This conference call transcript was computer generated and almost certianly contains errors. This transcript is provided for information purposes only.EarningsCall, LLC makes no representation about the accuracy of the aforementioned transcript, and you are cautioned not to place undue reliance on the information provided by the transcript.

Napatech
11/4/2024
Good morning, everyone, and welcome to today's Nappatech Q3 IMS. My name is Seb, and I'll be the operator for your call today. If you've dialed into the conference call and wish to ask a question, please press star one on your telephone keypad. I will now hand the floor over to Lars Boilersen, CEO, to begin.
Good morning. I'm Lars Boilersen, CEO of Nappatech. I'm pleased to welcome you to all to Nappatech's presentation for the first half of 2024. Joining me today is our Chief Financial Officer, Heine Torsgaard. Our third quarter 2024 report was released earlier this morning on the Oslo Stock Exchange and is also available on the Navitek website. For your information, a recording of this webcast will be available later today. There will be a question and answer session following the presentation. During and after these prepared remarks, you may submit your questions via text on the webcast paste, or we can take your questions on the phone. If you would like to ask a question, please follow the instructions on this slide. Please note that this presentation contains forward-looking statements that are subject to risk and uncertainties. Our actual results may differ from those discussed in forward-looking statements. Today's agenda will cover four main areas. First, we will provide a status for the third quarter of 2024 with focus on the major news items that show promising signs of progress by Navitek towards achieving our goals to transform our business. Next, we will provide a business outlook that shares information about the opportunity ahead of Navitek. We will then provide information on the third quarter financial results. Finally, we will then conclude with a question and answer session open to the attendees of today's call. We have made considerable strides in meeting important milestones of our strategic plan that strengthen our position as an early leader in the new and evolving market for programmable NICs. This includes both improving our core business and making measurable progress toward our design win goals that underpins our growth aspirations. This has gone according to plan and in some areas ahead of our expectations. Our priority for 2024 has been to execute our strategic plan as it underpins our long-term growth aspirations. To achieve those goals, we have three areas of focus. We must service our valuable installed base of customers who are the early adopters of programmable NIC solutions. They are proving ground for the technology and the value that our solution brings to their networks and businesses. We have accelerated our engineering in software and hardware technologies that extend our products into new areas that align with the highest projected growth segments for Navitek solutions. And we continue to develop an ecosystem through partnerships that expand our go-to-market reach via marketing, business development, and sales, enabling us to win new designs that would be challenging to achieve on our own. Before we get into the details, I'm pleased to report a summary of high-level progress in each of these areas. I would like to share a few highlights from the third quarter of 2024. We'll provide a summary here along with additional details in the coming slides. We have made notable strides to our design win goals that are the launch pads to new volumes of demand for our programmable NIC solutions. While design wins are long and complex, we measure important mileposts that indicate progress towards the end of goal of moving opportunities from programs and projects to production solutions. In the first quarter of 2024, we announced both a partnership with Intel Altera for the next generation infrastructure processing unit, as well as a design win with a Lighthouse customer who helped to define key aspect of the product. In Q3, our engineering completed a significant portion of the design on schedule and shipped these first engineering samples to the customer on time for the initial bring up and qualification. This is a very early step in a long engagement that will continue through 2025, but it is an important milepost that we met these challenging requirements and high expectations from the clients. We continue to expand our design wind pipeline with new opportunities and projects emerging on a consistent basis. This is expected, giving industry momentum for programmable NICs in the next generation data center architectures. The opportunities we are pursuing span a variety of applications, use cases, geographics, and scale. While each are unique in some ways, including requirements, volume, and pipelines, we are happy that all the design winds we have engaged in all continuing to move forward. We are particularly pleased to see that Intel Algeria continues on a monthly basis to provide new high volume design wind opportunities to NABATECH. With the design wind pipeline, we have created opportunities from both new customers and existing customers driven by our new solutions and our core products. Further, we now see the signs of stability emerging in our base business. Our largest OEM customers are nearing completion of their inventory reductions to be completed by the end of 2024. Many of our existing end-user customers are returning to historical demand with new projects and new programs. And our existing customers are refreshing products with new design wins for Navitek. Also, in the quarter, we extended our product and technology collaboration with Intel Altera. We announced two new products, including hardware and software, aims squarely at the high growth segment that define our relationship. Finally, we continue to see validation from prospect, customers, partners, and industry analysts that increase the market projection for Nabatech's products and solutions. This provides all the inspiration and motivation we need to press forward with our plans. Our third quarter results include revenue of $5 million and increase of 47% over the second quarter of 2024, with gross margin at a healthy 71%. As previously communicated, our 2024 year to date have been impacted by our largest customer, who traditionally have accounted for up to 50% of our quarterly revenue. If our biggest client have had a normal quarter, then this Q3 quarter would have been a strong quarter with revenue and new wins. We will provide additional financial details later in the presentation. The majority of our efforts have remained on areas where we have much greater control and ability to influence, which is the pipeline building and design win focus for our new business expansion. We remain excited and are pleased with those results. Since our last reporting, we have made progress on many fronts and have increasing confidence in achieving our goals for transformation volumes when these prospects reach production volumes. In the last quarter, we made progress in each opportunity presented to us In each opportunity, we have created a deeper level of engagement and more traction. While design wins are long and complicated, we can see the signs of progress in each of them. I'd like to share just a few highlights on the progress made in Q3, representing a small sample of a much larger set of opportunities before us. Note that this is just a sample of a larger pipeline. My goal today is to show the attractive mix of customer types, use cases, application, graphics and technology that provides insight to how power and flexible programmable NICs are as a platform for modern data center server designs. First, we announced the tier one global server manufacturer with proven success in cloud and enterprise designs and a strategic focus on artificial intelligence. Their product is packed with the latest technology from Intel Altera and is designed with optimization to meet the exact requirements of the server manufacturer. The cooperation is anchored by commercial agreements, associated statements of work that define the product specification and professional service fees from both companies to NABATECH. The collaboration includes these significant contribution and represents one of the largest opportunities in our pipeline. I mentioned in my opening summary, we achieved an important milepost in the third quarter, having shipped the first engineering samples to them. As we have progressed with this design, we have expanded to multiple groups within this tier one company who also may utilize the same hardware products and ultimately drive incremental volumes to NABATECH. Another example is a high-growth and innovative provider of high-performance and cost-effective data management solution for an exploding list of data center and its computing AI ML workloads. This is our previously disclosed design win, SigmaX.ai. They continue to develop their solution, powered by Nabatech and Intel Altera. We are several quarters into engagement with the global Fortune 100 banking and finance company, who is re-architecting their network to deploy security services into every server based on programmable NICs. Their existing solution is based on servers with standard NICs that are deployed by network firewalls. As the servers evolve with greater levels of virtualization, including virtual machines and containers, they require new levels of protection within and among the endpoint services. Customers like them are choosing programmable NICs for this task to deliver distributed host-based security and to secure that they maintain the performance and latency in their business services. We are several quarters into another opportunity with a U.S.-based supplier of networking equipment to U.S. government defense, civilian and intelligence networks focused on cybersecurity and network monitoring solutions. We have made considerable progress in engagement with a new design by a global tier one video streaming service provider. The solution for their content deliver network utilize programmable NICs to improve their performance and secure the content in their services. The engagement has evolved to expose multiple use cases beyond the original design that drive additional demand for Navitek. We have also made progress along with one of our 5G UPF offload and acceleration software partners on a design by a 5G packet core provider for a 5G mobile telecom network operator. We began a new engagement with a highly respected and innovative provider of artificial intelligence and machine learning solutions who need to scale their server-based product with a high performance and programmable NICs. As a final example, we have active engagement with many of the leading providers of 5G mobile packet core solution who now offer a Nabatech SmartNIC software and hardware option for their UPF applications. This option provides compelling ROI benefits, dramatically reducing the cost, size, and power requirements to deliver 5G services and increasing number of subscribers. We are continually inspired by our business development efforts in partnership with Intel, who is leading the way on many of these opportunities and regularly bringing new targets to NABATECH. While there is much work to do to secure new design wins that drive large volumes in their peak production years, these active opportunity examples give us confidence that we are making solid strides. In success, we aim to transform our business from a niche supplier to a high volume partner to customers in the largest and highest growth segments. Our historical volumes in our valuable but niche segment ranges from 5,000 to 6,000 units per year. The new type of design winds we are pursuing will result in annual unit demand of 50 to 60,000 units per year when they reach peak production. And what lies beyond that are a large and expanding pipeline of future design winds that drive even higher volumes. The use case of fueling the growth of programmable NICs has included cloud and edge computing, 5G mobile infrastructure, cybersecurity, and others. AI infrastructure is an emergent use case that is further increasing the amount for smart NICs and IPUs. Intel Altera is the leader provider of smart NIC and IPU solution to many of the top hyperscaler network operators. To grow and expand beyond the smaller number of large hyperscaler operators to the mass market of next wave cloud, telecom and enterprise networks, Intel Altera has partnered with NABATECH to provide complete commercial production grade solution to the mass market beyond these early adopters. So far in 2024, we have secured valuable new customers for solutions in fintech, cybersecurity, artificial intelligence, monitoring and recording, big data analytics, and more. Each customer and end-use case for a new design can range from 1,000 units per year to 5,000, 10,000, or even 5,000 units per year. Every new design from small to large gives us confidence in achieving our longer-term growth aspirations. While these examples are great signs of progress that give us confidence towards meeting our goals, we constantly remind ourselves that all design wins take time, and not all will be won. They require engineering by our customers after selection, but their size warrants the effort to develop them, and our customer will then take time to launch and ramp their products to full peak production. But their value warrants our patience. Our current active design wind pipeline includes more than 200 opportunities that could drive more than $350 million in revenue. While we cannot expect to win every one, we are aggressively pursuing each of them. NABATECH's programmable NIC play a crucial role in modern data center and server design. They provide the connectivity between the outside cloud and enterprise networks and the critical resources inside the server that deliver artificial intelligence, security, and other services. Navitalk products are labeled A and B in this image. Historically, the most important part of the server was the central processing unit or CPUs. The CPUs from companies like Intel act as the brains of the server and are labeled E in this image. The CPU are surrounded by memory labeled D. Memory stores data, including network traffic and application information for the CPUs to the process. A well-publicized area inside of servers are the expansion slots, shown as C. These expansion slots are most often used today for acceleration cars based on GPUs or FPGAs for artificial intelligence, machine learning, training, large language models, and many other workloads. These coprocessors are used to assist the CPUs in handling the most complex workloads. Numerous innovative companies are merging with new solutions to offload and accelerate CPUs from burdensome workloads, including artificial intelligence, machine learning, inference and training, video encoding and decoding, encryption and compression, and more. And many of these companies have implemented their solution on FPGAs, giving their high performance and programmability. These workload needs a home to run within the server and Navitek FPGA based products are ideally suited for this task. While it is in early phases, this is the example of another set of new cases that underpin the demand for FPGA solution from Navitek and are entering our design wind pipeline. There is no better way to illustrate the opportunity being created by these companies than to provide a concrete example. I'm happy to share with you more information about an exciting new design win and a partnership with a high growth company, Myrtle.ai. The increasing complexity in machine learning algorithms driven by an explosion in the use of machine learning to introduce artificial intelligence into all walks of life is massively increasing demand for significant improvement in latency, throughput, energy, and cost. The growth in machine learning is not sustainable using traditional software-only solutions and constrained by many general-purpose CPUs and AI inference and coprocessors. .ai produces machine learning inference solution which meet the exact demand of tomorrow's real-time inference workloads using minimum compute capacity and minimum energy. Specifically, they accelerate inference models on FPGAs to provide their customers with a latency advantage and cost benefit compared with alternative solution. Unlike DPUs and fixed function ASICs, their solution are powered by FPGAs, so their solution are programmable and can adapt to the latest developments. Example of benefits derived include faster, more intelligent decision making in finance, more natural sounding speeds through latency reduction of up to 20%, CapEx and OpEx savings up to 90%, superior image classification using poor quality images. Giving the explosive deployment of servers specifically for AI workloads, their solution has the potential to be deployed in high volumes in every data center of the future. More information on the Myrtle.ai solution powered by Nabatech will be available and on display December 3rd at a leading financial service data center networking show, STAC, in London. Although it took longer than we expected or desired, we encouraged as we begin to see signs of our stabilization in our core business. Existing customers continue to refresh the product designs, awarding new design wins to Navitek. In one recent example, CPacket, a leading provider network appliance used for networking monitoring and security, announced they had doubled the speed of their packet capture appliance. We have also seen awards for end-user customers in the government and enterprise for cybersecurity, monitoring, and financial service applications. On a particular note, we have seen a notable number of wins within the financial services and fintech markets, where FPGA-based programmable NICs are the preferred solution over competing alternatives. When it comes to programmable NICs for fintech use cases, an impressive list of leading financial services companies has chosen Nabatech. These type of winning accounts from our base business represent the early adopters of our product. The new solution we are developing allow us to expanding these accounts into the potential to capture a much larger percentage of their businesses. In 2023, we announced our strategic partnership with Intel Altera to develop a new line of programmable NICs, including smart NICs and IBUs, along with software solution aimed at the highest growth segments. We expanded that relationship in 2024. As those developments are nearing completion, we are fortunate enough to begin disclosing these new offerings to our shared customers. In the third quarter, we introduced the N3070X programmable SmartNIC. The new 400 gigabyte per second N3070X AI SmartNIC is based on the latest technology from Intel Altera. The SmartNIC is engineered for the challenge demands in artificial intelligence, infrastructure, as well as traditional SmartNIC use cases. Network operators evaluating early network architectures for AI have been challenged by vendor logging based on proprietary InfiniBand and RDMA over converged Ethernet implementations. NAPA tech solution conformed to a vision for standardized Ethernet infrastructure utilizing Ultra Ethernet. it has a creative design that allows it to be optional configured with an intel science cpu converting it to a high performance ipu the product was first promoted at the recent intel altera innovative conference held in october at their headquarters in silicon valley also in the third quarter nabatech released our link in line production grade software solution on or Intel-based NT400D13 programmable SmartNIC. This software enables 5G mobile network operators to offload and accelerate their user plane function. The new offering provides an improvement of more than 60% over previous solutions, helping operators to dramatically reduce the cost, size, and power requirements when delivering 5G services to an increasing number of subscribers. These product and technology announcement are important, but a small piece of the overall success we had in our relationship with Intel Altera. They continue to provide valuable assistance in our design win pipeline creation. I will conclude the business status portion of our presentation by noting that we are continuing to make strides in the hardware and software product development in 2024. We are now in a position to begin sharing information on products and solution with the world. We invite you to visit Nabatech at an upcoming event near you. Here are a few upcoming events where our new solution will be on display. Shuricon is a leading event attended by the top cybersecurity professionals and will be in Madrid, Spain in November. Stack is the top technology event for designers of networks and servers in the financial services markets and will be held in early December in London. And I'm pleased to represent NABATECH at the 27th Annual Needham Growth Conference in January in New York City. We hope to see you soon. Now we will move into the final section, and I will now give the word to our CFO, Heine Torskov. Please.
Thank you, Lars. Revenue in Q3 amounted to 34 million DKK and in USD 5 million. With 47% growth in revenue in Q3 compared to Q2, we saw signs of improvement in our sales, but the recovery in the smart net market, especially within the US finance and tech co-market segments, has been slower than anticipated at the beginning of the year. Compared to Q3 2023, revenue is still down 31%. We are expecting our traditional smart tech market to begin normalizing during the first half of 2025. For the first three quarters of the year, revenue in DKK amounted to 78.7 million and in USD 11.5 million. Gross margins in Q3 ended at 71.3%, well within our normal levels. The gross margins in the first three quarters of 2024 were 69.5%. Our staff costs and other external costs in Q3 amounted to 43.2 million compared to 32.1 million in Q3 last year. Year over year, our staff costs were up 27% in Q3 and our external costs were up 50%. This growth in cost follows our plan to accelerate our development activities significantly and bring new products to the market quickly to serve the increased design wind pipeline. EBITDAG in Q3 amounted to negative 18.9 million, which was an improvement of 9.3 million compared to Q2. EBITDAG for the first three quarters of 2024 amounted to a negative 72.3 million. Net cash flows from operating activities in Q3 amounted to negative 22.5 million, which is an improvement from 41.9 million in Q2 this year. Net cash used in investing activities in Q3 amounted to 3.5 million compared to 1.5 million in Q3 of 2023. Networking capital at the end of Q3 was 72.4 million up from Q2 as a result of our building an inventory for the sale of new products for immediate delivery. Cash and cash equivalents at the end of Q3 2024 amounted to 102.9 million compared to 44.2 million at the end of Q3 2023. Back to you Lars.
Our strategic targets remain unchanged. Our design win pipeline development is progressing well and our engagement with our lead prospect is maturing. However, the timing of securing design win is depending on the customer decision process and project planning cyclists. We still hope we can get a couple of new design wins before year end in order to get inside the guided range. On our financial targets, it has been very difficult to guide this year's revenue and results. We are clearly in a transition from our existing business towards new products based on our new Intel-Altera partnership. Very much our shortfall in revenues is related to our largest OEM in our existing business, who still is in a phase of reducing their inventories. We expect that to continue until the end of 2024. The good news is that most of our current customers were back to normal levels in Q3, and we expect that our current business should normalize during first half in 2025. We will not update our current guidance for the rest of the year, but we can determine that we will not be able to reach our guidance. We guide staff costs for 2024 in the range of DKK 107 to 180 million and gross margins in the range of 69 to 71%. This ends today's presentation and we will now open up for the Q&A session.
Thank you. If you would like to ask a question and you're dialed into the conference call, please press star one on your telephone keypad. If you're listening via the webcast, you can also submit a text question there. We do have a question on the conference call. This comes from Oystein Lodgaard at ABG. Oystein, please go ahead.
Thank you. Good morning Lars and Heine. First of all, I'd like to start with a question on the large design win that you announced earlier this year with the T1 server OEM. Can you say a bit about what are the remaining milestones that you need to go through with that design? And what do we think about when will these milestones be? So when should we kind of have increased confidence that production will begin to ramp up?
Yeah, thank you, Øystein. So as we said in presentation, the first thing you do there is we have to deliver the hardware. And we are delivering this in two deliverables, two carts. The first one we deliver on time in September. And the next one, which is a complete prototype, including software and everything, will be delivered on time end of this month. And then, of course, the client will basically go through a bring up qualification period, meaning that they will validate the hardware, they will test it, etc. And then the works next year starts to hopefully integrate this into specific servers in this company and then getting into production.
Thank you very much. And when should we start to see volumes begin to ramp up on that contract?
So that contract, it all depends on when this network card will be integrated into a complete server for this tier one. So basically what's happening here is that they are introducing a new smart NIC, which is our product. And then it has to be integrated into complete servers and then into production and start shipping. So we will hope that we get production in second half next year and high volumes in 26.
Okay, that's very clear. Thank you. And also, if you could give an update on what's the progress on Sigma X AI.
So Sigma X AI is one of these companies who is making let's say some very interesting software in this case they are basically helping clients to take away a lot of data before they send it to their uh to to to their to the data science company who analyze the data for them so they have they have some very interesting software which take away a lot of the data normally you pay fixed price for this data you send to you the company who analyze your data so so they can take away a lot of duplications a lot of data which is not really worth paying for and they can do that on the fly and in order to to do that they run their heart they run their software they need to run it on some on some hardware and in this case they have chosen to run this on um on our IPU 200 gig. So as soon as they win deals, we will basically deliver our standard product to them. But we are relying on them winning deals. We are in close contact with them. They have a very interesting pipeline. So we hope that we'll get some deals in the near future.
Okay, thank you. And lastly, I had some questions on slide seven where you look at these new potential design wins. I see there are some changes there compared to Q2. All these changes seem to be positive that you revised some of these estimates upwards. Can you just say something about why you've revised these estimates upwards?
I'm not sure what you mean by reverse. I think we are. What estimates are you referring to?
The U.S. government network equipment is revised upwards from one to three to now you say above three. And also, of course, the global video streaming services is revised upwards from one to three to over five.
So, first of all, we have more opportunities than the ones we described there. So we basically choose them to describe the variety of design wins we're working on and also to illustrate some of the interesting use cases. So that's how we select them. We do not really select them after the highest volume. But on the specific one, we have updated because it's simply just we see that the opportunity is bigger than we did last time. So that's the reason. In general, I would say that we have had really good progress on all our design wins, which we have been working on in the last six months. I think we have not lost anyone and we have made progress on all of them. So we definitely think that we are in a much better place now compared to six months ago when it comes to all the design wins we're working on.
Great. And last question for me, when these estimates that you give here, is that an indication of annual volumes that are kind of recurring or how should we look at that?
It's basically the volume in the peak production year, right? So that's how you should look at it. But obviously, most of these deals will be recurring, but it differs from opportunity to opportunity. If it's a, if it's a, if it, let's say it's a, it's a server manufacturer, right? They then basically most, they very often ship more servers every year, right? So then it will increase, it will be recurring and it will increase per year, but it differs from, from opportunity to opportunity. And that's why we say that right now we see like 40 to 60,000 units in peak production year.
Okay, that was all from me. Thank you very much.
Thank you. We have no other questions on the conference call, so I'll hand over to management for webcast questions.
Yeah, so we have a question here from Frederick Nikolais, and how much weight should we really put on design wins? It's a loose agreement on potential outtake that the customer can easily drop, or is it committed agreement? If so, how committed? So the way it works with these design wins is that that They basically design a product and we want to be designed in with our hardware and software, and then it has to go into production. So there is no commitment at this stage we are now. We are relying on that this product actually gets into production. And these discussion around commitment, they... that typically starts when you get designed into the server or et cetera. So it also differs from client to client. We have, for example, on the tier one biggest opportunity we have, there we are basically getting paid for the work we do because it's outside our classic roadmap. But hopefully we'll get into discussions on commitment in terms of volume at the stage where we get design into a real product. And it's not just about the network we manufacture. Next question. Lars Hansen, has there been any product shipment and revenue from AI products yet? How large is the revenue recognition yet? So we are in early phase. We are working really hard to get design wins, meaning being designed in. So for example, Moodle AI, SigmaX.ai are classic software AI products, and they run the software in our hardware. So they will have to basically win deals before we get revenue recognized there. And for more, the bigger opportunities we have where we design into server or bigger products there, the revenue recognition will start when they start shipping these products. So when do you expect revenue from the IPU Altera products? Is it possible in Q4 already? So right now, the only revenue we have there is when cost clients ask for sample products to see if they should design in our IPU. But if clients like Sigma X actually wins order, then then we will get revenues as soon as they will start shipping. So we do have some clients who are using our standard products here. And we also have, if you go through some of the design wins, some of them we hope to close before New Year's and they will need maybe a couple of hundred units right away. So we still are hopeful there might be a chance that we can get revenue this year. When we guided this year, we were very clear that we will not get any revenue this year, but we see that we have clients that might decide on our design and the product they're using is our standard IPU or our NT400 standard product. And that means that we can ship within one week. I think that was all the questions from the web. So if no further questions... Just confirming we have no further questions on the call as well. Okay. All right. No more questions or you had questions? Sorry.
Sorry. No further questions on the call.
No further questions. All right. Okay. So I think we end the presentation here. Thank you very much for participating.