E423: Castelion CEO on Elon Musk, Hypersonic Missiles & the Future of Defense

David sits down with Bryon Hargis, co-founder of Castelion, to unpack why rebuilding America’s defense manufacturing capacity has become a national security imperative. Bryon explains why decades of consolidation and outsourcing weakened the defense industrial base, why America could struggle against an adversary with significantly greater manufacturing capacity, and why Castelion is betting that deterrence ultimately requires weapons that can be produced affordably by the thousands.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00David Weisburd?

  Congratulations on the billion-dollar Series C led by JP Morgan, Carlyle, and Andreessen Horowitz. Why is developing hypersonic missiles necessary for the defense of the United States?

• 0:11Speaker 01

  There's a ton of work that goes into a successful fundraise, which this one's actually the first one that's been led by our CFO, Andrew Greitz. He did a phenomenal job. The strength of this round in particular hinges on that question, why hypersonics? We've been through a period of several decades where American dominance was so supreme that we didn't have to focus on the fundamentals. And what's happened in the past decade is that's becoming not true anymore. And unfortunately, just the inertia of large organizations wasn't necessarily addressed at the same timeframe that other people were closing that gap.

• 0:52Speaker 01

  And so we're in a position now where there's certainly sufficient munitions to do near-term operations, and you can see the administration saying that they have what they need. But I think there's a recognition now that in the long run, especially as you go up against, you say, more capable manufacturing adversaries, like let's take China as an example, they're going to have massive production capacity to produce weapons at scale. And you can't win a war if you run out of weapons. And in fact, it emboldens folks if they think that they can outproduce you.

• 1:33Speaker 01

  And so we have to get back as a country to being on top in terms of defense manufacturing, which is really what I see the purpose of Castelion doing, is we can fix defense manufacturing. We can be a large part in helping to address the time to develop these systems, the cost of them, and the scale. Because like, the only way that the government can afford thousands, tens of thousands, hundreds of thousands, whatever the number is necessary, is that the price of whatever it is has to be, times that quantity, has to be affordable. And the budgets have gone up, but they're not going to go 10x up. Like, that's not feasible.

• 2:11David Weisburd?

  This deterioration is such an underrated aspect of defense. There's a famous Sun Tzu quote, which is, "The military general of genius wins the war before it is started." So oftentimes, if you have this dominance in defense, you don't even have to fight the war. It actually leads to a more peaceful future than intuitively feels like as you're producing missiles.

• 2:32Speaker 01

  I personally 100% subscribe to that theory. Your position in the world and your ability to deter is really determined by your military strength at the end of the day. America's a cultural and economic powerhouse. And that's based on a lot of factors, including like the rule of law. But it's also based on the fact that America has had sufficient military capability to hold world order to the way it is now. And if you don't take care of that in the long run, all of those other factors can fall by the wayside. And so a lot of the strength of our military has been this kind of uniquely American mix of what they call like the military-industrial complex.

• 3:16Speaker 01

  And it usually gets talked about in a negative light, but I would say that it's been quite successful, this mix of both government and private companies to develop these type of capabilities and then to manufacture them at scale. Now, I would argue in like, the past couple decades since the Last Supper, we've really let the manufacturing part of that atrophy. And that's what Castelion and a whole range of new entrants are trying to fix by producing new systems at scale, whether they're a totally new capability, or whether they're a more modern replacement to something that currently exists.

• 3:54David Weisburd?

  And you at Castelion have decided to vertically integrate a lesson you took from SpaceX. Tell me about what else you learned being three engineers that left SpaceX to start in Castelion.

• 4:05Speaker 01

  I would actually say, I don't want to vertically integrate anything. What I would like to do is find somebody that can be a reliable supplier at a reasonable price that just delivers, and I will give them all of the work. We really vertically integrate when we can't find that. And in aerospace, a lot of the core of the problem is really not that somebody can't build that whatever device or widget that you're buying, it's that they do it at a price point that's way too high, that's not affordable, and they do it in a timeframe that's so rigorously slow, you can't iterate.

• 4:45Speaker 01

  And that just kills kind of like your development of a program, because complex systems are complex, and you need that iteration to get good at it. And so a lot of times, what it looks like is finding supply chain outside of aerospace that you can leverage to be a little bit more innovative, but can keep up with the development timeline. And if you can't find it in commercial because it just doesn't exist, then you have to vertically integrate, usually through great pain. It takes a lot of money, it takes a lot of capital, it takes a lot of time, because these are not necessarily easy things.

• 5:23Speaker 01

  And the experience at SpaceX really was eye-opening in terms of demonstrating that there's kind of like this unified way that almost every traditional aerospace company performs their systems engineering and system development. And it's very unintuitive coming out of like, college, like, why in the world would I do it this way? But okay, I don't know, I'm young, I don't know what I'm doing, so you're telling me this isn't how you do it. But it feels like this is just wrong. And I'd say like, one of the greatest lessons that I had at SpaceX was, if you just followed

• 6:06Speaker 01

  your gut and assumed like, that's not the most efficient way to do something, it turns out that indeed it's not. And you're seeing the limits of complexity in aerospace because of the process that they're following. So when you look at like, these large programs, like take SLS or Orion, you almost can't put in any more money and get a better outcome. Like, you're at the limit of the complexity that process can handle. And SpaceX demonstrated that there are other processes that you could follow and develop a more complex system and do it in a faster timeframe and more affordably.

• 6:50Speaker 01

  And I think that all of the aerospace industry is going to have to move in that direction. Otherwise, we're going to top out it, like, how complex of a system can we build as a country or humanity?

• 7:03David Weisburd?

  A lot of people misunderstand Elon Musk's strategy, and there's this famous Elon algorithm. It starts by elimination, and it ends with automation.

• 7:14Speaker 01

  Yes.

• 7:14David Weisburd?

  A lot of people assume it's flipped the other way around, which you want to automate, you want to scale. But a lot of it is, what aren't you doing? There's so many cases of this. There's a Tesla, and they were having issues developing a car, and Elon Musk asked the operative question, which is, why do we even have this feature? What would happen if we just took this out? And the first guy's reaction is like, we can't do that. This is like, a car must have this feature. And then when they thought from first principles, they're like, yeah, there's no reason for that. So they just eliminated it, and they solved this huge bottleneck in the production line.

• 7:48Speaker 01

  Best part is no part. Elon has a lot of great four-word phrases that are quite true. And I actually think this is like indicative of kind of the problem. Like, Elon would tell you that engineers will very easily hyper-optimize a process or a thing that just shouldn't exist, and they never stop to think, should I be doing this? Because it's just kind of like, the mental nature of an engineer. And I can say that because I'm an engineer. You have to like, back up and ask yourself, like, should I be doing this in the first place? And a lot of people don't think to ask that question. So he kind of forces that through every single meeting, which he meets with all of the

• 8:33Speaker 01

  technical teams, from my experience, weekly minimum. If there's some issue, it'll be much more frequent than that. And it's really helpful to keep that in mind, because one of the things that is the secret to SpaceX's success is really what not to worry about. And by that, I mean, the traditional aerospace process is, you gather up all of the engineering disciplines, and you put them in on a design review, and basically, anybody can come to that review and like, raise their hand with like, a possible issue, without really any...

• 9:11Speaker 01

  Quantitative data to back up that it is an issue. And you spend a lot of time trying to engineer out possible boogeymen without actually knowing if it's a real boogeyman or not. And the culture at SpaceX is very much, if you want to say something's an issue, that's great, but you have to bring the data because it can't be your hunch that it's an issue.

• 9:36Speaker 01

  And if not, if you can't prove it and we're all unsure, then let's just go build some hardware, do the test and see what happens. And if nothing bad happens, then we can move on and not worry about it. And you cut out so much work because you're just not worrying about the things that don't matter. And you're discovering something that nobody thought of that does matter and you're getting after that earlier. And that's kind of the secret of success.

• 10:03David Weisburd?

  So curious about this friction at SpaceX because there's basically two polar opposite perspectives you could take. You could take this purely quantitative approach, which also could be called product market fit to take into the extreme that it needs to be proven through customer feedback in order for it to be a valid strategy. And then the other side of it is using your gut, which is, I know people are going to need this, we're going to invest billions of dollars before we even know from the customer. And you could take this intuitive versus quantitative approach to everything. How did you marry that at SpaceX?

• 10:41Speaker 01

  I'm a huge proponent of talking to customers. In fact, in my view, you really shouldn't have like a product team and a sales team. It's really the same team. You need to leverage your sales organization to actually understand what it is that your company builds in terms of a product to then get to the core issues with customers to understand what their needs are and how what you can do as a company can fit what they want, while also not just directly doing whatever it is the customer set. And by that, I mean, like if you, that example back when Henry Ford was like making the

• 11:25Speaker 01

  Model T, like if I asked folks what they wanted, it would have been a faster horse. They need to have some level of creativity to be able to take the actual problem and visualize a solution to that the company can do that fits the expertise of what you have and what you can build at the company. And then present it back to them in a way that makes them feel frankly, like it's their idea. I worked on the Starshield program and did that quite effectively at SpaceX to how could we take what was being developed for Starlink, but apply it to national security? And the outcome of that was fundamentally game changing for the country's national

• 12:10Speaker 01

  security of like extremely proud of all of that work and the team that did it. And at Castelion, we kind of started the same way of talk to customers. What is your actual problem? What are you trying to solve?

• 12:26Speaker 01

  And then what should the solution look like? And Blackbeard, which is our first weapon, really came from discussions with the Navy of how do we get the F-18 back in the Pacific fight? And they gave us some criteria, like it was literally like three bullet points in an email and everything else stemmed from that without over prescribing the solution.

• 12:52David Weisburd?

  If you take that literally, is there a place for salespeople or are you just teaching your product people sales skills?

• 12:58Speaker 01

  Separating the two functions is a mistake.

• 13:03Speaker 01

  I don't care if they identify as a salesperson or a product person, but you want somebody that can do that translation. They need to have excellent technical understanding of the product, of what the company can do, because they need to make game time decisions on, yes, we could do that. And technically like it's challenging, but it's doable versus like, that's absolutely impossible and I need to push back on that concept immediately. And they need to have people skills.

• 13:37Speaker 01

  And for me, when I was running the government satellite sales at SpaceX, it was really

• 13:45Speaker 01

  caring less about somebody's specific background. But what I wanted to see is that they had persistence, they had technical understanding, they had a good personality that they could get along with folks and make relationships. And they had that persistence because you're going to be told no a lot. And you can't teach the personality traits that you need. So that's what you're selecting for. I can teach you orbital mechanics. I can teach you about whatever product that we're building. I could show you the different engineering divisions of the company and you can learn that, but I can't fix your kind of like core personality that you need to have to be successful.

• 14:27David Weisburd?

  Everyone I talked to on the show is chasing the same thing, an edge. And more and more, the edge comes down to your information, not just having it, but being able to trust it when the stakes are highest. AI is doing more of the information gathering for you every day and most tools are very good at sounding right. The summary reads clean, but can you trace it back to the filing, the transcript, the specific passage that drove the answer? Or are you just trusting the confidence of the output? For investors, that's not a minor concern. A missed filing, a missed weighted source, a context that got lost somewhere in the retrieval chain, those aren't edge cases, they're how decisions go wrong. AlphaSense is the AI market intelligence platform built specifically for this. They own the content, over 500 million curated documents from broker research and expert

• 15:11David Weisburd?

  transcripts to filings and earning calls, and they own the retrieval layer on top of it. So every answer links back to an exact verifiable source because the answer is only as good as what's underneath it. And with AlphaSense, you know exactly what that is. The edge goes to whoever could trust their information and prove it. See it for yourself. Start your free trial at alpha-sense.com/HowIInvest. That's alpha-sense.com/HowIInvest. It's an interesting thought experiment. What would you create, if you were creating the perfect salesperson from scratch, what would you put in that person? You would put technical know-how. The most underrated aspect to a salesperson, especially in a place like defense, is the

• 15:56David Weisburd?

  ability to generate enough rapport to get true feedback from customers.

• 16:00Speaker 01

  Yeah.

• 16:01David Weisburd?

  I think all things being equal, all things are not equal. The people that actually find out what the customer truly needs in very complicated and sensitive sales, for example, from limited partners or from the Department of War, those are the people that end up winning. And then they also need the technical expertise, which you could argue is part of that rapport of getting the right information. You're not going to get respect from other people if you don't understand the product.

• 16:26Speaker 01

  I don't know if there's anything else.

• 16:29Speaker 01

  I think that's like a good covering of what they need and a good archetype is like an extroverted engineer. But again, you don't have to be an engineer, but you have to have that technical understanding of the product to gain both the respect to the customer and of the internal team. Because like another huge part of the sales slash product role is selling the company on what the solution needs to solve so that everybody is like engaged to build the thing that will sell. And a huge part of like what I did with SpaceX was like argue internally for the characteristics of the system that had to be true for it to be truly successful

• 17:14Speaker 01

  in the market. Maybe one final piece I would say about traits of that person is just honesty.

• 17:22Speaker 01

  People sometimes underweight the fact that in a externally facing role, what you're selling is not just your company or the company's product, you're selling your personal reputation. Because people don't trust a company, they trust you. And so if you were telling them something, you damn well need to come through with it and then communicate clearly if something is about to be a surprise that's negative. You need to tell them and you need to tell them the truth. You need to not give them BS excuses because you'll lose that trust. And a lot of success of anyone in a sales role is being an advocate for the customer

• 18:06Speaker 01

  and someone that they trust because they're putting their faith in you to help solve their problem. Because frankly, if you hang them out to dry, it looks really bad for them that they put their trust in you and you let them down internally and other people are going to judge them for that. So you have to kind of have that perfect combination of traits to be truly successful. But if you, like, as you point out, if you do that well, you can frankly move mountains. And there's so few people that do that well, especially like in government sales space that it's kind of a running joke. It's just like how few people actually do most of the sales for these large companies.

• 18:48David Weisburd?

  I asked Jonathan Gray, President of Blackstone, this exact question, which is, what do you do in a downturn? And he said, in a downturn, all of his peers were basically hiding behind their desk. And he did the thing, which was jumping on a plane and sitting down with every LP and telling them exactly their views on it. And so many times, the reason these skills of reputation are so hard to build and so rare is because they're so hard and painful. To actually execute on when the time comes.

• 19:17Speaker 01

  It's 100% true. And yeah, you need that reputation and the trust for when things are bad, not for when things are good. That's easy.

• 19:27David Weisburd?

  The best way to communicate reputation is to have reputation.

• 19:31Speaker 01

  Yeah, you can't build reputation when you're in a bad spot and you need it. You can only build it when there was really nothing to be gained other than that's who you are and they trust in what you're doing and what you believe. And I think letting your true motivations, like, why are you doing this? Like, people need to know that's for the right reasons, too. Sure, it can be to make money for the company, but they want to know that you care. And if they don't feel that, you generally don't build trust with folks.

• 20:01David Weisburd?

  Do you worry about that? We just talked in the beginning of the interview, just raised a billion dollars. You're now gonna have a lot of mercenaries that are gonna be coming to you looking for positions. Is that always a deal breaker for you to hire a mercenary? Are there sometimes people that are so good that you just pay them to do the job?

• 20:20Speaker 01

  Motivations can come from a number of factors. In good times, somebody's motivation can just be money. But what happens in a bad time, like, they're just gonna leave the company because you can't continue to-

• 20:33David Weisburd?

  Last one on the boat is first one off.

• 20:35Speaker 01

  Yeah. And in my opinion, those are good consultants. Like, we'll pay you more. But the expectation is that you're not necessarily here for us in the worst moments. But we do a significant amount of screening for like, why does somebody want to join the company? And they need to have a good answer, not necessarily the same answer, but a good answer. And if it's kind of for me, if it's just like, hey, I want to manage a large team or direct what's happening, that's not a good reason. Because you're not going to want to manage a large team doing something complex and hard when we're behind schedule and things are

• 21:20Speaker 01

  not great. Your heart won't be in that. So it's much more important that somebody has some kind of meaningful reason why they'd want to swim through the trenches with you in the worst of times. And I think it's important that you load the company up with those types of people. Because frankly, they, just the spirit of why you're there that you can share with others is like, it builds on itself and it builds the company culture to be the correct thing.

• 21:49David Weisburd?

  Company culture is so fragile. One or two bad hires could really permeate throughout the organizations. How do you suss that out? How do you suss out somebody's true motivation in the hiring process?

• 21:59Speaker 01

  We ask a bunch of questions that perhaps seem unrelated, but it's all to try to get at motivation and why are they joining? And what is it that they truly want to do?

• 22:12Speaker 01

  Because you also get folks that'll try to get in the door in any job, but then move to some other job. And that's not necessarily a bad thing. But then you have to ask yourself, well, why do you feel like you just can't do that job now?

• 22:25Speaker 01

  But even if they do get through the interview, which does happen, like, an interview, I really, honest to God, hate interviews because it's like, I'm supposed to make some judgment of somebody's life's work in a few hours. And maybe they're just nervous, or maybe they're having a bad day, or maybe they didn't sleep. But even if they do get through, you can't hide your true motivation for very long. It'll come out. And I think that you have to be quick to address those issues because of exactly what you said. It does not take many people rowing in a different direction to really screw up the kind of choreography of a good team.

• 23:04David Weisburd?

  Speaking of a good team, SpaceX is famous for this responsible engineer culture. Tell me about that.

• 23:12Speaker 01

  Think of it as the CEO of whatever you're responsible for. It could be a complete system. It could just be a tiny box that's somewhere attached to that system. But you're the CEO of it. You own it. You own the requirements management of it. You own the engineering of it. You own the cost of it. You own the manufacturing of it. And you're given accountability to meet whatever goals were set out. But you're given the authority to like, do what it takes. You're the CEO of this thing. It could just be like a tiny box, but you own it, make it happen, tell people what you need, yell loudly if you're not getting what you need to be successful,

• 23:56Speaker 01

  and let folks know that are dependent on your success for their success. And so you end up with this kind of hierarchy model where you have responsible engineers of like, subcomponents that then feed up to a responsible engineer for a higher level system who maybe feeds up to like, a responsible engineer for an entire vehicle, whatever that may be.

• 24:22Speaker 01

  But you find in that process that good REs, it's not correlated to just like, years of experience in aerospace. Like, you can find folks that are fresh out of school. One of the unique things about that model is that once you determine that, hey, somebody managed this extremely well, let's give them a more complicated system to go manage next. And you move them up based upon their performance of doing a good job owning that system. And in that kind of model, you end up promoting based on merit and giving more responsibility to somebody that does well. And without regard to, I've been here 20 years and I should own that system.

• 25:06Speaker 01

  Like, that's not a good model. And fortunately, that's the model for a lot of companies. And that reduces your performance as a company to do it that way. And so the RE models, very powerful when implemented correctly.

• 25:21David Weisburd?

  How much of your experience in SpaceX instructs your work at Castelion?

• 25:25Speaker 01

  I think that we have taken a large number of the practices that work and continue to use them. It's not like a one-to-one necessarily mapping. Like, we have our own unique culture that it's not just a copy of SpaceX. But at the same time, a lot of the practices that even SpaceX follows are really what all of aerospace used to do in the Apollo era.

• 25:54David Weisburd?

  Sixties.

• 25:55Speaker 01

  Yeah, the sixties, like, the space program. Like, we made it to the moon in a decade. I honestly don't know if we could do that as a country in 10 years, starting now.

• 26:07David Weisburd?

  Sometimes the good old days are not just nostalgia. Sometimes there was actually more competence and more cohesion.

• 26:13Speaker 01

  The thing is, that's really frustrating. We have just so many incredible new tools and capabilities and technology that would have made like, getting to the moon easier. The problem in aerospace, in my opinion, it's not the tech, it's really the process that folks follow. And aerospace over many decades moved to be extremely risk averse. And you cannot make fast progress being extremely risk averse and thinking through every step of the process for a year at a time, just to move on to the next phase. Sometimes you just need to go build something and try it. And SpaceX, in my mind, really solidified that is the correct approach. Now, when I graduated

• 26:59Speaker 01

  college, and I joined Johns Hopkins Applied Physics Lab, straight out of Georgia Tech,

• 27:08Speaker 01

  I felt like that, like, quite frankly, like, I had the same kind of ideas in my mind, like, why are we doing it this way? This seems silly. But you get beat up by somebody that's been there 30 years. Like, that's not how this is done. And you're like, oh, okay, I guess I just don't know what I don't know. And next thing, 10 years in, you're beating the common sense out of some young engineer.

• 27:28David Weisburd?

  So everybody starts with this beginner's mind. They learn from an engineering principles. You almost start from first principles. Then you get infected through the cynicism for long enough that you have all your first principles thinking beaten out of yourself. And now your ego doesn't want to be the only miserable person. Now you start beating this-

• 27:49Speaker 01

  You're beating it out of the next person. And it feeds-

• 27:51David Weisburd?

  And the cycle continues.

• 27:53Speaker 01

  Yeah, sometimes I get asked, like, is there anything that we are doing that fundamentally, like, a Raytheon, a Lockheed, a Northrop, or Boeing couldn't copy and go do tomorrow? Absolutely not. They could go do it tomorrow. But the culture-

• 28:13Speaker 01

  Of the systems engineering process and just what is considered the right way is not something that I have any fear that they're going to just immediately change and then become competitive with the way that we're doing business because that culture is so ingrained and so hard to change. And once it takes root, it's just almost impossible to kind of steer that ship. And that's the differentiation between kind of us in a traditional prime. And so I got asked at a recent conference, like, would you ever partner with a prime? And I absolutely would if I thought they wouldn't impose on me a

• 28:56Speaker 01

  process that absolutely kills the advantage of our company. And typically, they're going to want to impose their process on us to make them feel comfortable. And we can't do that. And it's just like an impedance mismatch. So.

• 29:12David Weisburd?

  I think that's the biggest misconception about the innovator's dilemma. The innovator's dilemma was this concept by Clayton Crenshawson, which is that the previous generation of something disrupts the next one. Kodak was disrupted by the digital camera. Blockbuster was disrupted by Netflix. And everybody always thinks it's a technological issue, but it's not a technological issue. It's a cultural issue.

• 29:31Speaker 01

  It's a cultural issue.

• 29:32David Weisburd?

  And so much so that you've actually now the top founders and CEOs on Silicon Valley have innovator's dilemma on their bookshelf and they're aware of it. They're like self-aware of this bias. And now they continuously disrupt themselves. The latest example being Google deciding to embrace AI while disrupting their search business.

• 29:51Speaker 01

  And there was a period of time, at least for me personally, I didn't know if Google could do it.

• 29:56David Weisburd?

  And the founders had to step in. It's the reason why the founder has this role that can't be replaced by CEO. Even the smartest CEO in the world can't replace this alignment with both the equity and the long-term vision of the business.

• 30:08Speaker 01

  Yeah, you have to come in full Feldner mode and just be willing to rip things up in a way that would be too scary if you were wrong, if you were, I'd say, kind of a.

• 30:18David Weisburd?

  You have to bet. You have, literally have to bet the farm.

• 30:20Speaker 01

  You have to bet the farm.

• 30:21David Weisburd?

  No hedging.

• 30:22Speaker 01

  Correct. And I mean, Elon is tremendous at doing that. It feels like no matter how many times you're at the roulette table betting on black and it lands on black and you're like, holy crap. And then he's like, let's do it again. And you're just like, oh my God. But at the same time, it's so important that you continue to have the behaviors that made you successful. And the tendency over time, as you get to be a larger business and you put more risk management processes in place is that you actually kill the thing that made you successful. And then, once the culture shifts over a long period of time to be risk adverse, you're incapable of making the change, even if you can observe

• 31:07Speaker 01

  that it's a problem and it's there. And I spend a lot of time thinking about, as a startup, you start with a position of like, we have literally no process at all. And that's probably not optimal because you make mistakes on not having that process. But what's the right amount of process? What's too much? And it takes quite a bit of effort to try to control, like, yeah, this is necessary and we should implement this, even though it might slow us down a little bit, put this process in place to like, control quality or control whatever. And then recognizing like, hey, this is too much. We've gone too

• 31:49Speaker 01

  far and we need to come back to some happy medium. And I worry about that a lot, actually.

• 31:57David Weisburd?

  There's also this often underrated aspect about fast processes and slow processes. Jeff Bezos famously uses this analogy of the revolving door. Certain decisions are revolving doors and certain are not. And as you progress and evolve as a CEO, you start to learn that some processes should be slow, like hiring. You can't speed run it. You can't say, that was such an excellent interview, let's close them, let's get him before he joins another company. You have to spend time with him or her over several months trying to de-risk that decision because it's one of these irreversible doors. And other things like a product component, you just go and ship. And learning how to dial the knobs on fast processes and slow processes is one of the most underrated aspects, I think, of being a CEO.

• 32:41Speaker 01

  It's very difficult because fundamentally, you're in charge of what is happening and you're in charge of what is not happening. And.

• 32:50David Weisburd?

  The opportunity cost as well.

• 32:51Speaker 01

  And the opportunity cost. And one of the metrics that I always try to keep in mind is like, well, what's the consequence? If this was the wrong decision, what's the consequence? Is it something that's like, okay, that was embarrassing? Or is it company ending? Obviously, you have to be quite thoughtful as the consequence increases. But at the same time, if there is no real consequence, you got to get people like moving, like, let's just make a decision. I don't know what the right decision is either, but we're going to make one and move on because we can just reverse it and go the other way once we figure this out. And there's a natural tendency for folks not to actually think through how

• 33:32Speaker 01

  reversible is this decision? What's the consequence of it? And to your point of hiring, again, I just truly hate the interview process. It's kind of opposite all employment laws, but I would literally hire everyone and put them on a three-month probation and then just keep everybody that people liked working with at the end of three months. Unfortunately, that's probably not easily implemented.

• 33:59David Weisburd?

  Do you know how many interview processes Sergey Brin and Larry Page were part of before they stopped being part of each one?

• 34:04Speaker 01

  I have no idea for them.

• 34:06David Weisburd?

  The first 20,000 hires, they signed off on every single process. Doesn't mean that they were the first interviewer, the second or third, but they personally signed off on every one of the first 20. They intuitively grasp how important and how upstream hiring was of everything. There was no bigger and better use of time.

• 34:27Speaker 01

  I was just like, that's incredible. We don't do that. I don't do that because I'm on the road so much that if I had to interview absolutely everyone, I would actually slow down our hiring, which would be a strategic mistake. Although I do involve myself in any consequential hire, or at least meet the person, talk to them. But my favorite method is like, keep tabs on folks. Once they're here, talk to like their peers and see how it's working. And sometimes it's not like you need to be exited from the company, but maybe it's not the right role and that needs to be adjusted. It's always trying to find like, what's the strength of a person? How do you help them succeed?

• 35:08David Weisburd?

  I've had a couple CEOs say that they know almost within two weeks whether they made the right hire. Is that something that you see?

• 35:16Speaker 01

  In most cases, you can tell immediately.

• 35:20David Weisburd?

  Less than two weeks.

• 35:21Speaker 01

  In less than two weeks, you can tell immediately. Like you have, sometimes you make a decision and like, you're really just trying to hope for the best, even though you feel deep down like, this is bugging me. And I, like, one of the things that I've learned is like, if I feel that, it's like, this is wrong. Don't do it. Because it almost always turns out to be, yeah, that was an actual problem that I was picking up. I just didn't listen to.

• 35:49David Weisburd?

  Go do a sauna session, take a shower, eat some food.

• 35:52Speaker 01

  Yeah. Sometimes the pain is so acute to like fill some specific need. You're like, you just really want the help. But we are on the side of just continue to eat the pain with a smaller team until you can get the right person. But that is painful at times.

• 36:08David Weisburd?

  One of the most interesting models around this, Justin Khan, when he started his first company after Twitch, his very first hire was a recruiter, which is basically, how do we lower the error rate on hiring other people? I thought that was one of the most interesting first principles ideas. Have you ever thought about integrating recruiters and very elite HR people into your process?

• 36:30Speaker 01

  Yes. We've built up an extremely strong recruiting team. In fact, it's kind of funny that you said that it's, we identified hiring is probably our top priority. I think it was roughly six months into the company. And I was actually frustrated that I didn't bring that even earlier. Obviously, as you start a new business, there's just like, everything is on fire and there's nothing in place. But it became apparent extremely early that one of the most critical hires was to put in place good recruiting talent and to make sure, even folks that you don't hire, at least had a good experience with your company, which.

• 37:12David Weisburd?

  The reputation in the market.

• 37:14Speaker 01

  Like, don't leave people wondering what happened, et cetera. And it, very basic things, but it's actually difficult to do because you're outnumbered early on in a significant way. And so you have no processes and you have no tools that you're currently using. So it's easy to like, drop things unless you have a very particular mindset of a recruiter to just be very meticulous. And so, right now, we've actually taken one of our first engineers at the company, who was quite proficient at hiring with, I mean, he really liked the hiring process and what to ask and what to look for. And so now he

• 37:57Speaker 01

  runs all of our people operations, including recruiting. And I think that's tremendous because he also has a deep understanding of the engineering at the company and what we need and.

• 38:10Speaker 01

  That's super helpful when trying to look for the right candidate.

• 38:13David Weisburd?

  There is a bit of a disconnect in that every CEO will say that recruiting and hiring is the most important function and then they don't necessarily recruit their top people to run those functions. So there's this disconnect between what is truly important and what are we value internally? What do we show that we value?

• 38:29Speaker 01

  We regularly do a push where we just have our engineering team, like there's so much work to be done. Everybody's busy, but it's like, please take an hour out of your day to look at resumes or to think about what hires you need to make and write it down. So we could at least get like a job rec out. And I think it's very easy when things are extremely busy to let that, it's always like priority two or three. It's usually never priority one because it's not immediate. But you have to make that prioritization of like, let's take a timeout and actually review resumes that we have and try to get the right person in the role because it's the only way like we're going to dig ourselves out of this hole.

• 39:10David Weisburd?

  It's almost the universal advice you could give anyone. CEO comes to me and they say, I'm really exhausted. I'm really overworked. What do I do? And I say, I have good news and I have bad news for you. The bad news is I'm going to ask you to work harder for the next four to six weeks. I'm going to ask you to go out and add another task, which is finding somebody and helping. The good news is after six weeks that you're going to significantly offload a lot of tasks. So

• 39:32Speaker 01

  Exactly.

• 39:32David Weisburd?

  The paradox is that the more busy you are, the more you need to become more busy in order to

• 39:36Speaker 01

  To get out of it.

• 39:37David Weisburd?

  Offload that task.

• 39:38Speaker 01

  It's 100% true because until you fix it, it's just going to continue to get worse.

• 39:43David Weisburd?

  Speaking of things getting worse, your first year at Castelion, it was nine days to probably today. Tell me about the founding story.

• 39:54Speaker 01

  Well, really the concept of the company was, again, in that kind of customer oriented role. I had folks asking me while I was at SpaceX, like would SpaceX develop weapons for the country? And my personal background, I mean, I love the mission of SpaceX of getting to Mars, but of course, like my work there was always defense focused. And so my passion's always been defense of this country and keeping America strong. It needed to be fixed because of how bad the situation had degraded over the entire time I had been working in aerospace

• 40:37Speaker 01

  and defense. Cause this kind of message that the country needs to do better at manufacturing these types of systems had been said at every conference I'd ever attended since like I started my career in 2002. But then nobody was doing it. And so I finally was at a point in my career where I felt I had the requisite skills to tackle that type of problem. And so started SpaceX, you best twice a year, May 15th, November 15th. And so around those dates, everybody talks like, what are you doing? Are you going to do another six months?

• 41:18Speaker 01

  Are you going to do something else? And of course, like when people would ask me what I was doing, I told them. And so ultimately, Andrew and Sean, who were at the time both planning to leave themselves for different reasons and go do other completely unrelated things, decided that they would join me on this concept of building back American defense manufacturing and specifically starting with a affordable hypersonic weapon, which of course, like our customers had been directly asking for and very apparent the need. It

• 42:02Speaker 01

  was still just so critically important to me to attempt to plug the hole. It was worth the risk. And certainly back in 2022, when we started the business, hardware wasn't something that people wanted to invest in. And then when you added, oh, and I'm just going to do defense hardware. I mean, that was described to me by an investor is in the realm of the uninvestable.

• 42:32David Weisburd?

  He drew that two by two matrix and he said, this is

• 42:34Speaker 01

  Yeah. There was commercial defense, software, hardware. And he's like, you're on the quadrant of defense hardware and not even kind of like in the middle of the quadrant. You're on like the very corner. He's like, that's like the realm of impossibility. That was the end of the conversation, but that stuck with me. And I was like, but that's what's important. That's what

• 42:54David Weisburd?

  This is by the way, why most VCs don't give feedback.

• 42:57Speaker 01

  That's true. And I'm not saying a name. I frankly don't even remember the name. We had a lot of early rejections and quite frankly, when that winter of 22, I was like, well, maybe I just left a great job, which I'll never be able to get back because there were definitely some hurt feelings upon leaving to become unemployed. So certainly wasn't a great time. My wife was expecting our second child and it was like, okay, I guess we'll start selling the 401k to survive. And I certainly, there was a period there where it was going many months without a positive.

• 43:40David Weisburd?

  Close to 100 rejections?

• 43:42Speaker 01

  Yeah.

• 43:42David Weisburd?

  100 VC rejections?

• 43:44Speaker 01

  Sometimes multiple per day.

• 43:45David Weisburd?

  Who was the first person that said yes?

• 43:47Speaker 01

  David Byrd with Blue Yard was our first to give us our term sheet. And Blue Yard's kind of investment thesis was very different than I'd say a lot of the folks we talked to. David had recognized very early on, like the type of business he was trying to find because he thought that would be important to the future and would be a very good investment. And then we came along and we just happened to fit exactly what he was looking for. So quite frankly, like he was an instant term sheet. Our first money in the door, actually lately, let me correct the statement. That was our first term sheet, but our first like money in the door was Jonathan Lacoste with Space VC.

• 44:26David Weisburd?

  Good friend of the pod.

• 44:27Speaker 01

  Yeah. Who did put in the first funds into a safe note at 14 million cap.

• 44:35David Weisburd?

  So he's doing pretty well.

• 44:37Speaker 01

  Yeah. Yeah. So if you need to borrow any money, just talk to Jonathan.

• 44:40David Weisburd?

  How much of a special relationship do you have with those early investors?

• 44:44Speaker 01

  I'd say we're very close because they were fed on us when it wasn't cool. It was very uncertain. And I think that, again, like can't buy a relationship when you need it, like during a good time. And like, they came in and helped us at what I would say is a very difficult period of our kind of founding history. And for that, we'll be forever grateful. So like, if they want to invite me to an LP day to help them raise their next fund, I'm there regardless of how terribly inconvenient it may be. But I feel like that's the relationship you have with a good early investor is like, you help

• 45:29Speaker 01

  me, I will help you.

• 45:31David Weisburd?

  Fast forward to today. Now, certainly there's a lot of top investors that want to invest into you. When did that change? What was the turning point from you being virtually unfundable to now being massively oversubscribed?

• 45:49Speaker 01

  2024 was probably a turning point, maybe late 23, things started to shift. Certainly, Andreessen invested in us in late, I think it was like August, September of 2023. And once they invested, folks that had, I would say, passed over us, really started to come back and take another look, like what did we miss? The entire market was evolving too because now hard, like people were starting to invest in hardware. People were starting to realize, especially with Andrew's success, that you can make money in defense. And so

• 46:33Speaker 01

  I would say there was a market shift, but then of course, having like a fantastic tier one capital partner, putting money in got people's attention. And that really helped open doors for us in kind of interesting ways. Like, for example, we've always leveraged like debt as an instrument to help fund the business without always having to sell equity. And when we started the business, we couldn't even open a bank account. It was very difficult to find a bank that would take a defense company.

• 47:05David Weisburd?

  Almost like crypto.

• 47:08Speaker 01

  Well, worse than crypto. There were plenty of banks that are like, oh, you're actually in gambling and porn territory. And I was like, come again? Like what? And so we had a lot of folks that wouldn't even allow us to bank with them.

• 47:24David Weisburd?

  How did you solve that?

• 47:26Speaker 01

  We finally found two banks that said yes. And because well, one of my first questions was like, well, who the hell does like Raytheon or Lockheed or somebody else bank with? Like, how is that any different? So Chase said yes, and Silicon Valley Bank said yes. And so we literally took our $4 million seed round in early 23 and put $2 million into Chase and $2 million into SVB. And within the next week, SVB.

• 47:56Speaker 01

  Lost her money and it was up all night reading about what happens to a bank in receivership and how quickly we could get our money back, which was not a good night read because it was very clear it could take quite some time. And so we had lost half of our seed round, literally within a week. Luckily, we got it back quickly thanks to the country bailing out Silicon Valley Bank. But that was quite an eye opener. But once we had Andreessen as an investor, literally like we were struggling to raise any debt previous to that. Andreessen called banks on our behalf and we had like three or four offers within a

• 48:37Speaker 01

  day, just based upon the strength of our investor.

• 48:43David Weisburd?

  Times have changed since those early days, but you're still going after a massive market with significant challenges. What's your biggest bottleneck today?

• 48:54Speaker 01

  Labor, talent. We are very selective on who we hire and hiring for that right culture fit that we talked about. And it's very hard to get the right folks into the right roles. It's just a kind of a slow, painful process. It consumes a lot of time internally, but it's like, as we discussed, it's critically important. But fundamentally, I think we're gated by just our capacity to do work. And that's saying something because I think like on average, probably people work 65 hour weeks on average, with some working significantly more. So you're, some folks work like two years every year kind of work ethic. So you get a lot of output from the folks that are there and I'd say we're very well

• 49:38Speaker 01

  aligned, but we are always trying to get more great people into the company.

• 49:43David Weisburd?

  Tell me about Project Ranger.

• 49:45Speaker 01

  Project Ranger, the name came from the Apollo era program to scout out sites for where the landings would happen. And so for us, like the big culminating event is get Blackbeard into production. Where is that production going to happen? When you're talking about a explosives and energetics facility, it's significantly challenging to find where to put that and who will, who?

• 50:16David Weisburd?

  There's a NIMBY aspect. No one wants it in the background.

• 50:18Speaker 01

  Exactly. And so Project Ranger was the code name that we use for kind of doing the first site selection for our production facility. And so ultimately that facility is located in Rio Rancho, New Mexico, right outside of Albuquerque. It's a thousand acre site dedicated to hypersonic weapon production. I believe that makes it the largest dedicated site for hypersonics manufacturing in the world. It's 21 buildings at its current stage. Many of them are up and expectation is that we can produce 2,000 Blackbeard weapons per shift.

• 51:02Speaker 01

  We can run three shifts a day with the current facilities. We have room for expansion. And on top of that, we are developing a much longer range weapon system.

• 51:17Speaker 01

  That I'll talk about vaguely because we haven't announced the name of it yet, that will be produced at the Project Ranger site as well. And so while maintaining that 2,000 per shift of Blackbeard, so the site is sized to produce both concurrently. And so that effort, we started, we broke around in November of 25. And for a facility of that size, it's quite incredible how fast it's nearly completed. It should be complete from like a building and facilities perspective in September. And then we are moving all of the equipment into the buildings in the kind of final months

• 51:59Speaker 01

  of the year and intending to stand up initial production trials beginning of 2027.

• 52:06David Weisburd?

  And that's downstream of choosing the right jurisdiction.

• 52:09Speaker 01

  Yes. We started that process, like a nationwide selection process for where to locate that. The economic development team and Sandoval County in New Mexico were phenomenal. The governor's office, the thing that you're looking for, and I think is important for folks to understand when you do these large infrastructure projects, what you need is not regulatory certainty in the sense of like, you must just rubber stamp any requests we have, but what you really need is certainty on the timelines to receive permits, knowing who it is you need to talk to, knowing what permits you need.

• 52:52Speaker 01

  Like it's so complex for a project of this scope of how many things have to happen. It's super helpful when the locality facilitates introducing you to the right folks to get the meetings to happen, to drive a decision from the government organizations that make the call on some permit, just to get them to act. And what we found was New Mexico was the right fit for that attitude. And we were able to select that site and move into construction in under a year, which was, I mean, it's incredible. And now we're going to build the site and have it complete in a year.

• 53:35Speaker 01

  So that kind of scale in terms of speed for a construction project of that scale is just truly phenomenal that you can still do that because that's one of the things that as a country, we've really lost is that ability to get things done quickly. And in large part, it's not due to the actual construction or other activities. It's due to more mundane things like the permitting, the regulatory, can I get enough power to the site, et cetera, and get the permits to run power lines to this? I mean, it's just like, there's so many things. Can I get water? That can be very basic, but are not simple.

• 54:11David Weisburd?

  Hard to solve if you haven't thought of them ahead of time.

• 54:16David Weisburd?

  I know you're very ambitious, but can companies like Castelion and Androl truly become the next primes or are they always going to be second to the Raytheon's and the lock keys of the world?

• 54:29Speaker 01

  No, they'll, will absolutely become.

• 54:32David Weisburd?

  What needs to happen? What are the main risks?

• 54:34Speaker 01

  From a Androl and Castelion perspective, we're over the initial hurdles of will we be capitalized enough to make it? We have the capital to make it. And so now the risk shifts to execution risk. The thing that really seals kind of the deal for the government is to do the thing you said you would do, which in kind of both companies cases, I would say is really to enter production and to start fielding systems at scale that solve some of the country's critical national security challenges. Androl's obviously an older company than us and they're further ahead of us.

• 55:16Speaker 01

  But if you do that and you enter production and you do that well, the success is all but ours for the taking. And I think that it's like quite critical that at the same time, we also solve some of the issues with the speed to development, the cost of these systems and producing enough of quantity that it actually achieves the desired effect, which is deterrence. And if you can solve that, you're absolutely in a very strong competitive position with like the traditional primes.

• 55:46David Weisburd?

  And the Department of War outlined Castelion specifically in mid-range missiles. They named you guys as their go-to supplier for that.

• 55:55Speaker 01

  Our rationale for why we developed Blackbeard was looking at what are the capabilities that you would need in a weapon system to deter a conflict in the 2027 and beyond timeframe? That weapon was not simple. At least at the time, we were the only kind of new entrant startup to even consider doing something of that complexity. But in my mind, it was the right thing to go build. And it obviously would have been nicer to do something different. The Navy is one of our earliest customers for us, which was to put this weapon on F-18, a carrier-based F-18.

• 56:37Speaker 01

  And I can assure you that the first weapon you build, you would desire that not to be an air launched platform off of a carrier. The only thing that could possibly be harder would be a submarine launched weapon system, but it was the right thing to do and will have the impact. And because we are doing the right thing, it will pay off in the end, even if there are delays in the development, just due to the complexity of what we're trying to do. And then since that time, the army became interested and it would have been very easy to develop a different weapon system for ground launch, which is typically what happens. Like you don't typically use the same exact weapon.

• 57:23Speaker 01

  But we made a very concerted effort to maintain commonality of the weapon itself so that we could take advantage of the cost and kind of capacity and scale that that would enable in manufacturing, but to solve kind of both use cases. And that adds another layer of complexity that even like the large primes don't take on.

• 57:48Speaker 01

  But I think again, at the end, if we're successful in doing that, and I am very confident that we will be, that will pay off tremendously and our ability to produce at an even greater quantity and a lower cost.

• 58:01David Weisburd?

  Well, Brian, I was born in a communist country in Russia. My family came here as refugees. My grandparents are all from Ukraine. They were stuck between the communists in Russia and the Nazis in Germany. And what you're working at at Castelion and defending the American way and American values is deeply important. I really respect what you're doing, so really appreciate your time.

• 58:25Speaker 01

  Thank you. I really appreciate you saying that. And I do hope that we can make a difference.