Closing Bell - Manifest Space: Space as the Next Warfighting Frontier with Anduril Senior VP of Engineering Gokul Subramanian 11/29/24

Episode Date: November 29, 2024

Anduril Industries is a fast-growing defense tech startup best know for attack drones, robot submarines and unmanned fighter jets. The company is now forging forward into the next domain: space. Earlier in November, Anduril won a $100 million contract with the U.S Space Force to upgrade its military satellite sensor network. The company is also developing its own spacecraft, partnering with other venture-backed space startups like Apex and Impulse Space. On this episode, Senior Vice President of Engineering Gokul Subramanian joins Morgan Brennan to discuss the next frontier.

Andral Industries is a fast-growing defense tech startup founded by Oculus creator Palmer Luckey, and is best known perhaps for attack drones, robot submarines, and unmanned fighter jets. Valued at $14 billion last summer, the company has been forging further into another war-fighting domain, space. We see space as existential for the United States. You know, in the next 10 to 20 years, our adversaries are going to launch more spacecrafts into space than they have done in the last 70 combined. And so this is increasingly becoming a contested domain where we're going to have to have eyes and ears on everything that's going on. And that's really the area that we're seeing a lot of demand and why we're investing so heavily. Gokul Submarinian is Senior Vice President of Engineering at Anduril, focused on software and space. Just last week, the company won a $100 million program of record contract to upgrade the

U.S. Space Force's Space Surveillance Network to autonomously process and distribute data from a global collection of military space sensors. Anduril is moving fast. It already has payload on orbit, is developing its own spacecraft, and has been striking partnerships with commercial space startups like Apex and Impulse Space. We're seeing all of these small businesses, medium-sized businesses, who are working in the commercial space sector, who don't really have the ability to transact with the U.S. government and may not even understand the full set of mission needs that the U.S. government has. And so we can operate

as someone who can bring together and convene this set of companies. We're calling it the team of super friends who could come in and bring together an offering that is complete for the U.S. Space Force. On this episode, Submarinium breaks down the startup's burgeoning space portfolio, the strategy to build a, quote, super friends supply chain, and the plan to bring AI edge computing to, well, the edge of space. I'm Morgan Brennan, and this is Manifest Space. space. Joining me now, Gokul Submaranian, the Senior Vice President of Engineering at Anduril. It's great to speak with you, Gokul. Thanks for taking the time. Hi, Morgan. It's so great to see you and great to speak with you. So when I think of Anduril, I think of drones. I think of AI. I

think of robot submarines. I think about unmanned autonomous planes. I don't think about space and yet you're building out quite the space portfolio. Tell me about it. Yeah, thanks. Let me just take a step back for a minute. And yeah, I mean, when you think of Anduril, you think of all of those things. And really the through line that connects all of those things is Anduril has had a real focus on moving the paradigm, the traditional paradigm has been, the DoD has fielded for the last 40 years has been small do we move that? How do we invert that to large numbers of low cost systems? And you've seen us do this with the Century Tower.

You've seen us do this with the submarine that we're building, the school bus size submarine that we're building down in Australia. You're seeing us do this with the airplane that we're building. And in order to do that, two things have to be true. One is you have to have a real focus on software and autonomy, because if you're moving to high numbers of systems, then those things have to be able to come in and control themselves. You can't have one operator control every one of these things. And you have to have a real focus on production and the ability to build at scale. And that's the through line that connects every domain that we're active in. And that's that same through line that we're now applying to space.

And, you know, we see space as existential for the United States. You know, in the next 10 to 20 years, our adversaries are going to launch more spacecraft into space than they have done in the last 70 combined. And so this is increasingly becoming a contested domain where we're going to have to have eyes and ears on everything that's going on and that's really the area that we're seeing a lot of demand and why we're investing so heavily so when we talk about some of those investments what what do those look like what are the areas of expertise that are being crafted at andro when it comes to the space domain? Yeah. When we started in this, and I really took this on about a year ago,

our CEO, Brian, pulled me in and said, look, I really feel that for us to be able to be relevant, for even our systems to be able to be relevant terrestrially, we have to have a space capability and we have to have a space layer. And so we really asked ourselves then, what is the contribution that Anduril can make in space? Where can we be value add? Where can we do something new, unique, and different that the market isn't already getting from the traditional providers? And for us, that's a real focus on what we're calling space domain awareness. So how do we have eyes and ears on everything that's happening in space? Again, as that domain gets more congested and contested, there's going to be way more

things up there. And our traditional ways of maintaining custody are just going to start breaking down. And then from there, how do we enable more and more autonomous operations to occur in space? How do we enable systems to detect things, look, make decisions, move about those sorts of operations. And that's really been our calling card in every other domain. I call out a few things. I mean, we see this happening not just in space, but from the ground as well. We have a program that we're really proud of. We're working with the Space Force on called SDA Net or Space Domain Awareness

Net. And it is a network of sensors on the ground that are looking up to space and being able to identify objects. And we're integrating with sensors that have been on the ground for the last 40 years. These are legacy systems, but we're upgrading them and making them more autonomous, more resilient. And we're super thrilled about that work. And I'd love to share a story with you about that if and when we have time. In addition to that, we're also launching spacecraft. We have our first payload on orbit already. And we're launching a fully owned Andoril spacecraft next year that will have the ability to monitor objects in space. And I can tell you all sorts of things about that. Okay, well, there's two follow up questions right there. First,

I want to hear the story naturally. So maybe we start there. Yeah, absolutely. So SDA-Net is a program where we're integrating with the Space Surveillance Network. These are systems that have been online for about 40 years now. They're radar sites, they're optical telescopes. These are huge systems. So imagine there's one in Maui that's at the top of a mountain that's absolutely massive. It's the biggest telescope you'll see. And it's charged with maintaining custody of any spacecraft that comes over its field of view. Now, these systems, they're feats of engineering, and they were put into place in the 60s and 70s, but they're connected via what is basically dial-up internet, if that, even worse than dial-up internet. Now, your viewers may remember about two years ago, there was a huge fire in the area. There was a huge storm in the area and that there was a huge fire. And the result of that was that

connectivity to that site was totally lost. Our technology was able to come in and restore connectivity to that site in a matter of minutes. And the way we do that is we leverage two things. One is commercial capability. So in this case, Starlink. So instead of using landline cables that can easily be cut, we're moving data over Starlink and any other modality that's available. For example, LTE, the 5G connectivity that your cell phones use, or Starlink in space. And then the other thing we do is we move from a hub and spoke networking model to a mesh networking model. That's a fancy term for saying that instead of making these things have only one place they can send their data, we give them a resiliency where

they're able to send data not just to the central node, but to any other node in the system. And the data moves freely and ends up where it needs to go. And so with those two things, we're able to restore connectivity to that Maui site in a matter of hours and get that site back online when the fire basically took out the traditional way of networking. So we're in the process of rolling out that capability to dozens of sites all across the world. I mean, we're going to places like Ascension Island, where it's very, very hard to get to. And these sites perform a critical function of monitoring space from Earth and increasing the resiliency and the automation available at those sites. So to get connectivity on the island up and running again in a matter of a couple of hours, I mean,

how I guess how ready was that suite of offerings to be deployed to that situation? How much of that was you sort of just winging it right then and there and using what was available to you? I guess just how to think about how quickly that came together. Yeah, yeah. One of the really cool things about Anduril is we're a product company, right? And so we're building a suite of products that we use in every domain. And in this case, we're actually leveraging technology that we had built for many other customers. And the really great thing about that is that means that the product gets better and better over time, not just when one of the customers is paying the bills.

We're basically amortizing that cost over many customers. So we had a product offering ready to go, battle tested and battle hardened. In this case, we had actually done tests at the Maui site. We had deployed that hardware out to the Maui site and started doing tests to make sure it worked. When the fire happened, it basically sped up the rate at which we moved into operations. And we were able to very quickly move it to becoming the primary comms channel instead of becoming a redundant backup. And of course, I would imagine it gave the US military and other customers sort of that real life, real use case, here it is deployed in the field, look how well it's working, make the case for future investments and future purchases.

That's exactly right. Yeah, you know, we were happy to be called to the line of duty. And, you know, that was a great, you know, there's nothing like kind of learning in the heat of battle, so to speak. So we were able to learn from that. We were able to make improvements from that. The system worked. And now, you know, we recently announced that that program, the SDA-NAP program is expanding. We've won what's called a program of record in defense speak. So it's going to be a program of record that's going to deploy out to all of the sites and get them increasingly operational over the next three years. And now is that the $100 million contract that was just announced recently? That's exactly right. Program of record

is a big deal, especially for defense tech startups. I mean, I feel like I've had conversations for many years about how hard it is to get to the point of program of record. So how significant of a milestone is that, especially for a part of the broader portfolio that's relatively new? Yeah, that's right. And actually, I want to tell the story of how that whole contract came to be, because I think it is a huge success, not just for us, but for the industry as a whole, as you mentioned, the defense tech industry. And that's part of the thesis that we have for our broader space business. So this program, SDA Net, actually started as an SBIR, Small Business Innovations Research, that we won back in 2019.

You know, we went to the original pitch day that Space Force was hosting for small businesses to pitch ideas. And we came and said, hey, we've got this technology, got this mesh networking technology that we're using all over the southern border, we're using it for counter UAS purposes, we think there's a real application. And it started as I believe, a $300,000 small business research contract. And the thing about these SBIRs is very, very few of them move into operations. And I could say even fewer, maybe one to two handfuls end up ever as becoming a program of record. And it's been a real focus. You see this talked about

in Congress and in other places that the SBIR program isn't getting these small businesses over the hump into part of the defense industrial base. And so I think this program is a real precedent setter. It's really good for the industry. It's really good for all the other companies coming in behind us to say, look, here's an example of not just kind of spinning in SBIR land, but taking a concept in SBIR and moving it into production, into program of record territory and hats off to the Space Force, really. I mean, none of this happens without a willing partner who's leaning forward, investing, and seeing the value that can be generated and leaning into it. Yeah, it does feel like the Space Force has been particularly

forward leaning and kind of future thinking in terms of its ability to engage with and actually draw business from commercial companies and dual use technology companies. It does feel like it's kind of spearheading that process, especially at a time where as the Trump administration gets ready to come to the White House, there's more focus on this type of contracting and rebuilding the military in a smarter way that shifts more of the risk onto companies and off taxpayers. Yeah, I'm so glad you mentioned that. And we see and feel the same thing.

There's so much energy right now in commercial space. It's really kind of the next space age is what I hear from a lot of people. There's what Elon's doing with SpaceX, who's obviously the front runner in this, but there's so many companies coming in behind them. We're down in Southern California and we feel that energy every day. One of our theses with our space business is how do we harness that energy and bring new players and more optionality to the US government? And I think, as you said, the Space Force is leaning in. They are a new service and they're looking at how they can do things differently. Now,

we're seeing all of these small businesses, medium-sized businesses who are working in the commercial space sector who don't really have the ability to transact with the U.S. government and may not even understand the full set of mission needs that the U.S. government has. And so we can operate as someone who can bring together and convene this set of companies. We're calling it the team of super friends who could come in and bring together an offering that is complete for the U.S. Space Force. So we could say, hey, you build the best buses on the planet. You're a new commercial offering. You're doing this for commercial buyers. Let's figure out how to leverage that bus for a defense purpose. And you build a great radar. You build a great camera. We at

Andral, we build great autonomy. We have sensors and payloads of our own. Let's bring all of that together. And Andral, I think we've proven time and time again that we deeply understand our customers' problems and we know how to move at those problems at a ridiculously fast rate. And that's what makes us so successful in all these other domains. So I think a big part of what you're going to see from us in space is convening this team of super friends across the industry who want to contribute to national security, but maybe their company isn't set up or they don't have the domain expertise. We think we can bring all of that together.

Super friends. I'm going to take this term. I'm going to run with it. You say super friends. I'm going to say commercially rooted, new type of industrial supply chain or defense industrial base. Maybe that's right. Maybe that's wrong. But that's what I'm hearing from you is it sounds like a new way of not only contracting the government and bringing technology to the government, but also building out, you know, resilient supply chains. That's exactly right. So I'll point to two examples. So we've recently announced partnerships with Impulse Space and Apex Space. They're both commercial

spacecraft providers. They build buses, right? The core, when you think of a spacecraft, you think of these little blocks with two, you blocks with two solar panels sticking out the side of it. They build those things. Now, a spacecraft by itself is not sufficient to accomplish a government mission, but it is one of the core building blocks. Now, what these guys have is they have world-class technology. They're hungry. They're small startups. And they're moving way ahead of where the traditional technology has

existed. And they're building a commercial supply chain. They're building a commercial factory with a commercial line. And the analogy that I'd like your viewers to think about is, if you think about Ford, Ford builds commercial cars all day long. They build trucks, they build sedans, you know, they build the Crown Victoria. If you grew up in the 90s like me, they build the little Crown Victoria cruiser. Now, you could take that Crown Victoria and you can use it as a commercial vehicle or you could turn it into a police cruiser, which is what we probably all think of when you think of the Crown Victoria. Right. You take that commercial car, you change out the suspension, you change out the shocks, you put in a bigger engine, you put the police siren on the top of it. And all of a sudden you've taken a commercial vehicle and you've turned it into a national security system. And I think

that's the analogy that I would like to think about for what Andral can do. So we can help these commercial suppliers build for their commercial needs, but we can take a handful of those off the line and customize them for the dispense specific use cases, which they might not even know about. They don't have the infrastructure. They don't have the security clearances to know about, but we can leverage their technology. And what that gets the US government is lower price because these are commercial supply chains that we're tapping into that are being built on mass, right? They're building hundreds of these instead of paying an arm and a leg to have this one very unique custom spacecraft built just for your mission let's leverage these commercial

supply chains and then figure out how to repurpose them like the crown victoria into the police cruiser for these defense missions that's a good metaphor um so of course apex and impulse both very uh sort of buzzed about startups bc back startups how do you decide who to partner with? Part of it is just we look for companies that resonate with us, our culture-wise. Both of those companies, we feel like we could walk into their facilities, and it's just like being at Antical. And so I'll give you the example with Apex. When I first met Ian Cinnamon, the CEO of Apex, I said, hey, we're starting up a space business for all these reasons. Ian's like, couldn't agree more. We see it the exact same way. And I said, I want to do a partnership with you. Let's figure out how to launch one of our payloads on your spacecraft. And from the moment we had that conversation, we had a payload integrated on their spacecraft two

months later. From zero to 60, we went that fast. And we had that thing launched three months from that day we had that conversation, right? So, I mean, we could move at the speed of startup. And so that's one thing that we look for in our partners. The other thing is complementary capability. Right. We're experts at defense. We're experts at government missions. We're experts at building payloads for those missions. And I'd love to talk to you about some of the payloads that we're building. But we're not a bus provider. We don't build communications. We don't build all sorts of different things, radios. We don't build those things. So we look for partners who augment the things that we're not as good at, like I said, the team of super friends. So what are the type of payloads that you are focused on building? Yeah, a payload that

we're really eager to announce is we're building an infrared technology. So we have this on the ground. We call it the WISP. And that's part of one of the acquisitions we did at Andral Imaging. So we have a capability that was designed for the ground use case that uses infrared to sense the environment. Now we're going to take that same technology. We're already applying it to the air domain. That's an infrared search and track capability that we're going to take that same technology. We're already applying it to the air domain. That's an infrared search and track capability that we're putting on our airplane. We're going to take that same thing and we're going to move it to the space domain. And what that does is, one, we're leveraging existing technologies.

The U.S. government doesn't have to pay for the R&D from scratch. But two, it gives them a capability at a price point that we think is going to be really, really unique. So why does this infrared technology matter? It's because when we're looking for other country spacecraft in space, it's very hard to see them when they're on the dark side of the earth. Right. So when you're on the side of the earth where the sun isn't shining, it's not easy to locate where the bad guys could be. And it often is the case that bad guys will potentially make maneuvers during that period. They'll move around. And so our infrared technology gives US spacecraft

the ability to see in the dark, if you will. So we're going to be able to see the heat reflecting off of those spacecraft, and we're going to sense and detect maneuvers from our systems in space. We're going to look space to space and look for maneuvers. You also mentioned that you're getting ready to launch your first spacecraft. So tell me more about that. Yep. So this is a fully internally developed spacecraft that we're launching in partnership with Apex Space. So they're providing the bus. We're providing all of the mission capabilities. We're going to have that long wave infrared technology that I just mentioned that's

going to allow it to be able to see in the dark. It's going to be able to look space to space, look for other things in space. They're also going to look down to the ground and look for things on the ground. We're also launching our mission data processing capabilities. So think about this as taking our compute, our expertise in compute and AI and moving it into space. Most people may not know this, but spacecrafts tend to be pretty dumb. Most spacecraft have less computation ability than your iPhone. They don't have the ability to run AI. They don't have the ability to run all sorts of processing. So we're taking modern day GPUs, the foundational hardware that's necessary to run AI processing. And we're

getting that ready for the space environment. We're hardening it for the space environment. We're going to move that into space. And that's going to allow us to do more and more processing at the edge. So you can think of this as initially, we're going to run experiments, we're going to ask our spacecraft to be able to do things, you know, look for objects, you know, look at the ground, look for specific anomalies or behaviors without us having to joystick it, command and control it day to day. We're going to give it high level intent and let it do the processing and then send back the results. But you could imagine this evolving into more and more sophisticated autonomy

as time goes on. That would be a game changer. It's literally, you went exactly where I was going with you for my next question, because Lattice has been your jurisdiction. And that, of course, is the software. It for things like AI in space are going to evolve into. Yeah, I love that. So first things first, you know, we're launching our mission data processor and we want to provide that as an agnostic compute platform, a substrate for anyone to develop on. So we're happy if Lattice runs on it, but we're also super thrilled to let anybody run software on board, including our partners that we've announced and future partners that we're going to announce. Now, as part of that, though, on our spacecraft and everything we're going to release, Lattice is going to be on it. And Lattice is one of the things that we're going to ensure is compatible. We see Lattice as the common command and control

and autonomy substrate across all of our products. So every one of our products, they're operating in different domains, but they all can communicate over Lattice. And that's really what I'd like folks to take away. When you think about Lattice, it's the common way that things can move data between each other. They can move command intent between each other. How does an operator command and control these systems? And think in terms of how operators think, right?

You shouldn't have to learn the manual, like a thousand page book in order to operate a spacecraft, which is gonna be different from a submarine. Operators should think about, hey, I wanna move here, I wanna get eyes on this thing, I wanna do something about a target that I saw and let Lattice take care of the underlying, how do I do machine to machine communication to get that outcome to happen? And then the final layer above that is autonomy. So

instead of commanding, now you're giving intent. I want you to patrol this area. I want you to look for this thing and do this if you see that. And those are all things that are going to be enabled by Lattice. The spacecraft that we're launching next year will indeed be running Lattice on board our mission compute framework. So how much of this, and we're talking about it from a national security lens, but do you see commercial applications for this as well? For Lattice? Yeah, and also just for some of, for the spacecraft that you're building out, this idea of edge compute, you know, AI edge compute, we'll say edge of earth compute, you know, AI edge compute, we'll say edge of earth compute. You know, are there commercial, do you think there's customers in the private sector for this too? Or no, you're just, you're specifically focused on the missions at hand. So we are a

defense company. I mean, our focus, what gets our team out of bed every morning is focusing on national security and supporting the US and its allies. That said, you know, we are building our products as commercial products. So the Mission Data Processor is a commercial product. The Longwave Infrared Sensor is a commercial product. Even Lattice is a commercial product. And we have made commercial sales. We're in the process of discussing with a couple commercial space companies the ability to license and use Lattice and also things like our mission

data processor for their own needs and we're super thrilled to support them on that but our team my team our focus is always going to be the U.S. national security mission and so we touched on it before but just to go back to it what a Trump administration is going to mean for space policy and how you're gaming that out it's a little too early to speak on that right now. I mean, obviously, I'm very thrilled. President Trump was the one who created the Space Force. So we're really excited about what this could mean in the future. And I know everyone's really excited just about the energy that we're feeling on this. But given how early we are in the future administration, it's probably premature for me to talk to any specifics. Okay. And then I guess just in terms of assessing the threat environment in the space domain and what that means in terms of how you're thinking about those investments and those future

product pipelines. Yeah. I mean, I could break this into multiple levels. So one is, you know, about a year ago, our CEO, Brian, he was very clear with us, which was our ability to operate as a nation is going to are not in the Space Force. And everyone is now saying, hey, space is increasingly contested. The term is it's a warfighting domain, and we have to secure and protect space. So I think we've never felt it's more critical than now. The second thing is, I mean, you can just look at just the sheer commercialization, not even the defense side of space. The number of systems that are going up there is exploding. It's exponential. And so it's never been more important for the U.S. to maintain eyes and ears on everything that's happening, just in the same way that we monitor shipping ports or shipping lanes.

We need to provide that capability in space to protect just the commercial investments. And then if you layer on top of that, national security is, of course, incredibly paramount. So much of how we operate on the ground that we take for granted, right? Like you and I take for granted GPS. I can't drive anywhere at this point without GPS and Google Maps telling me where to go. That's all powered by the space layer. And if you think about our armed forces, their ability to go where they need to go and do what they need to do on the planet Earth depends on that space layer to know where they are to communicate. So protecting that layer is incredibly important as the number of systems, as I mentioned, is going to explode over the next 20 years.

All right. Gokul Submarinium. It's great to speak with you, Evandro. Thank you for taking the time. Morgan, thank you so much. It's great to meet you. That doesandro. Thank you for taking the time. Morgan, thank you so much. It's great to meet you. That does it for this episode of Manifest Space. Make sure you never miss a launch by following us wherever you get your podcasts and by watching our coverage on Closing Bell Overtime.

I'm Morgan Brennan.