Weebit Nano Limited (WBT) Earnings Call Transcript & Summary

This meeting is being recorded.

Okay. I have this thing on because we're also recording this and having this on Zoom, so there are a lot of people on Zoom. Thanks, everyone, for coming. I appreciate, I guess, taking the time off during the working day and coming down into traffic and all of that. So it's great to see some familiar faces with that here. So welcome. I like to walk around, but I guess I'll stay close to this mic, so it will work better. So I guess, some of you guys might have seen already that we just released some news, and I'll talk about it during the presentation. I think we had a very important milestone today hitting the 125-degree for the automotive level. But starting from the beginning, you guys know we bid. So I'm going to go very quickly through this slide and really just focused on the fact that we're continuing the work, continuing to develop the number on the patent here in the bottom box is going up, and we're working more also on getting our IP protected. Except for that, I think you guys know this slide by heart by now. We've been doing a lot of work as we bid. I think if we look at the last half year, there's been really very good progress, and I'm very proud of all the stuff that the team is doing. We got our wafers back from SkyWater, that was an important milestone. And kind of fast forward, if we look at the June announcement that we actually, it takes time to qualify, and that's something that I think you've already learned if you need to put something in the oven for 1,000 hours. It just has to be in the oven for 1,000 hours, and it takes, however. The few months that it takes to do it and then you need to do it on multiple rounds and everything. So these are long processes that we have when we're dealing with semiconductors. We're very happy to have that qualification done. Now it's much easier when we talk to SkyWater customers, of course, yet another big obstacle has been removed and makes all the discussions that much easier. We've -- in the beginning of the year, we talked about how we are much more friendly to the environment than other technologies. That's very important, I think. People don't -- people are so busy now with, okay, what's happening with commercialization, what's happening with the technology, et cetera, et cetera. But I think it's important for environment. We don't use rare earth materials that we use less power to manufacture our ReRAM relative to other technologies. At the end of the data so we're talking about how warm it is outside. Well, that's part of the thing. So that -- and then the other study that was done by the University of Florida that was released in March, showing that we're immune to the radiation. That was another important thing to also show that our technology is robust. It's good a very wide range of applications. And especially the aerospace guys, this announcement was very important for them, and a lot of the SkyWater customers are in that military and aerospace domain, and we're looking at it. So those announcements were very important. Of course, taking out on SkyWater and 22-nanometer with a major, major milestone for us. GlobalFoundries is one of the [indiscernible] foundries in the world. And doing that tape-out was very important. We're still waiting for the wafers and they're going through their manufacturing process is long. And unfortunately, now that you don't always manage to get the top priority in the foundries when they have so much production happening there and the things that generate money for them, but we're getting through that, and we have a good relationship with them moving forward. And obviously, the rates that we did was very important for us. I guess you saw the 4C, we now have $88 million in the bank, which we're very cautious about how we use them, but it's very good to know that now we have the funds to enable us to go forward at least 2 years, 3 years forward and do what we need to really ramp up the commercialization. So these are all things that happened over the past half year. This morning, we announced that we achieved qualification at 125 degrees. This is really another very important step. There are several industries that for them, this is a basic requirement automotive, the temperatures inside an engine, industrial IoT. When you have on the manufacturing source, so some of the machines on the manufacturing stores reach very high temperatures, and all of the monitors that keep monitoring the machines are operating properly, et cetera. All of these things, they need to operate in high temperatures. So there are a lot of applications where temperature is critical, and achieving 125 degrees with retention of 10 years, that's a very important milestone, and we're very proud to have achieved it. And talking just for a second about automotive. We have, I think, about a year ago, Infineon, which is one of the leading automotive suppliers announced that all of their future products will have ReRAM. That was a major announcement. Yes, they are working with TSMC, but the fact that the whole automotive industry saw these guys are committed to ReRAM, it really shook things up. And so today, we are having discussions with quite a few of these automotive suppliers that for them, a continuum is doing need to do it, of course, as well. And so that gave us a big boost. And now with this qualification, it's going to make life much easier when we talk to them. They saw the supporting data that we already had up to now that showed them that we shouldn't have any issue to reach 125 degrees, but the fact that we now officially have it is, of course, much better. So these are all important milestones. And by the way, moving forward now and if you look at the end of '23 and what we're planning for '24, there's constant R&D work. There's constant improvement. We're going to continue qualifications to qualify for better endurance for higher temperatures, for more and more extreme demand, and this is an ongoing activity. There's going to be a lot of work continuing to improve the technology as we go into the mass production phase. The quality levels constantly need to be improved. You always need to have fewer defects. And you always have -- I mean there's never a 100% yield but you want to have as few as possible defects and of course, we're constantly working on improving that and improving the technology and the functionality and so on. I hope you think is working properly. So that's the key for us. And what everyone is thinking about, we are very, very focused now on bringing on board the customers and the founders, we have this triangle. We have, on the one side, the foundries that we're working with, and we're pushing forward on that. And on the other side, we have the product companies, what's called the fabless companies, and we're making progress, and I'll be talking about it as we go through the presentation. So regarding the team, and I'm showing the slide here because as everyone knows, Fred Bart resigned, and we have Mark who stepped as a nonexecutive director now. So he's filling in that post. We are looking for an additional Board member, and we're having good discussions with candidates, and I hope we'll have news on that in the near future. And of course, I should point out here, and I should have pointed it out already, from the beginning, we have Ashley Krongold here. He flew in from Melbourne to join us, and he's been with the company actually from the very first day. So way before I joined or that or anyone. And he's been very active in helping set up the company in the early days. The management team, you guys already know this slide, so I don't need to really talk about it. I will comment we just hired another key person to the team,[indiscernible]. She started off as a young engineer in power. I don't know what years ago or 20-something years ago. She grew -- she went up the ranks until she became a manager of one of the fabs at Tower. So she really knows that business very well. She knows exactly what goes on, on the floor of the fab, and we're going to have her support us in all the engagements we have with the foundries. So that's a key hire that we have. Maybe this is a good point to also talk a little bit about hiring and the team and so on. So we, the team today, is about just 30 people. So we've been growing slowly. I'm very [indiscernible] about how we use the funds and easy to say, oh, we have money in the bank. We can really go and hire a big team. And that's really a, but I'm very cautious because I know how fast the money can disappear. So there's a balancing act here between hiring people and having projects, so we have a very high level of challenge that we will be closing these deals. And I know that these projects are coming. So right now, the basic approach is if we have really good people that we can hire, we're hiring them because we know we'll need them. Once you close the deal, they might not be there and then your stuff. So I'm very happy that we hired last. We also had, I guess, our fortune, their misfortune, an MRM company in France went bankrupt recently. We knew some of the engineers there that were really, really good engineers. So we took the opportunity and hired 3 engineers there. We know we'll need them as soon as we close some of these deals. So that's kind of our approach slowly hiring, not going crazy, but if we see really good people and the key point, we don't compromise. We are really focused on hiring the best people. And I think I can say we're, if not the best ReRAM development team on the planet today, we're clearly one of the top 2 or 3. I can't really think of any company that has such a strong and focused team. By the way, in the R&D, we have more than 10 PhDs that did their PhD in non-volatile memory. So I mean, it's really a very, very strong team. And I can assure you I wake up every morning. I can't believe that we have this group. So amazing. Now in terms of what's happening, I guess, in previous presentations, I was talking about the level of investment of governments and of companies in semiconductors showing the importance of semiconductors. I think everyone knows already just how critical semiconductors are to the world, and anything you touch today has semiconductors in it. The people here and some people here, I remember even when I was talking in 2019, and I would say, cars are having almost 100 electronics components in them. I think last year, I was already saying we're getting close to 1,000 components in a car. Well, I just read that the more advanced cars today are already 1,300 electronic components in them. I mean it's just going crazy. And one of the key drivers, and I think everyone's been hearing about it now and ChatGPT and all of these things, AI is a key driver now in the explosion of the need of semiconductors. So -- and we can see here on the left, it's a little small, the chart basically, it talks and the chart on the left is how many devices in the billions of tiny ML, what's called machine learning, devices are in the world. You can see the growth here going up to 14 billion in 2027. It's huge growth. I mean in a car now, autonomous cars, they have so much AI in them already. All the sensors are collecting so much data. They bring it into the processing units. In the past, a lot of this was loaded to ascend to the cloud and to the big data centers. But in a car, you need a very fast reaction. You need to -- I mean, if someone jumped in front of the car, you need the sensors to catch it, the process to process it, to react the braking system and so on. So everything needs to happen very fast, you need to have a very fast memory like a ReRAM that you can work with. You wanted to -- these cars are operating now on batteries. You wanted to be lower power memory and stuff. So we have our contribution into this. And AI no matter how much memory you give AI, they always want more. I mean it's really endless. The more you give them, the more processing they can do, the more they can analyze. It's unbelievable what goes on there. And they're going kind of on the right side here. You can see the process that's been going on. Until now -- so first of all, AI, because of the quantities of memory they need, they need to go to the smallest geometry as possible because that's where you can put more and more and more on a single chip. Say, they're really the ones that are pushing the industry 7, 5, 3 nanometers. Now they don't have a nonvolatile memory. So what they do, and you can see here the two-chip solution, they basically have an external flash that they kind of glue on the chip and have that communication, it's a very inefficient way to work. I mean it's slower, it's more power-hungry. It's the security level, you can use drop on that communication and stuff. So now there's a movement [indiscernible] ReRAM or internal NVM, there's also MRAM there at this point to go and start using RAM. Moving forward in the future, we talked about it in the past about more and more computing and how the memory can be actually in-memory compute and much more advanced and ReRAM is really great. The ReRAM memory cell, I remind you guys, the ReRAM memory cell operates in a very similar way to a synapse in your brain. We have already for several years now in reMORPH demo that we show how ReRAM can use for reMORPH in computing. And we have agreements with quite a few research institutes who are using our ReRAM to develop more reMORPH in computing. So we're definitely involved in this. This is long term, okay, but it is something that we're -- we have our eye on. And I think in the future, it's going to be a major activity for Weebit. You guys know the slide, all of the advantages, Weebit, better endurance, energy efficiency, speed and everything. And I put it up because this is what we're so proud of. The technology, it's not just a small step forward from flash. It is a big jump forward. It really gives us a very strong position. When we go and we talk to companies have plans beyond the fact that, of course, they've been going down to be lower nanometers and more and more and more projects are now at the [ 28, 2020 ] goal. But all of these other advantages are very important for them. So we are in a very good position. And when you look at the different markets. And we have here just 6 markets as an example with the mix signal -- power management, by the way, has become a very, very hot topic because today, so many things are operated on batteries. So you need to have those chargers there, and this is really a major thing. By the way, they also need high temperature, not 125, but sometimes, they even get to 125. You have the IoT, which is everywhere else these little devices all over the world, the Edge AI. Again, that's becoming now an important thing. Many, many devices want to do some AI locally before they send everything to the data center. And this is becoming more and more popular. So we have the advantages there in the automotive space, the aerospace, the medical. In each domain, it's not just that we have one added value. It's really we have a lot of added value for them so we can have very good discussions with these customers. You know that we're working -- the big focus of Weebit today is on the embedded space. And we -- we're working with the companies that are developing their system on a chip, and they embed our memory into their system. And that's where the big focus of Weebit is. Again, we have very good discounts with these guys. But as we talked in the past, Weebit is not for getting the discrete memory market, stand-alone memory. And you know that there's been a lot of work on improving the selector there, and that's ongoing to get that happening. Again, that's still a lot of R&D work to make that work. So again, in terms of the size of the market, once a year update their numbers, and I consider them the most reliable, the best source for that analysis of the memory space. By the way, this you can actually see here. I don't know if you guys can see it. Yes, you can see the cursor here. So they are still quite conservative saying that only this very small percentage, 3% of the nonvolatile memory market, the embed non-volatile memory market will be using emerging memories basically ReRAM and MRAM, most -- there's PC here and stuff, but PCM is really a very small element. So there even when you're thinking about just 3% of the market, it's already 1 billion units by the end of '28 or by 2028. And they're predicting that 60% of the wafers produced in 2028 will be -- of those that use in the emerging memory will be with ReRAM. Now remember, MRAM has been out there in the market for several years already. So they already have been there, they have already established themselves, and still, you can see how ReRAM is starting to take over the prediction is that ReRAM will take more and more of this market. ReRAM have so many advantages over the MRAM that you always now showing it in a very clear way that they are seeing how ReRAM will be taking over this emerging memory market. And I think that's maybe the most important point on this slide. Again, once we get into position and start ramping up and people see devices with ReRAM in mass production, the devices are out there, the consumers are using them and that is working in the field, the 3% here will start rapidly taking over a lot of the rest, and that's where the big growth. I mean you can just see the size of the market that we're talking about. We say the embedded -- I'm sorry, the non-volatile memory market is USD 124 billion is expected to be by 2027. You can see the growth that the market is now still smaller, but the potential growth is crazy, and that's really what we're going after. You need to have some patience because this market grows, it's not software if you just click a button, you compile something, you have a product. As you already know, it takes us time to manufacture, qualify at all, but the numbers are there, and you can see where we're heading. I want to focus more on business model. Now that we're going into commercialization. I talked about it in the past, but I want to elaborate a little more because now it's really very important to understand how this whole thing works. And again, I'm focused here on the embedded market right now. So you know already that Weebit is producing technology, IP of ReRAM. Now we need to manufacture it. So there are foundries. And just to make sure that everyone has the terminology right, I'll try to go through this and try to explain it. The manufacture semiconductors in what's called fab, fabrication facility, but we just call them fabs. Fabs are very, very, very expensive to set up. And the more advanced fabs today are tens of bids of dollars to set up. The number of companies that can manufacture at below 20 nanometers, I think it's 7 or 8. It's definitely less than 10. So not many companies can afford to set up such facilities. It's very difficult. So you have what's called foundries that build fabs for the general public. They basically offer a manufacturing service to the public. So because even Apple, Facebook, Google, Qualcomm, Broadcom, Cisco, they don't have enough money to set up their own fab. It doesn't make any sense for them to do it, so they go to the foundries to manufacture their system on a chip. So these companies, the Qualcomms and Ciscos of the world are called fabless, and they go to the foundries that have the fabs to manufacture. There are not many companies that are called IDMs that have their own fabs. So Intel manufactures its own Pentium processors, but that's the exception. So there are not many companies manufacture their own products, Intel, TI does it at higher as larger geometries and so on. So that's kind of the terminology that we use. Weebit needs to have foundries that can manufacture our era. So SkyWater was the first one that we engage with. If you remember in the past, everyone was asking, why don't you engage directly with Samsung or Intel and I said, it's better to engage with someone small closer to your size, you can look them in the eye more and cooperate better. I'm glad we did it. It helped us really ramp up, transfer the technology, go through this once so that now we have that experience and we now qualify that SkyWater start working with them. And that enables us now to talk to the big guys, and we are working with the big guys. The foundry in the general mode, and I'm not talking about the first or second deals, but once you start working with the foundries. For every geometry that they manufacture because most foundries don't do just 1 geometry. If you look at a GlobalFoundries, they go from 130 to 90, 65. They have a lot of fabs at different levels. They have the 22-nanometer that we talked about, but they're also working at 12 nanometers and at 28. So for each one of these geometries. We need to work with them to transfer the technology to that fab. We will require them to pay us a license fee for licensing the technology. We will want them to pay for all of the engineering work that needs to go into doing all of that transfer. And I guess you know by now that also were standard products, standard tool, standards, everything, it still takes several months, at least, to transfer the technology to show that it's working. Each one of these fabs has slightly different equipment. They buy maybe from different manufacturers or they use them in a little different way. This slide has changed has impact on the manufacturing. So we always need to transfer what we call the recipe and adapted and make it work in that sense. So they need to pay us a license fee. They need to pay us an NRE, which is for the engineering work, nonrecurring engineering. And then we finally, have it working in the foundry. Now we need to get the fabless companies, the product companies on board. The normal business model in the industry is what you see here, which is basically the foundries take our technology, and they put it on a shelf for their customers to use. So foundries care about selling wafers. That's all they care about. That's where they make their money. And they just want to sell as many wafers as they can as fast as they can. That's their focus. In order to do that, they want to make it easier for their customers to develop their System-on-a-Chip so that they can develop them faster and better and manufactured by wafer. So they have this -- I refer to it as a supermarket where they have on the shelves, they have several different types of processors and WiFi and Bluetooth and all these other things and they're going to put our ReRAM up there as well on the shelf so that people can use. So the connection between the product company will actually be to the foundry, they will be paying the fees to the founder because they're engaged with the foundry anyway for the manufacturing. And then the foundry will be paying Weebit the fees, the royalties, the license fee from the customer and so on. So this is a very common business model in the industry. I like this for the smaller companies where, I mean there are many smaller companies that require support and stuff, but the ROI is not really there. You need to invest more in supporting them and the fees that you collect, the royalties are not that big. So it's really -- it's easy for us to have the foundries take care of those guys. And that leads Weebit to focus on the big product companies. And so there's the second business model where I call it the triangle where Weebit have the same level of relationship with the foundry. However, now with the big fabless companies, we even engage directly. I am a big believer in giving people the best support possible, in making them as successful as possible with your technology. If Weebit works directly with the product company, we know exactly what their needs are. We're working with their engineers. We can tailor our solution to their specific needs and really give them the most optimal solution, give them that added level for, they will be happier. They'll come back to us and we can really make it a very successful relationship. So in that case, the product company will already pay Weebit directly the license fee. They'll pay us for the engineering, and the royalties will come directly from their revenues, okay? So I'm making sure that our agreements with the foundries will include both business models so that we can choose which one we want with the customers. A lot of the smaller companies will be using the first one, the previous model. The bigger ones we want to work with directly. So that's to give you the feel of the business models here. So kind of going back to the technology a bit. We've qualified today, we just announced it. So we qualified for 125 degrees. We're very happy with that. I already talked about automotive and industrial IoT. But this is really a very exciting, and we have several of the foundries, which told us that this is a basic criteria for it. SkyWater, I guess you know we're making progress. Right now, the big focus is working with their customers. We are, by the way, bringing some customers to them. Some companies that were not working with SkyWater in the past which runs ReRAM. We're now bringing the SkyWater. In other cases, Tidewater introduced us to their customers. We're now having 3-way discussions with those customers. So we're working on it, as I explained in the past, and I gave you the example of this bridge that we built with the bungee on it and we go to tell people, "Hey, you're the lucky guy. You're going to be the first one to jump from the bridge." And yes, I mean, it is exactly that. You see, there's no one else who already manufactured with the ReRAM. We can't see any other mass-produced products out there with ReRam. Unfortunately, not yet. There will be very soon, but not yet. And so kind of see that normal hesitant human nature, but we are making good progress because they are seeing the results, and we are removing these hurdles. As we go, the hurdles are being removed and the qualification with SkyWater was major hurdle that we passed. So we are expecting to close agreements with customers in the near future. We were hoping to already have it in our hands. It's taking, the economy is also a little difficult, and companies are more hesitant about making these big steps, but there's definitely progress there. And I can assure you, we're working very hard to drive this and to get these deals done. Going down to smaller nodes. We are working now already with other foundries, some of the big foundries, on smaller geometries. This is something that we keep pushing forward. So R&D work is constantly going on. And of course, the foundries, every foundry that we engage with is still there and involved, and we're pushing forward with them. So there's really a lot of interest. You hit some things that are completely unrelated to what sometimes with these this work with the foundries, with one of them. We were already working on agreements and stuff and the lawyer quit and suddenly, they're looking for another lawyer. And until they find one and until they ramp them up and whatever you some those a few months, I mean frustrating. But sometimes it's those kinds of things. You have a foundry that says, yes, everything is standard, standard, tool standard. One of the tools that you want to use is at full capacity in the foundry. We need to find a solution for that. How are we going to deal with it? And then, so there are a lot of things, the level of complexity in these agreements and in these engagements is crazy. I can just tell you, we're making progress. We do plan to get an agreement signed, at least one agreement signed with one of these big guys in the next several months, and we're doing our best to push it forward. But everyone is -- the level of excitement for ReRAM is there. I talked about the discrete market. Up to now, I've been focused on the embedded all the time, but we are looking at discrete, and we have several cases where people came to us and said, your existing selector is already good enough for us. We would like to have a discrete chip. We just have your existing embedded technology, but on a single chip in a larger array on a single chip. The NOR Flash market here. It's a little niche market. It's only USD 3.5 billion okay relative to 124, it's a small niche. But hey, $3.5 billion, it's a pretty nice market for us. So we see that we have very good advantages over with our existing technology, and we are looking at doing these things. Again, there's a lot of work happening. We're focused, our priorities right now on getting the deals with the foundry, getting an embedded customer on board, but there's definitely work also going on with some of these customers on how can we get a discrete device out there and Weebit will have its own product also to sell. So a lot of work, and it's not just NOR Flash, we're looking at the others. So it's very interesting. That's more or less it. We're still working very hard. I do believe that will even have some initial -- not anything big, but we will have some initial payments that we should be getting. During this year, already this calendar year, again, we're pushing for that. I think that with the level of engagement we have with the different players, I believe we will be able to show some initial payments this year going into 2024 and starting to ramp that up. We're still focused on getting that Tier 1 fab and a customer, at least, one. Hopefully, we're working on more there and continuing to improve qualification all the time to higher levels, continuing the R&D work and scaling down. So the team is very busy. It is a very exciting time, and you just need to hang in there and then really keep focused on just closing these deals. I mean that's what we're doing now. And I hope to be able to announce a deal as I say here, before the end of the year, we should definitely have some announcements. I think that's it, just a summary here. So I'll open it up for questions.

Got a microphone here for if we need it, just so people online can hear.

Yes. I realize these discussions are very complex because you need you're obviously talking about multiyear deals when you see deals, could be for decades.

Yes, when you engage with foundries, it's very long term.

What I'd really like to know is can you give an indication of how many you're talking to in terms of the IDMs and the fabs we talking 20 or 30 different companies? Or is there any way you can give us an indication so far?

So the number of foundries in the world is small. Okay. So that's, I think, and I started off with. I think I even said the number of foundries that can deal with less than 20 nanometers is a little more than a handful. If you look at even below 30 nanometers, the number is much larger. It's still around 10. So there aren't that many foundries, and the challenge. I mean, it's not like I said, "Oh, that guy, there's your put up focus on someone else." You just have to keep pushing with the few that you have because that's what you have. I mean when you look at these lists here. These are the leading guys. We're talking to more than half of them. We're really focused on this list. This is already a lot. I mean each one of them -- when we will engage with one of them, that's already going to be a huge project for Weebit. We're going to have to put a lot of resources into it to make sure that it works and it's successful, and there's a lot of work here, okay? You guys can see already from SkyWater, from the moment we signed an agreement with them until we actually have qualification at 85 degrees, that already took us more than 1.5 years. So it's a big project. We are engaged with these guys who are already working where we've done evaluations. We're doing evaluations with them. It's really intensive work with these guys, but there aren't -- even when you look at the very large geometries, there are tens of foundries in the world, not hundreds and thousands of them.

So we should be very comfortable you're talking to a lot then.

Well, we're talking to a lot -- a large percentage of the market. Yes big. Yes, we're talking to many of these fabs.

Sir I think it's also important to note that it's also they're talking to us. We're not chasing them anymore. So when TSMC came out, was it last year that they came out and said they've got ReRAM, almost 2 years ago that they've got ReRAM? All the others then jumped and went, "Oh, okay, well, with TSMC, the largest fab in the world has come out and said they're qualified ReRAM. They've confirmed that ReRAM is a solution for them, then everyone else started not just being chased but coming to us and saying, when will your ReRAM be ready because TSMC will not give their ReRAM to the other foundries.

Yes. I think that's a really important point. Some of the big names here, they called us. We didn't call them I think that's a very important comment that Ashley is making. It's not like we're begging them. Please work with us. Some of these big names, they called us. So those are really nice.

Yes. Thank you, Coby. That was a very informative page. So would you be willing to say, of that list of top 10 foundries, how many of them are you talking to?

Well, I think the fact that I'm saying that half or more than half is already a good indication, I won't be going into specific names.

I don't want specific names. I was just wondering about the number.

No, we're talking to a lot of the names that we see here. Yes.

So both in the foundries and the IDMs?

Yes.

Just a follow-on from your previous answer. Like obviously, we understand the importance of Weebit and signing these deals to the foundries. Can you talk to us a little bit about what it's actually -- and like was or the foundry? Like how are you pitching to them? Why they're calling you -- why are they calling you? Because obviously, they've got a ton of workload that they need to do and squeezing you guys and it takes time, so what's the big payoff for them? And what's the reason why they're going to want to sign quickly?

Yes. Actually, a very good question. For the foundries make money by some way, okay? I always have to remind everyone because people sometimes get distracted by this. They have specific teams which are called the enablement team, okay? They want to enable their customers to do designs as fast as they can in the most efficient way so that their customers come to them to manufacture products and buy more wafers faster, okay? These enablement teams, their job is to go and find all of these things that can make the life easier for their customers, okay? Now I talked in previous talks, I went into detail of what exactly the embedded means. So because everything has shrunk so much in semiconductors, people can put a full system on a single chip. So they put the processor. They put the WiFi Bluetooth and sensors and all these other things that may need memory, okay? Now the foundry, if they will not have because when a design company, a fabless company comes with a system on a chip, if they will need to design their own processor, their own WiFi, their own Bluetooth and everything, all take them. I don't know how many years to do it. So they are looking for companies like Weebit that provide IT so that, I mean WiFi is a standard. Everyone -- you don't need to do your own WiFi, just go and buy WiFi that exist and you put it in there. You want to focus on your core logic, your core smartness in that system. And all the rest, you can bring from the outside. You don't need to invent -- reinvent everything. So for the foundries, that is the key. They want to bring all of those other elements. The standard processors, the standard WiFi, standard Bluetooth and have it on the shelves there so that people can just take those at their core logic and going and manufacture. If they don't have ReRAM and a customer wants ReRAM, they'll go to another pound, okay? When Infineon announced that they will be using ReRAM all their future devices, and now the whole automotive market is looking at it and saying, ReRAM is the best solution for us. That's what we need for our product. If some foundry doesn't have it, automotive companies won't come. So that's their interest, and that's why they are really pushing forward and looking. When TSMC announced that they have ReRAM, all the other foundries, they have to compete. They have to have that technology.

And I want to add when TSMC came out and said that they had ReRAM. they had it designed and -- but from a small little Taiwanese company. Think of TSMC as a supermarkets. They've now got TSMC Home brand, ReRAM, sitting on their shelf. But the minute that a IDM, a product manufacturer comes and says to us, we want to use your ReRAM but we want TSMC to have it on their shelf. TSMC will say, "no problem. We just want to sell wafers." If we can sell more using ReRAM, Weebit's ReRAM, then we'll take our home brand, put it on the shelf that people are less likely to see and put it Weebit's ReRAM on the eye-level shelf. So it's very much a commodity-driven thing for the fabs. So we've now got the fabs who need ReRAM because TSMC have got their own, and they can't use TSMCs, and we'll now have the product manufacturers saying we think your ReRAM is better than down the track. Because your better than TSMC, we want TSMC to have yours on the shelf.

[indiscernible].

So thereof now, we don't know. The answer is we don't know. We know that there is already one chip with ReRam units from TSMC and iPhone, okay? Actually, that's right now the only product in the world that I know that has the -- that's in mass production that has the ReRAM from TSMC. They still are pushing it forward. Infineon might already have it in some products, they've been working on it for quite some time. It takes time to roll these things out. That's -- even big, huge TSMC takes time to roll these things out. But it is -- the market is happening. You could see the numbers from the old. It's now rolling.

When it happened, when TSMC came out and said they've got ReRAM, just for a little blip it was like, oh, okay, so what does that mean for Weebit? We were celebrating with champagne. We're saying, okay, they've just confirmed that our ReRAM is going to be and is a solution, not will be or hopefully will be but is now a solution on the market. So it was a fantastic day for this company.

Welcome back, Coby. I have a question more regarding the capabilities of your implementation teams. So if, for example, you signed a customer to before Christmas, you would all like to see, how really your implementation teams? And how secure is your ability to get more end up this day?

Yes. So that's a good question. That's -- when I was talking about our team and the hiring plan, that's really that delicate balance that I'm working on. So first of all, we have a very strong team. really, as I said, I think it's the best in the industry, definitely. Maybe TSMC say, I don't know how many people they have on it. I would think, except for TSMC anyone comes anywhere close to us in terms of the strength of the team, both if it's revised its process if it's analyst and digital design across the board, we have stellar team. We've been hiring cautiously key people, anticipating that we are going to close such a deal. So we have the basis we can deal with 1 or 2 customers the team can deal with today. The team that we have today, right now, they're working on improving the technology and the product that I can immediately shift them to a customer project when it comes. We also have quite a few contractors that are really good contractors that we've worked with in the past. And in case that we have too much work in the beginning, we can offload things to them while we hire more people. So we have the plan on how we can deal even -- and hopefully, I really wish that we will have that problem of having 3 and 4 and 5 customers come on board, and we really won't have enough resources. We know how to deal with it. And one of the things that I kind of say different countries have different cultures, the one thing about the Australian culture that whoever gets to the we're really is the word impossible doesn't exist in the vocabulary. So when things happen, we find a way to deal with it and we find a way to make it work. And we will -- even if I have that huge problem of 5 customers, we'll deal with it.

Great presentation. And I might be a little bit greater possible to take two questions, if that's okay. The first question is regarding if we do get -- when we do sign customers. So with the fabs it's taking what maybe 12 months to 18 months to get everything completed and ready for production, but right, for the customers, we think the same time frame from [indiscernible] to finish?

Yes. Very good question. I wanted to say do it in the presentation, so thanks for reminding me. Yes, that's another key thing that I really want everyone to understand how this thing works. When a customer engages with Weebit and makes a decision that they want to embed our technology into their chip, now I mean, it's not just Weebit, but they need to make decisions on which processor, which Bluetooth, et cetera, et cetera. And now they start working on that System-on-a-Chip. The average time to do such a design is about -- some simpler ones can take maybe half a year, some more complex 1 than take 1.5 years, but a year is a good number to have in mind. Once they have that they need to go to manufacturing. And as you already know, that can take several months. I would say, half a year is a good number to have in mind. And then they need to do testing and they need to do qualification. And again, I said it in the past, you can't have a baby with 9 women in 1 month, 1,000 hours is 1,000 hours, and it's going to take 1,000 hours until you get the results. So again, that's about another 0.5 year. So in total, for a project from the day that they decide which IPs they want until they actually start mass production, 2 years is a good number to have in mind. So the ramp-up takes time. And that's why you see here also when you all is talking about how many units are going to be using emerging memories in 2028, and they're saying it's only 3% of the market, it's only 3% of the market. It's because of that delay. I mean in 2028, there are going to be a lot more projects that are going to start using ReRAM. But the day until they get and they grow this from 3% to 10%, it's going to be another 2 years or whatever. So that's what you need to understand in these markets. It's not that they are going to be only 3% of the projects designing with ReRAM in 2028. It's already in mass production. And so those are projects that before 2026 started to work on it.

Okay. Perfect. Yes. So about 24 months is the...

Yes.

And then the other question I've got because I think there's maybe some confusion around this is we're looking at this slide particularly. So when you talk about the embedded market, there's ReRAM being, say, 37% of $2.7 billion, so I should say, $1 billion to run it off. Now that's the sale of chips, right? So say we're on got 100% that market just in. The ongoing revenue for ReRAM would be the royalty -- or so the royalty payment percentage of that $1 billion?

So okay, in the business model, there are 3 main elements. When you're talking about the fabless companies, the product companies. There's license fee, there's the NRE if they want us to do any modification to the module and then there's the royalty. Now by the way, this is for every project of every customer, okay? It's not like a customer engages with us, they pay onetime license fee, that's it. And for eternity, that's it. Every time they start a new project, they pay us a license fee for that project, if they want to do some modifications, we do engineering work, and they pay us for the royalty from the sales of that budget, okay? So what I'm going to say on this. So yes, so those are the 3 payments. And as more and more and more projects start using our technology, there will be more and more license fees and royalties and most likely also NRE. One of the things is as we do more and more versions of the module for different people, we started having a large variety. So we need to do less engineering work to generate new versions because we already have so many versions out there. So over time, we'll start having a lot of different versions. For the big guys, we're also tailoring things for them.

[indiscernible].

Yes. So with the -- so in this business model, we work directly with the product companies and the royalty is directing from their revenue. In this business model, the foundries sell wafers to the customer at the cost of a wafer to take into account all of the IPs that the customer is using and then they give us a percentage of the wafer cost -- of the wafer sales that they do to that customer.

Back to the time line, last time when you mentioned about to speed up the process we're going to make chips to the customer specification so that we don't qualify what this hits has done so far? So is GlobalFoundries for at the moment, printing wafer at 22-nanometer scale, to the customer specification? Or are we doing the same like what you've done with SkyWater to print out a system on a chip and then qualify, how is GlobalFoundries doing now? Are they contain to customer specifications, so we qualify them and scale up the process?

No, we're doing the same thing that we did with SkyWater. We did our own demo chips, that's what we taped out in the beginning of the year, and we're waiting for those wafers to come out so that we can test them, and then we'll be starting to work on customer specifications.

Yes. Just a question about the -- so you've got on the team another top fabs. Is there going to be some level of exclusivity in there? Or will you license to multiple?

No exclusivity. No exclusivity at all.

So just a quick question, a couple of last ones. So setting up of a fab. So obviously, you need a critical mass, so how many units would you need for any one of these guys to consider tolling up? And then what is the tool up? Is it like a very complex tool up or they're basically just tweaking something they've felt already?

So that's really, when you look at the Weebit Board and the level of experience that we have, the key guidelines that I have from day 1 was you need to have the best ramp possible which is commercially viable. And that end is critical. And that's what some of our competitors did think about when they were doing their stuff, and we've been very focused. So in order to make our ReRAM commercial viable, we have really focused and limited ourselves to work only with the standard materials that the fabs work with today -- with the standard tools that have the fab work today. So that we will remove all of those hurdles. So theoretically, if a fab is not at full capacity, we can go in and use their existing production line and existing processes and just moving. So now the challenge starts when they -- like a machine is at full capacity right now and they need to bring in another machine just because it's at full capacity or things like that. But in general, that is one of the things that we've been really focused on, is removing all of these barriers to entry to a fab and doing everything we can so it's as clear as possible for them to have us come in.

And then...

So each -- yes, I didn't answer that point. Each one of these foundries and IDMs works differently, they have different approaches and stuff. So each one of them has a different criteria of what they consider the minimal critical mass in order to justify bringing in a new technology. You're right. I mean they won't just bring in a new technology just because it's fun, right? They always want to see that they see a strong enough market. I think all of these guys can see today -- even if they don't have specific customer demand for specific products and whatever they can see that the markets want ReRAM, okay? The Infineon announcement was really a major milestone for the semiconductor market in general, not just automotive but in general. Today, many of the big sales companies are already requesting ReRAM. We've already talked to foundries that told us that they lost one of their key customers because the customer asked for ReRAM and they then have ReRAM get to them. The situation is already there. The people are seeing that they're going to be losing business if they don't have ReRAM. And that's why, again, I mentioned earlier, some of the big times, they called us even before I managed to call them, they called us already and said we want to talk. So they know. They're seeing what's happening in the market.

We'll have some time for one more question from the room here. You're probably another 10 minutes, sorry for Coby, hang around here for people to ask any questions. I did see one question come online. While we've got this slide up just asking around the discrete market opportunity and what if you can define what short term means in terms of how long that would be, and pose to the 200-plus people online that we couldn't get to the questions.

So again, you guys are already getting the feel for what short term in semiconductor means. Short term is a year or 2 kind of looking forward. We'll be working on NOR Flash right now. We already have customers that have come to us with specific requests saying, "We really would like to have the existing technology that you have for embedded, can you make a chip that will be using that technology to replace a NOR Flash at that we have in our system? So we are already working with them and engaging with them on seeing how we can address that need.

And so and having to work on that I think also if we can just quickly mention that there's a perfect storm at the moment in this industry -- in the semiconductor industry. You've got a massive world shortage of chips. And that is a big compiled with that there's a huge log jam between how do you produce those chips in the amount of fabs that there are in the world when they don't have the capacity, coupled with that you've still got Flash memory. They can't just say, okay, let's just take ReRAM or let's just use ReRAM or they still need to be able to run everything from the life in this room, memorizing what the dimming setting years to your phone. Remember when you turn it on to the TV station -- when you turn on your TV and it remembers what you last watched. Flash still has to be there for the time being, but it's not able to keep up with the requirements. So there's a huge problem in the market. And that is what Weebit, what we're doing at company level is to make sure that once we've come up with the right solution, we're not going to hit other problems that prevent us from going into the market. For example, as Coby mentioned, that our, we're using materials that are ESG friendly that are, that we're not digging out things from the earth, MRAM doesn't do that. MRAM users rare earth. We are making sure that when we can go into a fab that we can go straight in and use the machinery and the fixed assets and the plants and equipment that they've been using for years and years. So it is -- we're trying to make sure that once we get our technology ready and qualified, that we can actually get it into the fabs, and we can get it out to the market, and the market will be really happy to -- they won't have barring to entry to use our technology within their products.

Just a quick question again on discrete, but the discrete not using the selector. You've mentioned that people have come to you requesting it. Is that where you think that first revenue will come from? Is it a big focus for you? Or is the focus still on embedded and getting agreements with Tier 1s?

We are working now in parallel. On the one hand, we're trying to get a foundry on board and then with customers -- and for me right now, I want to the first agreement. I want to see a first customer come on board. So we're pushing forward with some of these guys that want to have a discrete chip. We're pushing forward with some of the people that want to do System-on-a-Chip was embedded. The first one that we'll launch is the one that we took to focus on and really push forward. So as I was talking about this natural hesitance of using a new technology, and it's all of us -- when someone will come with something brand new that was never used before, we'll always have that reluctance to jump in there, right? So the first one that jumps in, I don't have a specific preference right now. The first one that jumps in, I'm going to latch on to that and push them as much as I can. I'm going to give them the best support possible. And obviously, we'll try to get that revenue from them. So I don't know which one of the customers we're talking to today will be the first revenue, but one of them.

[indiscernible].

It was -- you're talking about the old this one, yes.

Yes. Can I just ask this question so ReRAM [indiscernible] is [indiscernible]?

Yes. Now by the way, so what you're seeing here, there's -- on this side, there's revenues and this side, there's a number of units, okay? In terms of revenue, because MRAM is so much more expensive to manufacturer, even with smaller number of units, they'll actually have more revenue. I think that's what this thing basically shows.

[indiscernible] question is once the ReRAM [indiscernible].

Yes.

How much of that is going to [indiscernible]?

Of course, as much as we can, that's the best answer I can give. If you look at that, TSMC obviously, will have a big chunk. We shouldn't underestimate. TSMC is, by far, the largest foundry in the world. They have huge capacity -- they manufacture. Just TSMC manufacturers, more than half of all of the semiconductors in the world, okay? So just by that, we can say that at least half of this will probably be TSMC. But when you're looking at the other foundries, that's where we come in, and we are a key player. So there are, there's TSMC, UMC announced that they have a ReRAM. We're getting indications that there are issuance with that ReRAM, but I'm always paranoid and I'm going to assume that they're going to manufacture also quite a bit. So -- but we'll still have a nice chunk. It's really kind of just a wild guess right now at this point. So it's -- even if we have a quarter of that number, I think it will be very nice. Hopefully, we'll have more than that.

Eric sort of giving us the windup because Coby has got a head off to other meetings. But I just wanted to say something to the people here who have come into the city at [indiscernible] to get into CBDs these days with the traffic, but also to the, I think it's 200 people who are online at the moment is. What's an amazing occurrence is that the semiconductor industry in Australia has been nothing much. We haven't seen anything of it. In fact, we are meat, wool, wheat, coal, oil and gas. This is what we've been as a resources-based country. But you guys, the retail business into Weebit, a lot of you have been with us since day 1. And we would want to say thank you, and how exciting it is that the anomaly that in Australia, our retail investors, our moms and dads who bought into this new world-changing type technology have a better understanding of the semiconductor industry than our institutions, which is quite an anomaly. I mean when...

[indiscernible].

Absolutely. I mean when we go to -- when we come to these types of presentations, you guys asked the best questions. We go to institutional meetings, and they say, "Can you please educate us a little bit in what is actually the semiconductor industry?" So it's an achievement. And I think that is what is so great about Weebit listed in Australia is that it is, we're at the forefront of world-changing technology that our initial investors, our retail investors are more knowledgeable than most people in the world on this. So well done to you all, and thank you very much for attending. Thank you, Coby, from all the way, Coby. When we brought him on board as our CEO. And it was -- this is a very much a phased strategy with Weebit. We were very, very small in the beginning. And it was very hard to run a very small company that no one knew about and no one understood the industry, and we had no real money. And suddenly, we were listed. I've always thought that as the founding CEOs of a technology business, have to understand right from the start that they're not going to be the CEOs when you get to the commercialization phase. And that's our initial CEO, understood that. We made sure they did. And when we bought Coby on it was to take us through to that absolute next level and beyond, which is where we are today. So I feel like we're this gangly teenager that has now entered the ASX 300, so we've gone out of the teenage years to we're in our 20s now. For that reason, being in the ASX 300, we've made sure that as a corporate lead now constituted various committees on Board level. We've now got -- we've put together a Strategy and Technology Committee, Generation Committee, Governance Committee, Audit Risk and Finance Committees. These are what you need to do when you're an ASX-listed company. We've made sure that we are with the right consultants on Board, from IR, PR, legal, auditing, accounting, making sure that everything is run absolutely the way a corporate should be. And that is our role as well on a Board level to make sure that it's not just about the technology. It's also about that you're running the business correctly. So thank you for attending today. And any questions, feel free to ask Coby,but you can always -- I'm here in Australia. If any, do you feel like you want to call up and ask a question or have a chat, I'm always around any time.

Thanks guys. Thanks for listening in. Thanks.