Pod Insights Summation with Auren Hoffman Anduril Co-Founder Trae Stephens - Hard Tech & Modern Defense Strategy About this episode Trae Stephens is a partner at Founders Fund and co-founder and current executive chairman at Anduril, a defense technology company that produces software-defined, hardware-enabled major weapons programs. In this episode of World of DaaS, Trae and Auren discuss: Revolutionizing defense technology manufacturingBillionaire founders in defense techSupply chain resilience for military hardwareTechnology's impact on modern relationships Looking for more tech, data and venture capital intel? Head t...
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:02 Auren Hoffman? Hello, fellow data nerds. Welcome to World of DAS. I'm your host, Orin Hoffman, CEO of Incubate and GP of Flex Capital. Discover more episodes, get weekly data as a service news, original content, articles on data and more at worldofdas.com. That's world of das, D-A-A-S, dot com. Hello, fellow data nerds. My guest today is Trey Stevens. Trey is a partner at Founders Fund and co-founder and current executive chairman at Anduril. Trey, welcome to World of DAS.
0:33 Speaker 01 Hey, Orin. Good to see you.
0:34 Auren Hoffman? Great to see you as well. Now, what does a new, modern day defense company look like? Or what should it look like?
0:43 Speaker 01 There's probably a lot of different things it could look like. We saw over the last 20 years what happened with SpaceX, which is a space launch company, satellites as well. And then you have a company like Palantir that's all about data infrastructure that has a huge business in defense and intelligence. Anduril is kind of a very different form of company in this space. We're focused on software-defined, hardware-enabled, major weapons programs. So everything from surveillance and reconnaissance to kinetics, to autonomous systems, undersea, on the surface of the sea, on the ground, in the air, and in space. The biggest lever that you'll see out of the defense technology space is really about how to create products
1:24 Speaker 01 that are leveraging private capital rather than taxpayer capital to develop capabilities and then sell them at lower costs than the alternative of doing like an in-house build inside the government, which is the way that these things have been done for the last 60, 70 years.
1:38 Auren Hoffman? And one of those three companies is that even though they may have some very good people in the DC area, almost all the executives and the product people are based outside the DC area. Is that where you think more and more of these companies will be going in the future?
1:55 Speaker 01 All of these companies have a large presence in Washington DC. Palantir, our second largest office over the entire history of the company, was in DC. That's probably not the case of SpaceX anymore, given Brownsville, but it's probably the third largest office for SpaceX. And it's definitely the second largest for Anduril as well. There is some engineering work. That has always happened in DC at these companies. But the core engineering, research and development capability is back at headquarters generally. The reason for this is largely cultural. I think there are talented engineers that are based in the Beltway. They're just trapped inside of a bureaucracy that doesn't value them in the same way that they're valued in a traditional Silicon Valley company. And I don't mean Silicon Valley geographically, I mean culturally Silicon Valley.
2:41 Speaker 01 I wouldn't anticipate that that would change.
2:43 Auren Hoffman? A lot of the defense primes, you have to clock in and clock out. Everybody actually has to time keep. I don't know, it just seems very much of a nanny state thing. And maybe there's a reason for that. I wouldn't want to work at a company like that just for that one reason.
2:58 Speaker 01 This was like a major thing that we tried to work out culturally at Palantir in the early days. And when you have bodies on contracts, you do have to time keep, submit your hours and things like that. And we had to figure out processes and machinery inside of the company to abstract that away from the engineers who were morally offended by the idea that they needed to do that.
3:20 Auren Hoffman? They would just quit if that happened.
3:21 Speaker 01 Exactly. So that was like a real thing that we had to deal with at Palantir back in the day.
3:26 Auren Hoffman? If you think of Palantir, SpaceX and Anduril, one of the other things they all have in common is they're all founded by billionaires. I don't know what we're thinking, but if we're long-term wanna be a great society and have great defense. I don't know that all of our great defense companies are gonna be able to be founded by billionaires. Or maybe that will be the case. How do you think that is gonna be the case going forward?
3:46 Speaker 01 I don't think that's random chance. I think it is actually a bug in the system that is what it is, unfortunately.
3:53 Auren Hoffman? I don't think it's just the billionaire. It's also that they were rock stars. The founders are like these famous people that I imagine generals wanna meet. Maybe it wasn't just the money, but it's the fact that they are famed and they're known for being these creative individuals that was able to get them through the very hard early stages of procurement.
4:15 Speaker 01 You've nailed it. I mean, this is one of two things that I think that brings. The first thing the billionaire founder brings is it makes it much easier to raise capital, for better or for worse.
4:24 Auren Hoffman? And you need a lot of money to make these things happen.
4:26 Speaker 01 And you need a lot of money. Both Palantir and SpaceX took over five years to get to 10 million in revenue. You really need to have some sort of capital advantage. But then to your point, the second major advantage is that members of Congress want to meet you. It's much easier to score meetings with the people that are moving the money around in the system. That is definitely an early advantage. In the Anduril case, we were the fastest since the Korean War to go from company stand-up to program of record inside the federal government. It was just over two years. The craziest aspect of this is that we, me, Palmer and Brian, actually met with the customer, the person that signed off on that program of record, three days into the company.
5:09 Speaker 01 So that's the sort of advantage that you get from having the billionaire co-founder and people that had experience in the space before is that you're not playing, how many degrees separated from Kevin Bacon do I need to get to the decision maker?
5:22 Auren Hoffman? Who would not want to meet Palmer Lucky? Of course you want to meet him. He's just like amazing. But the Palmer Lucky before he was Palmer Lucky, or Elon Musk before he was Elon Musk, they're going to have a very hard time starting something in defense world. Or do you think it's still possible? They just have to deal with the more difficult things that they have to deal with?
5:42 Speaker 01 I think it's possible. It's certainly like being able to stand on the shoulders of the companies that have gone before you is really helpful. Our run at Anduril was largely predicated by lessons learned from SpaceX and Palantir, and that was a huge asset.
5:56 Auren Hoffman? And now we have three, and then it's just a little bit easier to have the fourth once you have three.
6:00 Speaker 01 Exactly. But at the same time, as you well know, category investing is tough. If you were a space investor and you didn't invest in SpaceX, you probably lost money. If you're a crypto investor and you didn't invest in Coinbase, you probably lost money. If you're a social media investor, you didn't invest in Facebook, you probably lost money. And so I think there is a bit of a weird dynamic going on in the marketplace right now where people have this belief that, oh, defense tech is wide open, we're going to build. Twenty Palantirs, and that just flies in the face of everything that we've learned about the parallel nature of venture capital over the last fifty years. I just don't expect that that will be the case. The more competitive the environment gets, the higher the prices for these companies get bid up, the harder it is for venture investors to make returns, and everything is just going to be a race to the bottom.
6:45 Speaker 01 So I'm not really sure that, from a category perspective, it becoming slightly easier is necessarily the boon that everyone thinks it is.
6:54 Auren Hoffman? Now, for a company like Andor, it really is about defense. It has to sell to the government. The government is the main customer. You can reimagine a Palantir where the first customer is JP Morgan, and then like over time, it sells more to the defense and national security type customer like Snowflake is doing now, where Snowflake probably sells into the national security customer today. If you saw a company that was trying to sell into both, would you just say, don't do it, just sell into one first and get to a hundred million in revenue and then try to sell into the other? Or what would be your advice to a company?
7:29 Speaker 01 Both Palantir and SpaceX, an enormous chunk of their business is commercial. At Palantir, it's about fifty fifty, I think, from a revenue perspective.
7:37 Auren Hoffman? Today, but it started off much more B to G, right? Or no?
7:42 Speaker 01 It did, yeah, which is where I was going with this. I think you can choose your adventure at the beginning. You're either hard commercial or hard government, but trying to mix the two of them is not generally a great idea. One crazy story that I'll tell, that was a bit of a research project for me a few years back, is I was trying to understand where this term dual-use came from. It gets thrown around a lot in government circles. There's like entire acquisitions programs that are focused on acquiring dual-use technologies. If you go back to during the Cold War, you'll find that this term was only used in reference to nuclear technology transfer. Okay, we talk about it as nuclear has a dual use. It has a power usage and it has the weapons program usage. That was it. That was the only appearance that it made an congressional record.
8:27 Speaker 01 Famously, in the summer of nineteen ninety-four, there was this seminal event in defense history that has become colloquially known as the Last Supper. And the Last Supper is where Secretary of Defense pulled together all the CEOs of the defense companies and said, "The Cold War is over. Spending is going to go down. Consolidate or die." And basically challenged them to go out and merge and become more efficient. October of that year, so just a few months after the Last Supper, you started seeing the term dual use getting thrown around a lot. It started appearing in Congressional Record, but used in a different way. It was being used as a defense for all other applications, commercial and defense applications for all sorts of technologies.
9:07 Speaker 01 And my hypothesis, given the increase in usage of that over the last now thirty years, is that the defense primes have run a masterful campaign. They've said, "We're going to do our part. We will consolidate, so there's only five or six of us that can win these major weapons acquisition programs. But the agreement that we're asking for from the Pentagon is don't let anyone else come to the table." And so dual use is a pejorative. It's the defense primes' way of saying, "These are people that can sell as vendors on our contracts, but they cannot compete." And you saw how that came up with SpaceX, with being blocked from bidding on national security launch programs by ULA. And it came up with Palantir, with the distributed common ground system contract with the Army,
9:50 Speaker 01 where they were saying, "Look, Palantir, maybe it's a dual-use capability. It can't win a major program." And so I try to always remind people, dual use is not real. If you are a seed stage or a Series A company and you're running around saying like, "I'm a dual-use capability. I'm equally selling to federal as I am to commercial use cases with oil and gas companies or whatever," you have been hoodwinked. The primes have done what they've wanted to do, which is prevent you from being a credible threat to them on their cash cows. Anduril is not doing that. Anduril is, "We will look more like Lockheed Martin than we will Palantir or SpaceX." We're going to have ninety-seven plus percent of our business government, and we are going directly at these major defense acquisition programs, head-to-head against traditional players in the industry.
10:34 Auren Hoffman? Now, why does Department of Defense outsource almost everything? I get like the super high-tech stuff like Anduril. You gotta go buy that. That's not something you're gonna build internally. They have all these contractors that are like cooks. They pay contractors to buy ammunition for them. It's historically, you'd have like cooks in the army, or you'd have somebody in the army who like knows how to buy ammunition. Not everybody shoots guns in the army. They do other stuff. But it seems like the army really wants to outsource everything except the people that shoot.
11:04 Speaker 01 This wasn't always the case. The government literally used to own and operate munitions factories.
11:10 Auren Hoffman? It still kind of technically does in a weird way. They still own a bunch of factories. I don't know if they operate them, but they do own a bunch of factories around the country.
11:18 Speaker 01 Not to mention test ranges and the labs are building the kinetic payloads for missiles and stuff like that. There's just like this complete change where during the Cold War, there was this realization that things weren't being run super efficiently. And they decided to shift as much as possible to these cost-plus contracts because they felt like they were being extorted by industry. So if they could just agree on like a fixed margin on top of whatever the cost of a program was, well, surely that would make us more efficient. Now, what it didn't do is it didn't protect against the deterioration of the incentive structure where there are always cost overruns and time overruns because the incentive structure being set up rewards that behavior.
12:00 Auren Hoffman? Cost plus is so stupid. I understand like the reason why they did it, but I don't understand the reason why they've kept it.
12:06 Speaker 01 Well, at this point, I feel like it's very, very hard to change. These are huge publicly traded companies. Their investors have like a clear expectation for how they're supposed to be doing their business, the margins that they should be getting off of these programs, how they do accounting on these programs. If you were to snap your fingers and say, Lockheed Martin, I need you to go from one and a half percent. Revenue I R A D spending to 20% I R A D spending and then do everything firm fixed. How would that work in the public markets? There would just be no understanding of how that would happen.
12:37 Auren Hoffman? Right now, Department of Defense, they'll go hire a 25 year old consultant from Booz Allen, who doesn't live in the DC area. Maybe they live in Chicago or something and they fly them to the DC area every week. They put them up in a hotel and then that person helps them buy ammunition or whatever that person does. And I'm sure it's a smart person, but they could just hire that person. They're probably paying all in like 1.2 million a year for that person. They could hire a very smart person for a lot less than 1.2 million a year, especially a 25 year old. Now, I understand part of the reason that they do it is now they can fire them and they could hire at will and they could do all these things that they don't, maybe they don't have the flexibility of doing
13:21 Auren Hoffman? in the normal government, but why don't we just like give them the flexibility? And then they could do what they want and not have to pay 1.2 million a person.
13:29 Speaker 01 It's certainly hard to like create some sort of arbitrary justification for why they're doing things the way they're doing, because I think most people would agree that it's broken. On the Booz Allen side of things, the crazy thing is, did you know they hold the largest AI contract across the entire federal government? Booz Allen, not OpenAI, not Palantir, Booz Allen Hamilton.
13:49 Auren Hoffman? There's no amazing AI person that's going to go work for Booz Allen.
13:52 Speaker 01 No, of course not.
13:54 Auren Hoffman? I know some nice people at work. I've never met like a great AI person who's like, oh yeah, I'm going to work for Booz Allen. There are zero great AI people at Booz Allen. I know someone's going to sue this podcast for me saying that, but I will go on record and be very, very confident that there's no great AI people who work at that company.
14:11 Speaker 01 It's actually funny. We hired a lot of people out of Booz Allen at Palantir back in the day, including my co-founder at Androol, Matt Graham. We used to say that Booz is a great first job. The best way to find really talented people to hire for Palantir was find people who were super smart and realized how dumb it was to work at Booz Allen.
14:31 Auren Hoffman? But if they're there for more than a few years, it's too late.
14:34 Speaker 01 This is actually the same thing with officers. A five year contract in the military as an officer, you should end as a captain. If you go through the wickets and you progress and get promoted, you should end those five years as a captain. In some ways, if you leave as a major, it means you stayed after five years. Why did you do that? What was motivating you to extend your contract beyond five years?
14:56 Auren Hoffman? That one I understand more because there are certain things you can do in the army that you can't do anywhere else in the world. I could understand someone staying all the way to general because they really a dedicated public servant. Booze is a different example. I just don't understand that one.
15:11 Speaker 01 It's a very generous explanation. I started my career in the intelligence community and I think if I had stayed there for six years instead of two and a half, I would have walked away with a very different culture about my expectations. Once you've been living in a system and you become indoctrinated in a system, it's really tough to break out, change your expectations around performance and things like that. So when we hire people out of the military, our general guidance is the captains are going to be your best option. That's like the highest talent because they stayed exactly as long as they needed to. And then they realized that there was something that was broken that they really cared about going and fixing in the private sector.
15:47 Auren Hoffman? I have hired a decent number of captains in my life. They all actually were really good.
15:52 Speaker 01 It's a bit of a shortcut.
15:53 Auren Hoffman? It's a good point. That's pretty interesting. On the geopolitics war, it seems like the foreign policy establishment, at least this is my opinion, I don't want to put words in your mouth, has gotten most of the very big things wrong over the last 25 years. Maybe it got a lot of little things right, but the big things, Iraq and some of this other stuff, wrong. First of all, do you agree with that? And if you do agree, why is it the case? These are smart people. Why is that the case?
16:20 Speaker 01 They're pretty idiosyncratic. I'm not sure that there's like a single thing that led to these individual cases going wrong. Obviously, we had to do something in response to 9/11. I'm not sure that occupying entire nations in the Middle East was the right approach, but the mission alignment with what we were trying to do by eradicating the state-sponsored terrorism was the right instinct. It just obviously didn't play out as we probably would have expected it to. A lot of what causes these things to happen is a misunderstanding of what our role in the world should look like and an incredibly volatile, but intentionally volatile, political environment where you're constantly flipping back and forth between parties. You have political appointees that realistically probably only end up
17:06 Speaker 01 Spending two, three years in their role. We have no long-term plan. If you went in the way that China has done, it's more of a marathon than a sprint. And they've laid out a many decades plan for where they want to be. When you have that sort of articulated strategy and you have pretty consistent leadership. My guess is that it's easier to go out and achieve some sort of longer-term goal. I'm not saying that that's what we should be aspiring to. I actually think the democratic system is much better.
17:33 Auren Hoffman? Even if you just say the Bush years, which were somewhat of a consistent government with a consistent goal and et cetera. Over those eight years, it seemed like they got a lot of things wrong. And then when they even did those things, they prosecuted them wrong too. So it's like two by two. You want to have good strategy versus bad. But if you have bad strategy, at least you want to have good execution, not bad. And I think they had like bad and bad in many cases.
17:55 Speaker 01 The mission accomplished banner was definitely misplaced. I actually think that the George W. Bush administration, they really believed what they were doing was the right thing. They really believed people deserved to have democratic institutions and to experience Western views on freedom.
18:11 Auren Hoffman? They were Wilsonian at the core.
18:13 Speaker 01 Totally. And I think the real challenge with that is that I think we all learned during that time period that you can't fast track the enlightenment. We had 500 years of enlightenment.
18:23 Auren Hoffman? There was no Thomas Jefferson there and Alexander Hamilton and stuff.
18:27 Speaker 01 Exactly.
18:29 Auren Hoffman? Is it our hubris? Is it we're getting lobbied by different internal and external forces that move us into these positions? Why do we make these mistakes? And by the way, because we're so big, they're all recoverable. If a smaller, more fragile country made these mistakes, the whole country would topple. For us, it's just like a little blip. It was like a one-week news cycle when we pulled out of Afghanistan.
18:54 Speaker 01 I definitely think that hubris has a role to play with that. A lot of times that gets kind of wrapped in with a question about our role on a global sphere. Given the power dynamics, at least the power dynamics that existed since the end of the Cold War. Obviously, there's like a very new paradigm emerging with the rise of China. But not to riff on a great movie from 20 years ago, Team America. World Police, but I think that there are a lot of people that actually thought, yeah, maybe we should be Team America World Police. We want to maximize justice and freedom in the world. And when we see injustice, we should go tamp it down. You could still today in 2025, make a credible argument for why that's the case. Obviously, the current administration is drawing a line and saying, we are not Team America World Police.
19:39 Speaker 01 They have their own reasons for believing that, their own philosophies. I believe that because we have this political system that shifts very rapidly, both in Congress and in the judiciary and in the presidency, you get this crazy vacillation between these poor philosophical foreign policy ideas. And does that increase the rate at which we make mistakes? Maybe. Does it decrease the likelihood of explosion into total war? Maybe that is kind of a stop on the system as well.
20:06 Auren Hoffman? If we think of the big thing that I assume that everyone is worried about in national security is this great power conflict between the US and China. These are near peers. They're a very smart adversary. How do we build the resilience and the defense without tripping it over into a hot war?
20:28 Speaker 01 The core concept that is what we've been working on at Androol and it's what we believe deeply philosophically is in deterrence. As Reagan said, you want to have peace to strength. If we have such a monumental advantage technologically, militarily, in theory, it would be really unwise for adversaries to test that advantage. We've seen the longest stretch of relative peace since the end of the Cold War. And I think that was largely the cause of the United States having this monopoly on power in the world. That said, I think the other part of this is that we've agreed as a culture and society that these threshold conflicts are somehow more acceptable. That as long as you have a conflict that runs over many, many years
21:12 Speaker 01 and is much lower intensity and doesn't trigger us into nuclear conflict, maybe somehow that's better.
21:19 Auren Hoffman? Essentially, it's really about the body count. As long as the body count is relatively quote unquote low and quote unquote acceptable, and there's not as many people complaining about it, and we don't have to do a draft, all these other types of things too, I imagine are part of it too.
21:34 Speaker 01 I think that's sort of the case, not as much as you would expect because the body counts get really high in these threshold conflicts. Probably close to a million people at this point have lost their lives in Ukraine. The casualties count is really high and it runs for multiple years. And obviously, we're seeing what's happening in Israel and Gaza.
21:52 Auren Hoffman? From the US sense, they care about the US casualties.
21:56 Speaker 01 I mean, certainly, no one wants to write letters home to mothers and fathers of US service members that are caught in combat. Aaron, you're obviously really familiar with this. Rene Girard, the philosopher that developed memetic theory, he actually pointed this out as a real problem. It turns out that these sorts of threshold conflicts are memetic violence. We're just locked into no one wanting to cross the line, and it leads to these massive body counts over time. And actually, probably if you were going to strike and you were going to compete with certainty, you would actually just go total war from the first moment and end the conflict as quickly as possible. The theory would be that you would actually keep the loss of life much lower than it would be if you let that conflict run over a long period of time.
22:40 Speaker 01 Peter Thiel talks about this in Zero to One as well, in a business context. I think it's chapter four. He says, if you're going to compete, act decisively and win as quickly as possible. I think that's how you break down and short circuit the memetic rivalry that ends up being far more violent in the long term.
22:58 Auren Hoffman? The flip side of that would be, let's say there's an 80% chance that you end it quickly, but there's the 20% chance that it's 100x worse.
23:09 Speaker 01 If it leads to nuclear war, then that's a bad gambit. That's always the question.
23:13 Auren Hoffman? I mean, you never know how these things end up. I don't know if I trust the decision makers in the US government to make that call to go all in. Because I feel like if it spirals out of control, it could spiral out very, very, very badly.
23:28 Speaker 01 But then you have the question of is the way that we've supported Ukraine just at all? We haven't given them the capabilities they needed. We've given them the old stuff in the Cold War inventory. We've prolonged the war, but we haven't given them what they would need to decisively end it. There are questions on the moral side on both of this debate.
23:45 Auren Hoffman? Totally. I don't know what the right answer is, but I don't know that it's an easy answer.
23:50 Speaker 01 It's not a need to answer. That's true.
23:52 Auren Hoffman? That's interesting. One of the goals of Anderall is to reduce the cost of engagement. High cost of engagement makes it hard to react harder to make decisions. But they're both these flip sides of a lower cost of engagement world.
24:06 Speaker 01 The reality is if we got into a hot conflict with a great power, we would run out of munitions in a week. Gone.
24:13 Auren Hoffman? By the way, I have friends in the military who they barely are able to shoot rounds in practice. They're not allowed to do that, both because the rounds are way more expensive than they should be. And there's a limited number.
24:25 Speaker 01 That's right. We've built these capabilities that are incredibly exquisite. They are incredibly custom. They have really complicated supply chains. What that means is that our partner nations, Saudi Arabia, for example, is fighting this kind of ongoing conflict with the Houthis. They've got stuff being shot into their sovereign territory and creating havoc on a daily basis. And we cannot sell them enough Patriot missiles. They literally have to go to other partner nations and try to buy their inventory of Patriot missiles. And then you see situations like Ukraine where we deplete not only the available inventory, we also deplete our own inventory of the capabilities that we're sending them in order to support the war effort with no ability to actually resupply.
25:08 Speaker 01 The primes are literally calling people out of retirement to rebuild assembly lines to make some of these capabilities.
25:15 Auren Hoffman? I've actually dived into some of these ways that we make shells in the United States. It's very archaic and sometimes it hasn't changed in like 100 years.
25:23 Speaker 01 That's right. We're still using incredibly antiquated manufacturing processes. And this is one of the things that we spend a lot of time on at Anduril is we want to leverage commercial supply chains to the extent possible. If you look at something like our sentry tower, which was the first product that we use to defend military bases as well as national borders, basically everything on the sentry tower is a commercial product. The radar is a commercial product, the camera, the solar panels, battery systems, the power management systems, the tower itself, everything is a commercial product. Because of that, we have options as well. If one of our solar panel vendors cracks the bed, we can always go and get another one. We can go to a different optical system if we need to, so on and so forth.
26:04 Speaker 01 And I think this is not the way that these things are traditionally done, but if we can build our supply chain in that way at Anduril, we should be able to move significantly more quickly and we should be able to ramp scaling and production faster as well. Another great example of this is our Barracuda missile that we launched a few months ago now. The fuselage of the Barracuda missile uses the same hot pressing production process as bath fittings, as bathtubs. So we're not using some crazy new technology to build it. No, we're just leveraging the way that industry has done these things for a long time, making the system much more modular.
26:40 Auren Hoffman? I'm imagining these bathtubs being shot out with like a guy in it. Rubber duckies going everywhere.
26:49 Speaker 01 That would be incredible. Instead of dropping ammunition payload, it drops thousands of rubber duckies.
26:54 Auren Hoffman? If you drop a rubber ducky on people, people would freak out. They would probably just run away because they would have no idea what happened.
27:01 Speaker 01 It's very Dr. Strangelove. I love it.
27:05 Speaker 01 We want to go from thousands and thousands of independent parts to 10x smaller number and be able to do that so we can produce at scale.
27:14 Auren Hoffman? Just like Tesla has done in their own factories where just the number of parts has gone down dramatically.
27:19 Speaker 01 That's right. That's exactly what we should be working on.
27:22 Auren Hoffman? That also helps with any type of supply chain disruption because just the chance that any one part is going to disrupt goes down if you have fewer parts.
27:31 Speaker 01 Exactly right. That doesn't mean that we can be lackadaisical about supply chain. We have an entire team at Anduril that's entirely focused on supply chain resilience. We understand that that's obviously a very core part of getting the scale production story to work.
27:45 Auren Hoffman? On the venture side, Founders Fund, probably more than almost any other well-known firm, has invested deeply in like hard tech. And that includes things like hardware and some of these other types of things. How do you guys think about that vis a vis? Is that just a counter positioning versus other venture firms? Or really just because it's more interesting, it's core to the philosophy? Why has that evolved that way?
28:09 Speaker 01 There used to be this tagline on our website that said, "We want to imply in cars." And instead, we got 140 characters. I think this is definitely built into the ethos of the firm. Part of this is just the anti-competition idea that Peter has been preaching for the last 20 years. A lot of potentially interesting technology startups are in software categories where it would not be that difficult to build a direct competitor, which just leads to this memetic violence that happens in the marketplace. When you build harder tech, it's much less likely that those things become highly competitive. We have a philosophical tendency towards that. That said, we're also large investors in Palantir, Airbnb, Stripe, and a bunch of software companies. So we're certainly not focused on that as a category.
28:54 Speaker 01 The other thing that's worth mentioning is that we have invested in a bunch of hard tech companies. One of the commonalities that we found between the success stories in the hard tech ecosystem is that for every incredibly talented tech founder, you really need an incredibly talented business founder alongside. It's not sufficient to believe that it's the field of dreams and that if you just build the product, everyone will magically show up and it'll just be this massively successful thing. It is not sufficient.
29:22 Auren Hoffman? It's not the same person.
29:24 Speaker 01 It's usually not the same person. It can be in some cases, people like Elon are uniquely talented.
29:29 Auren Hoffman? Or Colson Brothers or something.
29:32 Speaker 01 Exactly. But generally speaking, it is undervalued in hard tech companies.
29:37 Auren Hoffman? It's commercially oriented.
29:39 Speaker 01 Exactly. We try to make sure that that filter is cleared before we get terribly concentrated on the tech side of things.
29:46 Auren Hoffman? Interesting. I imagine the hard tech, just the feedback loops are much longer, much harder. You can go year over year and have less understanding of how the company is progressing, what's going on. Whereas like in a SaaS company, it's easy. You look at net retention and your, your growth and maybe one or two other metrics and you have a really good sense of how things are going at the SaaS company. Is the best way to do it just really diving into the product and seeing the product? Or how does one know if it takes four years to get a big contract? How does one know if things are progressing the right way?
30:24 Speaker 01 Obviously, on an absolute versus relative basis, it's kind of hard to define what a long time means, but the best hard tech companies tend to figure out a way to commercialize milestones. So you look at SpaceX, obviously their long-term goal is to send people to Mars, but that's not the only product. They're selling Falcon 9 launches, they're selling Falcon Heavy launches, they're selling Starship, taking payload to space, lowering the cost of sending a kilogram of mass to space.
30:52 Auren Hoffman? They sell connectivity now, Starlink.
30:55 Speaker 01 Yeah, they sell connectivity through Starlink. And then you have companies like Neuralink, obviously like getting to full-scale human BCI is the long-term goal, but they're selling basically to people with chronic illnesses that solve some problems for them along the way. At Anduril, we just recently won a contract to build the collaborative combat aircraft, which is like an autonomous fighter plane. We didn't start by selling autonomous fighter planes, we started by selling sentry towers.
31:20 Auren Hoffman? Which are easier to get to market, essentially.
31:22 Speaker 01 Way easier.
31:24 Auren Hoffman? The founders have to be smart enough to get their first product a little bit lower ambition, a little higher percentage chance of being able to get it done.
31:33 Speaker 01 And everything you learn from those earlier stages is critical to the long-term goal. It's not that you're just building something random to finance what you're actually building. It's like you're figuring out a way to commercialize every stage of development. Obviously, there are some cases in which that's much harder, like nuclear fusion. I'm not really sure how you would have milestones commercialized along the way.
31:53 Auren Hoffman? Sometimes in the bio, it's hard because you need FDA approval or something.
31:57 Speaker 01 There are some cases, but there are many cases as well where it's just like a failure of imagination that leads to these companies requiring way too much capital and way too much time that makes it difficult to leverage venture capital as an asset class to help you accomplish your goals.
32:11 Auren Hoffman? Now, Proster is now closing like a $3 billion fund, which seems just so gargantuan when one thinks about the venture world. How do you think about this money as an advantage as an investor?
32:24 Speaker 01 We have a growth fund and a venture fund, so we're currently investing at a billion dollar venture fund. We just raised growth fund three, which is three and a half billion dollar growth fund. There are a lot of really interesting growth deals out there. Obviously, companies are staying private longer. There's enough capital in the private markets that there's continued interest in doing that. And we can write bigger and bigger checks, more concentrated checks to acquire more ownership prior to the company going out to the markets and having some sort of liquidity. The advantage really there is that the founders know that we're in it for the long term with them. The economics of growth don't affect our ability to deploy additional capital in your businesses over time. And that strategy has worked really well. The vast majority of our growth fund deals are concentrating
33:05 Speaker 01 into our existing portfolio rather than new investments outside of the venture fund. But at the same time, it is really important for us to have a venture fund as well. We're not just writing checks into mature growth stage companies. We are still doing seed, series A, series B investments. That's kind of the top of funnel in many ways for concentrating that capital at a later stage.
33:26 Auren Hoffman? How do you set up the dynamics where that $10 million check is still so valued?
33:30 Speaker 01 There's kind of a gray line for where a check would come on a venture rather than growth.
33:34 Auren Hoffman? And there's some partners, just a one versus the other or something.
33:37 Speaker 01 We're all a team, but there are a handful of people that are very growth focused. The rest of the fund is very venture focused. And our expectations on returns are different between those two funds. Venture capital is a power law, particularly at the early stage. We have to have a credible belief that every investment we make is going to have the opportunity to return the entirety of the fund. In the growth fund, we're not underwriting to a 10x multiple or anything like that. You have just a different understanding of what targets you're looking at as you're evaluating deals at different stages.
34:06 Auren Hoffman? Most people know you as this really deep thinker about venture capital, about national security, about defense tech. I also know you as someone who thinks a lot about love and family. You wrote this piece recently talking about how technology might be destroying marriage. How do you think about technology in the world of dating and modern relationships?
34:27 Speaker 01 I had this realization a few years ago that coming out of COVID, there was this expectation that we would reestablish community and everyone would be so happy to be out interacting interpersonally with other people. And it didn't really happen. We still have this loneliness epidemic that's ravaging Western society. I started seeing data that was coming out of the online dating companies about what types of people are getting right swipes, what types of people are getting messages. And then you look at some of the most heinous violent events that are happening in our society around mass shootings and stuff. Very often, these people are incels. They're incredibly lonely people who feel like they have no community, who've been rejected by their desired sexual partners, whatever it might be.
35:10 Speaker 01 A hundred years ago, or whatever, people tended to pair off in local communities. You have people in your school, or your church, or your Elks Club, or whatever. And people were matching because maybe they weren't the most attractive, but people thought that they were the kindest, or the smartest, or the funniest, or the most charming. When I go back and look at the small rural town that I grew up in in Ohio, I was honestly shocked leaving high school to find out all of the people from my high school that weren't in the same social networks in my high school, they ended up getting married. And it was because they had ties. There was connectivity that existed. That kind of local maxima way of pairing off has worked in human society for thousands of years.
35:48 Auren Hoffman? Most people married people within a couple of miles of them.
35:51 Speaker 01 Exactly, yeah. This new paradigm that we're finding ourselves in is we've increased the dating pool by a lot, by entire geographies. It's much easier to get around, so people can drive or fly short distances with relative ease. What's happened is that it's created this belief that there's always someone better out there. When you always believe that there's someone better out there, there's no issue with going. And experiencing as many people as you possibly can and being indecisive and getting stuck in this loop, this doom loop of loneliness. I don't think this is good for humanity. And the solve that technology has been looking at, because I think this is pretty well understood. At this point, people understand that this has been bad for humanity.
36:35 Speaker 01 The solve is AI girlfriends. It's not how do we get people back to connectivity? It's we can't do the connectivity thing anymore. And so we should solve it by giving them a person that will love, affirm and acknowledge them in all ways, without any challenge for the rest of their lives. But that's also not what real relationship is. Real relationship has conflict, it has emotion, it has drama. I think that we're losing part of our humanity by not figuring out how to really invest in bringing people back to intimate community with one another.
37:07 Auren Hoffman? Is the solve just that? We need more communities, whether it's people going to church together or people coming together over some sort of other in-person, some sort of other shared thing that's bigger than themself. Is that the solve?
37:22 Speaker 01 There's a bunch of different technology versions of this solve. There are the local social stuff was trying to match people together to go out and grab coffee when they virtuously realize that they're close to someone else. I think there's the non-technology solve to this. What you said is people, you should just go to church. I don't think people realize just how good the community is from religious experience. It's actually an incredible unifier. Even just like the more secular version of this is go out to a bar on a Friday and talk to a bunch of people and see what happens. Maybe that works. I know for sure that has worked for a lot of people that find themselves in healthy marriages today. On the technology side, I think it's like asking ourselves tough questions
38:02 Speaker 01 about how to address some of the core shallowness that exists in social networks. If you look at the right versus left swiping on dating apps, for men in particular, it's very height oriented. All of the people that are 6'2" or over are getting lots and lots of positive feedback. And anyone that's shorter than 5'8", they're getting nothing. There's just no positive feedback at all.
38:28 Auren Hoffman? Which is so silly. Of all the things, I understand you might want to discriminate against someone for a lot of reasons. I don't know, height just doesn't seem like it would be my top 10.
38:38 Speaker 01 You're speaking like the man that you are, Aron. I think if you ask most women, they would probably disagree with that. There is something about height in men that's very primal for women. I don't really know what it is.
38:48 Auren Hoffman? I guess you're right, but it's just kind of funny. I can understand that a long time ago, when maybe you needed to be defended or something. And so what if the guy's 5'4" and you're 5'5"? Who cares? Yeah, I understand if you're six feet tall, you need someone who's at least close to that. But if you're around the same height, who cares? I'll meet these like 5'1" women and they're like, I need a guy over six feet. I'm like, why do you need someone who's a foot taller than you? It makes no sense.
39:14 Speaker 01 You are striking at something primal that probably can't be explained. I think technology should be able to help us with this. We can find ways to surface what makes people awesome that isn't just tied into some super shallow metrics. I don't know exactly what that is, but there are dating apps. There's one called Keeper. We're not investors at Founders Sons, so I don't have to issue a disclaimer there. But the whole purpose of Keeper is finding a long-term spouse. Their whole goal as a company is helping people get paired off and get married, which is just fundamentally different from an incentive structure than Tinder or Hinge or whatever, where they are incentivized to keep people in the system for as long as possible.
39:52 Auren Hoffman? But how are you incentivized then? How does it work? Because if they do the most amazing job in the world, they like start losing all their users. Unless you get like paid every year they're married.
40:02 Speaker 01 It's a success fee. You don't pay them anything until they get married. And then when they get married, you pay out like a real bounty.
40:08 Auren Hoffman? It should be like 10 grand a year, every year you're married for the rest of your life or something. Inflation adjusted.
40:15 Speaker 01 Well, I mean, this is the college tuition ISA thing.
40:18 Auren Hoffman? Totally. Yeah, yeah. I love it. Yeah, an ISA for love.
40:22 Speaker 01 Yep, exactly.
40:23 Auren Hoffman? That's crazy. Couple personal questions. We're starting to see more and more tech and business leaders that are becoming more open about talking about their faith. 10 years ago, it was extremely rare. And now it's more common to hear that. What do you think is changing in the national conversation?
40:40 Speaker 01 We're still talking about really small numbers. The weekly church going attendance in San Francisco, the city of San Francisco is still 5,000, 800,000 people in the city, 5,000 that are weekly church going.
40:51 Auren Hoffman? Non Easter, non Christmas, average, random Sunday type of thing.
40:56 Speaker 01 That's right. Yes. Anecdotally, there's something happening. My wife and I lead this Bible study that over 100 people show up to every Monday evening. I can't even imagine that happening five, six, seven years ago. Obviously, this is all completely anecdotal, but my sense is that Silicon Valley went through this weird Easter mysticism thing for a long time. Became a trope even on the HBO show Silicon Valley where CEOs having like Buddhist monks tail them around for their days, going on ayahuasca retreats. And I think that people just found that along with money, fame and power comes an emptiness, a hole that is God shaped. And you can jam all sorts of stuff into that God shaped hole.
41:36 Speaker 01 You can jam psychedelics and mysticism and ecumenicalism and all sorts of stuff.
41:42 Auren Hoffman? Effective altruism.
41:43 Speaker 01 Effective altruism, whatever it might be. There's no stickiness to it.
41:47 Auren Hoffman? It's not Lindy.
41:48 Speaker 01 It's not Lindy. There's something about traditional orthodox religious beliefs, particularly Judeo-Christian. I don't want to say that people are coming to like complete faith or anything like that in an instant, but I do think that there's something that's much stickier about traditions that have lasted for thousands of years that claim to have some sort of monopoly on truths. People are way more curious about this than certainly I expected them to be when I first came to Silicon Valley 10, 11 years ago. It's been a really interesting thing to watch for sure.
42:18 Auren Hoffman? There's two things that like literally every human I know is. Every human I know is religious about something. They believe in something that you can't really prove that's bigger than themself. And again, it could be Christianity or it could be effective altruism, but they're religious and deeply religious about something. The second thing is that they have some sort of yearning for community. They may not be in a community, but they have a yearning for community. And then obviously like a traditional religion fits both types of things. But what's the bigger one? Is it the religious piece or the community piece that you think is so important to people?
42:51 Speaker 01 I think you're absolutely right that everyone is religious, whether or not they think there are atheism is a religion.
42:56 Auren Hoffman? Correct. Certainly, yes. Sometimes even like veganism can be religion if you're evangelical about it.
43:02 Speaker 01 Or if you're orthodox, you have a set of rules and beliefs that you hold other people in your community to be accountable for.
43:09 Auren Hoffman? And certainly communism has been a religion for the last hundred years.
43:13 Speaker 01 And in many cases, I feel like the faith that especially really intelligent people usually criticize people inside of traditional religious practices for having is not that different from faith in science. If you believe that the world emerged from nothing, you have just as much evidence for that as I have to believe that that's not the case. You have faith in some system or process to deliver you the truths that you desire. I agree with you, like everyone is religious at the end of the day. On the community side of things, I think this is one of the most beautiful parts of traditional religious expression is that there's something deep inside of our souls that needs to be connected with people who share values with us.
43:54 Speaker 01 It is an incredibly powerful way to build a community that is going to support you and that is going to be there to like hold you accountable and to bring you along as you mature and you learn. I do think that that is a really powerful aspect of people coming into these communities and finding that they are getting that hole filled.
44:13 Auren Hoffman? There's something about chanting or singing together. I still remember every day in first grade, I pledge allegiance to the flag. We all would get up and we'd all stand up. And first grade, I probably didn't even know what any of those things meant, but there was something very communal that we're all doing it together in a way that it felt very reassuring to me back then. I assume anything we were chanting and talking together and saying the same words together over and over again, it's very primal, it's very human to do that.
44:42 Speaker 01 This is like the nature of the sacred. You see this emerge in traditions all over the world. There's like a spirit orientation that happens in these community moments that they can really change people's lives for the better.
44:54 Auren Hoffman? Really cool. Couple of other personal questions. You have mentioned that you have a doomsday bunker. Beyond the obvious food and water and stuff, what do you have in the bunker? What's most important to keep?
45:05 Speaker 01 My bunker is about as traditional bunker as possible. We did get like these really cool Murphy bunk beds. You're not just like sleeping on cots. They do fold out of the wall, which is pretty great. Definitely has an armory, which is a necessary part. The next version of this that people need to start thinking about now that these systems are much better than they have been historically is you can live a pretty modern lifestyle off the grid with battery systems and efficient solar. If you're building a bunker, you should be thinking about how do you make sure that everything that you have in that space can be run effectively off electricity because you don't want to have reliance on natural gas or anything like that. That's like a big part of it. But for us, when we did this, it wasn't just about the bunker. It was also about having a really great retreat for the
45:49 Speaker 01 entire family in a rural place because we live day to day in the middle of San Francisco. That's really what we were optimizing for is the survivalist thing, but we also have a really cool house that me, my wife and my kids love spending time in.
46:02 Auren Hoffman? All right. Two last questions we ask all of our guests. What is a conspiracy theory that you believe?
46:07 Speaker 01 The nature of most conspiracy theories is that it's pretty impossible for large organizations to be competent enough to maintain a conspiracy over a long period of time. So all the traditional stuff, I think is just completely ridiculous because the government is incompetent. I don't think that they could cover up the Kennedy assassination stuff. I don't think that they could cover up a fake moon landing.
46:27 Auren Hoffman? It's almost more depressing that the government can't do a conspiracy.
46:32 Speaker 01 It is way more depressing. If the conspiracies were about Peter Thiel, I would be like, okay, yeah, that's believable. He's incredibly competent and he could probably pull it off. But the federal government? Come on, guys. That's ridiculous. I think most of the conspiracies that I find the most interesting are the ones about really competent organizations running long-term campaigns that have geopolitical value. Was FC in a Mossad agent? That's an interesting question. He definitely didn't kill himself. We all know that, but that's not a conspiracy.
47:01 Auren Hoffman? It could be some sort of misdirection campaign or something like that. Okay, I don't want to have the pipeline because I own a train company or something and I'm going to use other actors to help me do that. That would be like a conspiracy that could be done by a smart actor or something.
47:19 Speaker 01 Things on a smaller scale where there's less people involved, for sure.
47:22 Auren Hoffman? Okay, last question we ask all of our guests. What conventional wisdom or advice do you think is generally bad advice?
47:28 Speaker 01 I grew up, as I said before, in a rural public school in the middle of Ohio and I wouldn't be where I am today without the help of a lot of people. People that gave me their time and their attention when I didn't deserve it and I had no connectivity, I had nothing to offer. And I'm very, very grateful for that. I think the conventional wisdom is shoot your shot, go ask people for help, be forward. I agree with that generally, but the contrarian take on this is there are limits. I think some people have taken this advice to hustle and shoot your shot to an extreme that borders upon completely inappropriate. If you're listening to this and you're like, yeah, I'm going to cold email the hundred people I want to talk to on a daily basis or an hourly basis for the next year, you've gone too far.
48:14 Speaker 01 If they haven't responded, they're probably not going to and you shouldn't turn into a stalker.
48:19 Auren Hoffman? In any type of exchange like that, it needs to go both ways. I would always give some advice like, what can you offer? The other, and sometimes offering them is the good feeling of them helping you. I'm sure there's some seventh grade teacher that helped you and they did it really because they were a good person and they saw something in Trey and they wanted to help Trey become bigger than Trey thought Trey could be. And they gave you extra attention to help you along the way. But they got something out of it too, which was they got this great feeling of helping you as well. And now you're a different Trey. And so if someone wants to reach out to you today, they're probably trying to get you as their mentor. And then it's like, well, okay, what can they offer you? Probably a 16 year old can offer you something, depending on who they are.
49:03 Auren Hoffman? I think they need to think that through. Okay, here's, I'm going to bring this to the table. And so the exchange is going to be, yeah, I'm going to learn something from Trey, but Trey's going to learn something from me. And they're going to be very happy they had this meeting with me.
49:16 Speaker 01 That's a great way of putting it. If it doesn't work, if you don't get the response that you were hoping for, that's okay. Be gracious.
49:22 Auren Hoffman? It's like dating. Yeah. Not everyone you date is going to be your spouse.
49:27 Speaker 01 It's like dating if you're over six feet tall.
49:31 Auren Hoffman? Totally.
49:31 Speaker 01 When you have lots of options.
49:34 Auren Hoffman? All right. Well, thank you, Trey Stevens, for joining us on World of DAS. I follow you at Trey Stevens on X. I definitely encourage our listeners to engage with you. This has been a ton of fun. I really appreciate you coming on World of DAS.
49:45 Speaker 01 You bet. Thanks, Tom. Take care.
49:49 Auren Hoffman? Thanks for listening. If you haven't already done so, please subscribe to World of DAS wherever you consume your podcasts, YouTube, Apple, Spotify, and more. And please help us get discovered by leaving a review and check out worldofdas.com. That's worldofdas, D-A-A-S, dot com. And of course, connect with me on Twitter, at Auren, that's A-U-R-E-N. Would absolutely love to hear from you.
Previous Fmr Verisk CEO Scott Stephenson - building a $40 billion data company Next AppNexus Founder Brian O'Kelley - adtech, AI, advertising and more AI Page updated November 3, 2026 · transcript and summary are machine-generated