if you can get more functionality into something you already wear, it's a huge benefit. The other thing is the glasses have cameras and that can look into the world. So now with Gen AI, when you look into the world, when you see what's out there and with the ability to -- also have microphones, you can just do a lot more. And the micro displays, we haven't really seen glasses in high-volume production right now that have a very good augmentation, right? I mean the glasses today is still right now that are in high volume, mostly have cameras and audio and some form of speech recognition. But they don't -- they're not really AR, right? They're not creating that AR experience. And when you create that AR experience, what happens is you have a display inside your glasses that people can't see, but you can see. And we've seen demos of those at CES and other markets. But we are now seeing that the customers we are sampling to are building things like that. And when you do that, here is what happens, the processor has to be on and that, more importantly, the GPU has to be on to create that image on that augmented display. That consumes a lot of battery and the memory has to be on. That consumes a lot more battery. So the amount of energy density you need with true augmented reality glasses where there is a secondary display that you see inside your eyes is much, much more. And that's where we are seeing a lot of traction for our batteries, is the next generation of those. And that's kind of where I believe that the energy density is so important in a small form factor.

Okay. Let me -- at the risk of blowing this thing up, let me -- what else an engineering point to make. So, this is an Enovix picture of smart glasses and all the systems he was talking about are there. Now, if you look at the side of the glasses on the left, you can see the small battery compartment. And if you look over here, you can see a small battery. And then this is a picture of our battery. And so, it's kind of a long, skinny, weird shape. So one thing, the way we make batteries by laser cutting anode and cathode and separating them and stacking them, you can make that thing right away. Second thing is, think about a small battery like that one shown and that skin that's on there, is 3 layers thick. And it prevents the electrolyte, which is a flammable liquid, from getting out. It's tough, metallized 3-layers thick. So when you have that skin, think about making it smaller. So the skin gets smaller and smaller. And if you go through it, the skin gets more and more of the area -- volume of the battery. And at some point, when that battery still is about that big, but it's really super thin that there's nothing left inside, none of the anode cathode stuff that makes a battery. So all I'm saying is all batteries as they get smaller because of the overhead of connecting to them, get lower in energy density. Like these numbers are lower than the cell phone of 900 because of that effect. Point is this, that battery, we can jam a lot of watt hours per liter into that battery because we have compression. The material in that battery is compressed and there's a lot more of it in a given volume because we have a compression system -- a patented compression system, which happens to be made out of steel leading to point two. Point two is it's got metal on it, which means the heat coming out of that battery comes out across the entire wide face, the Enovix face of that battery and radiates into the air. And if you try to make a battery, just stacking battery junk inside of a bag, that thing is a bag, You're just stacking stuff inside of a bag. You generate a bunch of heat. Lithium-ion batteries can get hot and work fine, except you don't exactly want something that's 50 degrees Fahrenheit up against your [indiscernible]. So the batteries are kind of naturally aligned to make high-quality batteries and safe batteries as well for this application. This always actually was a better application. That's why we tried for watches. This always was a better application for our technology where it's a sweet spot, if you will, for our technology.

Our next question comes from Ananda Baruah from Loop Capital.

Two, if I could. Raj, can you remind us how we should think about when you'll -- when the volumes that you guys will get visibility to, how that whole dynamic starts to layer in as you go through sort of finalization of the qualification with the first smartphone OEM, and then you start to go into the second smartphone OEM? Just walk us back through how you guys are anticipating getting visibility to production volumes? And then I have a quick follow-up as well.

Yes, absolutely. I mean I think this is an important question, and I get it on every earnings call. Basically, the way the battery industry works is a little bit different from semiconductor industry. We first sampled standard-sized batteries and like rectangular batteries that we believe is the right size, not exact size, but right size in terms of capacity. And the customers test them and they give us a series of tests that they would like to do, and we do those, and we do that in our factory to make sure that it passes everything and we give it to them. And they test it for a few months typically. And then they give us the exact dimensions of the battery. And they -- one customer has given it to us right now. And then we make the battery to that dimension, then they're going to take it, put it into their phone, and then they're going to do the cycle life test. As I mentioned, 1,000 cycles, and that takes months to test, and that's what they do. And then as TJ mentioned, there's a lot of tests like crush test, drop test and so on that they do, and that takes some time. And then what they do is, typically a particular model of a phone has multiple SKUs that they launch into different parts of the world, some launch in India, some launch in Europe, some launch in U.S. and so on. And then they take a new supplier like us and put us in one SKU and start with that and make sure that they feel comfortable introducing new technology. Then very quickly, it's going to second SKU and a third SKU in that model. Then once we are qualified and we are a valid supplier, very quickly, it moves into multiple models, right? And again, this is not that different with any other supplier. I know that Micron, same thing, we produced the UFS chips, got into SKU, then few SKUs, multiple SKUs before you know it. Finally, sometime later, we're shipping millions, right? And that's exactly what I expect should happen in this case. So we are in the middle of that with one customer. Now we've gotten a lot of interest after we launched the AI-1 platform launch, because now we have the technology. By the way, I don't think people quite realize, I think TJ wrote in the AI-1 press release, the amount of complex R&D work we had to do to take a 100% active silicon anode, put it under pressure, put it under the temperature stress and for the first time get a 900 watt hours per liter with 3C fast charge and 1,000 cycles and storage at high temperatures and work across temperature range. It has never been done before with 100% active silicon. So as soon as we launched that, I got a lot of requests from many customers. And in a couple of weeks, I'm going to be back in Asia. and we will sample this to other customers, and the same process will happen. They will test this battery, then they'll give us the exact dimensions that they need, then we're going to test them internally. We'll go to put them in the phone and then we expect the production to happen next year. That's the same process. And the second time around, it will be much faster because we are a valid supplier now. We're not the [indiscernible] supplier. But the first one takes some time.

And okay, let me -- can I ask you this, TJ? I mean -- well, TJ, if you'd like to jump in also, really, really admit Raj, but any context would be good. To the extent that you're -- you can like responsibly answer this question. So if -- like, let's say, the first call happens, you kind of wrap it up in like 3 or 4 months and then you get inclusion into the first SKU, and then -- so let's say that's kind of going through the fall and then maybe going into next year, you get on, like what you said, Raj, like the second SKU, the third SKU going into next year. Is it -- and let me know if this is off base. Is it reasonable to think that like going into spring you could be what you call, I think you said like a, sort of like a valid supplier like -- and then you could go across more SKUs? Is that reasonable? Is that the right way to think about the timing as distinct from the volumes that you get with that?

Yes, it's the right way to think about it. And like I said, I think the most important thing is this particular shape of the battery, how many SKUs does it satisfy. If we need to change the battery shape, then it will take us a few months to -- which I said in the prepared remarks. We're now reducing the amount of time it takes to make different sizes. So we'll need to make a different size if it's a SKU that needs a different size battery. There will be some lead time associated with that. But that's just the way the progression continues, right? But the thing you got to remember is that this is a huge market. I mean we are talking about 1,200 million units plus market, and we are talking about selling a few million batteries. So the opportunity is clearly there. The first one is a hard one, but once we get there, it will start moving.

Our next question comes from Tony Stoss from Craig-Hallum.

I wanted to follow up on the investor letter where you talked about you've passed the airline safety certification. How important was that? And is there any additional hurdles that you need before you can start shipping in volume to that first smartphone customer? And then I had a follow-up.

Yes. So the certification of 8.3 is a very important certification because without that, it's hard to ship batteries on a plane. That's the first thing you got to remember. You're taking all kinds of exceptions to get batteries out there. The second one is when you make a battery for the first time in our factory of a certain shape, passing that gives us the confidence that this is an extremely safe battery and we are able to pass that. Now there's other certifications we need to do on top of that, and we are in the middle of that. But this is a huge first step for us, that we be able to pass that on a giant new battery like we made for the first time in our factory. So it's a very important step. Like I said, the other qualifications are not -- that the customers do are not just industry-based certifications, but very specific tests that they have. As TJ was mentioning, we got books that big on how they would test the battery. For example, one of the tests they do is take the battery, put it in a phone and drop it from a certain height multiple times. The other one they do is they put it in like a big washing machine like thing and they spin the battery for like many, many times. And they also store it at high temperature. So there are lots of different tests that they do. And every customer is a little bit different, but we do have all those tests with us, and we try to do them ahead of time so that to make sure when they get samples, they surpassed them, but then they will do it again on their own. So it's a pretty rigorous process.

Got it. And then as a follow-up to –-

TJ is going to add something to it.

I just want to talk about, given my background, ramping the factory. I remember when we started Cypress, we started 2 runs a week, 225 wafer runs, 50 wafers a week. And it took us 1 year to ramp our first Fab. So this Fab has got new and it's new equipment. We learned a hard lesson with our first Fab where the equipment wasn't what I would call semiconductor quality equipment. And we were in a place that doesn't make semiconductors anymore. We were in Fremont, California. And we call it "Silicon Valley", but they haven't put a new Fab here in 30 years. So you can't hire people that work in the neighboring company and want a raise to come over to your company. So what we've got going for us is we're in Penang and Penang, Malaysia is the center, the epicenter of semiconductor assembly and test. So when you look at semiconductors today and you see modules that big with chips integrated into them, chips stack vertically with each other. Those are the kind of dimensions that those people live with. And we have people from the semiconductor industry in our plant. I'm really impressed with the quality of people in our plant. So we've got good equipment from a good manufacturer, costs more money than I wanted and good people. So we will ramp and that -- we're not ramping to our own guess specs. We're ramping to their specs, and they're looking at stuff real time. You can send them stuff and say, does this pass fire test, and they'll give you a quick read. So I don't expect we're going to have a catastrophic problem like we had in the Fremont Fab. But ramping a Fab is a big deal. And there will be a surprise or two. But when you're really close with ultra-high tolerances and good manufacturing quality habits, those are things you can accommodate. All I'm saying is, don't expect this is like, make an ice cream and now you say if I buy 5 more ice cream makers, we're just going to ramp it. We've got a battle. We'll have more people in the plant than we have in the rest of the company. And we've got -- but we've got good people. So I just want to warn you, we got to do that. But this little pretesting we do when we get data from them and they tell us about the crush test as opposed to the nail penetration test. That's going to make the probability that we get through it much higher, not much higher, it's going to work. You put your head down and you make it work. We got this much money involved. We're going to get there.

Thanks for the color, TJ. If I could just sneak in one more follow-up just related to the eyewear customer. Raj, a few of the eyewear companies that are launching now, the ones that are expected to be high volume over the next several years, how difficult -- since they've secured other battery suppliers already, how difficult will it be for you to displace those folks? Do you have to wait for many generations? Or could it be in the next year or 2? How quickly before you think that turns into a real revenue for you?

Yes. I mean, look, we are sampling now, right? I mean we're sampling batteries now and we're testing them. And I think -- if I think about the eyewear, it's a market that I'm very familiar with, the whole AR/VR space. I used to run that group when I was at Qualcomm. That's a market that's just happening now, and there are many, many more iterations to go before we see very high volume of that product. There will be different versions of it, by the way. I mean I think, if you think about now, it can be sold as an electronics device. It can be sold as a fashion device. It can be sold as a medical device. It can be sold as an industrial device. So it's in the very early stage of that. So I believe that there'll be many, many versions of that, and there will be products that enable lifestyle, that enable productivity, that enable utility and so on. So I believe that the heart of all of them is higher energy density because as TJ mentioned, they're very small batteries and the demands are high. So I believe that when we produce a battery that's got much higher energy density than anything else and much safer, it shouldn't be that hard to convert because it's very early stage. And people love it. And these people we have given samples to love the batteries we give. It's a long ways to go to get into multimillion units, but we're happy with the technology, and we're happy with the position we have and the feedback we're getting from the people we have sampled, right?

Our next question comes from Derek Soderberg from Cantor Fitzgerald.

Two questions from me, quick ones. Raj, how much are smartphone OEMs paying for batteries for the flagship smartphones today? And then second, on the industrial handheld opportunity, it sounds like maybe an RFID scanner label printer. Is this opportunity sort of in the millions or so of devices per year? Is that the best way to characterize that? That's all for me.

Yes. I mean I -- how much they're paying depends upon what the energy density is. It's actually disproportionate. If you -- when you increase the energy density in the same volume, the premium actually goes up higher. So what I mean is if you produce a 900 watt hour per liter battery versus 700 watt hour per liter battery, you get a different price per amp-hour, right? I hope that makes sense. So rough numbers, I would say $1.50 to $2 is probably what people pay per amp-hour. So if you got 8 amp-hour, that's probably -- you can think about anywhere between $15 to $16, right? That's probably roughly what you can think about.

Our next question comes from –-

I'm sorry, let me answer that second question. The second question –-

Yes, I'll jump in on the IHH opportunity. That's absolutely customer engagement, we think has potential for multiple million units on an annual basis.

Yes. One other thing I'll say about that market, the IHH market is, again, a market I'm quite familiar with. It uses the same processors as smartphones, is that there is also an opportunity to selling replacement batteries because that's a market where people can launch a product. It lasts for 4, 5 years, but people also buy batteries to replace the existing battery because they work on contract. I mean think about the UPS person that comes to your house and you sign something. So there's a size of the market and then there's a size of the replacement batteries for the market, and that's a millions of units.

Our next question comes from Gus Richard from Northland.

I was just wondering as you get closer to buying, production, will you wait for as purchase order to start building inventory? Or will you stack some inventory, I mean, I know because these are each custom devices in terms of dimensions?

Yes, no. I think some people have a little bit of misconception in terms of the sampling we are doing now. This are not [ 1

TJ, [indiscernible] add a little bit to it, yes.

Well, my comment on that is from a Board/money perspective, you can't pile up inventory. First of all, the beauty of this machine is auto line, is linear. If this part stops, everything in front of it stops. You can't pile crap up in semiconductor fabs for a long time, that was a problem. So, this thing will have a cycle time less than 30 days. And if our purchase orders have 30-day take-or-pay window, which would be a typical purchase order. In the semiconductor industry, we never could get that. We never could start a [ wafer ] and ship product in 30 days. In this industry, you'll be able to do that, especially with the equipment that's automated, linear equipment, and especially because it's custom stuff. So no, we're not going to need the inventory. We can't afford inventory.

There are no further questions at this time. With that, I'd like to turn it over to Dr. Raj Talluri for closing remarks.