00:30:33 · Trae Stephens
[00:30:33] S0: No. Actually, not at all. So because I didn't know anything about venture, but I understood government tech pretty well, When I was just running through that first 500 pitch meetings, I decided that I wanted to try to see if I could find the next Palantir SpaceX. So incidentally, government acquisitions is all public. And so you can find out everyone that bid on contract or bid on RFPs. And so I just went and started scraping down all the data I could about companies that had raised venture capital that had also bid on federal contracts. Oh, interesting. And then I outbounded to get meetings with these companies. And I met with a couple 100 companies in the first two years that were that were doing that. Ended up making one investment into a company called Expanse. It was started out being named Cadium. This started by Tim Junior and Sean Maguire, who's now one of the partners at at Sequoia. So we we led their seed round in their series a. Ended up being a good acquisition by Palo Alto Networks. Again, it's like... It doesn't really move the needle for a fund of our size, but it was a a great company. I'm glad we invested. But that was it. I didn't find anything else. And over the course of these 200 plus pitch meetings, the realization that I had come to is that we needed a hardware like a hardware defined company that was driven by software to pair along with Palantir, which was just doing software. And one of the things that I had learned doing software sales at Palantir is that the government has no idea how to value it. They think of it as just hours. It's like hours and lines of code. And so you could deliver the most perfect software ever as a product, and they're always going to think it was too expensive. But hardware, there's a bill of materials. And so it's actually much easier to sell hardware. And so I came back to the team, and I said, well, I've taken all these meetings. Unfortunately, I don't think there's a next Palantir. But the thing that I'm looking for is a software defined but hardware enabled company that works on low cost autonomous systems. That would be like the thing that I would be excited about investing in. And everyone just sat there silently. And Peter was like, great. It sounds like there's nobody better to build that company than you. And I'm like, oh, no. No. You did this to me once.
00:32:44 · Trae Stephens
I'm working at Founders Fund. I'm not saying that I wanna build this company. And he was like, no. You should build the company. And I thought I had the ace in the hole. I was like, well, you know, Peter, Palantir and SpaceX were both founded by billionaires. And I think that's pretty important to building a government company because it takes a long time, and you need a lot of hype. You need a face to the company. And I'm like, I got him on this one because I'm not a billionaire. And he said, well, you know a lot of billionaires. I I was already friends with Palmer. We did the first institutional round in Oculus. So I reached out to him, and I pitched this idea. And he was like, that's dope. I definitely think you should do that. You should come to my house in Woodside. I'm building a... I drained my swimming pool, and I'm building a ramjet in my swimming pool. Of course. Of course. Yeah. And so I went to his house and checked out his ramjet. And I'm like, so seriously, you should... You know, you're working for the corporate overlords now at Meta. And he's like, Zuck is VR boy number one. Like, I'm VR boy number two. Like, we're we're gonna do this VR thing forever. I'm like, oh, well, I guess that... That's one billionaire down. I need to go figure out how I'm gonna do this. And then a couple months later, he called me out of the blue, and he was like, you still wanna do that company? And I'm like, yeah, man. What changed? She's like, I totally got fired. So that's how we started Anduril.
00:34:08 · Aditya Agarwal, Trae Stephens
[00:34:08] S1: Oh, man. That's amazing. It's it's actually based on what I know of Palmer and kind of that situation and, like, it's all tracks. [00:34:16] S0: Some stories are are stranger than fiction. [00:34:18] S1: That's true. What does it mean to... I mean, I I get the I get the sound bite. I understand that a software defined kind of hardware platform that kinda works across, you know, autonomously across a bunch of different form factors. But what did that mean actually in terms of building the first product or the first set of products? Was it that you needed to make sure you're working with the same software foundation across all of the different things? Like... And how is that... You know, I I haven't worked in defense. I haven't really worked at a hardware company either. So what does that mean in terms of what you had to do for the first couple of years? [00:34:51] S0: Yeah. So we we had this, like, Google Slides pitch deck that we used for our seed round. This was before there were three rounds before your seed round. It was just... Our first money in was our [00:35:01] S1: seed round. That old school.
00:35:02 · Trae Stephens
[00:35:02] S0: Yeah. So weird. And there was a slide in this deck where we had, like, a list of all the things we wanted to build, and it was, like, crazy stuff. I mean, it was, like, fully autonomous fighter planes and, like, submarines... Autonomous submarines and, like, low cost cruise missiles to compete with Patriots, like, all this crazy stuff. We had this recognition that SpaceX did not build Starship on day one. And this is actually an important lesson for entrepreneurs. SpaceX did not build Starship from day one. They built Falcon one. They built a rocket that had a single engine, and then they built Falcon nine, and then they did reusability, and then they did Starlink, and now they're doing Starship. It's an iterative process to get to Mars, but it's not a one step jump to Mars. And so that was the approach that we took as we said, we know that all of these things are going to require JARVIS from Iron Man. We need that AI brain, but we also need a quick way to deploy that, that we're going to be able to test on real world data, that we're going to be able to deploy quickly, that doesn't have this, like, massive basic science problem associated with it. And so we wanted to build something that we could get deployed into the field in six months or less. And so that was really the origin as we build this counterintrusion product called Sentry. That's just a... It's a camera on a stick. It's a little bit more complicated than that. But basically, solar power, a big steel pole, solid state metamaterial radar, an infrared camera, like a EO... Standard EO optical camera, a pan tilt unit, and compute at the edge with comms. And this sounds like not that complicated. Turns out it's really complicated. No one had really deployed, like, NVIDIA GPUs at the edge with bad comms before doing, like, object recognition.
00:36:55 · Aditya Agarwal
[00:36:55] S1: So doing edge
00:36:56 · Trae Stephens
[00:36:56] S0: inference, basically. Yeah. Doing edge inference, doing, like, object detection, tracking at range in, like, desert environments with, you know, environmental distortion from temperature. Like, it was it was really hard, but it was doable in six months. There was no basic science. We... In fact, at the time, we we didn't even think we needed to build any of the components ourselves. It was just a hardware integration problem. We ended up having to do that, which is a fun story that I'll I'll tell. So pan tilt units is, in theory, should be the simplest part of the entire stack. It literally is just like a thing that moves the sensor payload. But it turns out that if the normal way that a pan tilt unit is operated is with a joystick, where it's just like you move until you see something, and then you stop. And you look at the thing. And if you miss, you just slew back a little bit a few degrees in the other direction until you you have target on. We had a computer doing all this. And so the pan tilt unit, we'd say, go 117 degrees. It would turn 116 degrees. It would just be like a few clicks off for some reason. Now you multiply that against 100 movements. Within an hour. You're no no longer even looking remotely close at the thing that the radar is telling you to slew to. So no one had built a pan tilt unit that needed to have the level of accuracy that we needed it to have because we were using a computer to control it. So we had to build our own Pantel unit. Super annoying. Wish we didn't have to do that. This is still the philosophy of the company today. We don't want to build hardware unless we need to build hardware. And we were able to get that deployed. The company was started in June. We had a test unit in September. We had first towers deployed in December. And then twenty two months from founding, we had a multi $100,000,000 program of record with the federal government to operate Sentry Towers. And that program is still running today. It's worth billions of dollars. We have hundreds and hundreds and hundreds of towers deployed at military bases, national borders, maritime environments, both in The US and abroad. And it's the best margin profile of any product in our entire portfolio, which is with maturity and you can tighten things up. That's the way that all of our products should work over time.