Not Boring Founders: Crusoe Energy (Chase Lochmiller and Cully Cavness)

27:36Like generally things that drive large sources of emissions we benefit from in one way, shape, or form. Steel production drives a lot of greenhouse gas emissions, but it also gives us steel, which gives us skyscrapers and all these useful products that we use from steel. In the case of flaring, it's literally a negative externality. It's like a tax on everyone in the world because it's leading to tremendous amounts of greenhouse gas emissions, a warming climate that was leading to higher probabilities of big storms, fires, all of these natural disasters that are creating more uninhabitable climate for everyone.

28:10And it's this hidden silent tax on everyone. Our goal as a business, and mentioned, well, I'm an environmentalist. I'm also a libertarian capitalist, was like, this actually is an opportunity. Because this is useful in one situation, but not this situation. If we could figure out a situation that we could make it useful, that market is enormous. Being able to sort of tap into essentially a free energy resource, who wouldn't want that when your primary cost driver is energy? That was really the catalyst of the business. And our goal always was to enable many different computing applications because seeing these things play out between digital currencies being developed and really requiring a lot of computing power and the transformations taking place in AI, I really felt like the biggest growth sector for computing was really going to come, or the biggest growth sector for data centers was really going to come from these more energy intensive computing applications, whether it's proof of work blockchains or things for high performance computing or scientific computing or artificial intelligence research that really just required tremendous amounts of power consumption.

29:16Yeah, I was having a conversation at dinner last night with somebody in finance who works out of an office that can kind of do anything. And so we were talking about these arbitrage opportunities that exist, but it's not like Like arbitrage opportunities that you like wish would exist where like there's just these mispricings on different markets or whatever. Like you need to do these heavy physical things. But if you can do these heavy things, look around you everywhere, then there are arbitrage opportunities that can be taken advantage of. This seems like a pretty huge one that you'd pay a certain amount of money for that same energy somewhere, but you can just put the data center there, get the energy for a much lower cost and then turn it into something useful.

29:59How does that work with the oil and gas company? Are they paying you to take the energy? Are you paying them? What does that look like? We have a pretty standard deal at this point where we'll pay for the natural gas. We're typically not paying as much as a pipeline would pay if it was at a pipeline interconnect point. But you understand why now? I think because we're in a place that doesn't have access to a pipeline and we're bringing out a lot of very capital intensive equipment to that site to consume that energy. We're bringing out power generation systems, modular data centers, and we own and operate all the compute within the data center too.

30:31So these chips are not cheap. I mean, anybody who's opened the Wall Street Journal recently can sort of tell that there's a lot of capex running to the computing asset as well. But we don't charge them anything. It's a free solution for their clearing. They're making some revenue on the gas to the extent we're generating carbon credits, which in some cases we do do that as well. We'll work out a relationship where they're getting some of the carbon credit economics. and then we managed the computing asset and sort of monetized the computing asset. We're effectively a gas buyer. It's effectively a gas sales agreement like they would have with a pipeline company, but different terms and conditions and a different end customer on the other side of it, basically.

31:14I actually love the way you framed it as essentially an arbitrage strategy. In my previous life, we were building essentially statistical arbitrage strategies. cheese. I think this is certainly another manifestation of something that's valuable in one case, but not in another. How do I move it from that unvaluable case to the valuable case, take ownership, and then move it from the unvaluable case to the valuable case? That is the capturing the arbitrage. And that's really what we're doing in terms of unlocking value with the stranded energy with computing. And there's all these beneficial side effects in terms of the environmental impact and reducing methane emissions, creating a revenue stream from an otherwise wasted resource.

31:55And in our mind, really being able to offer a sustainably cheaper cost of computing to the marketplace by harvesting this otherwise wasted product. Yeah, it's beautiful when you think about it and see it now in action. I'd love to go back to the early days of the company because this is one of those ideas when it's up and working and when GPU demand is through the roof and when everybody needs your product, it seems obvious. What were those first conversations with, let's start with investors, with VCs, like where you're saying, look, we want to do this capex intensive thing. We're doing Bitcoin.

32:33We're doing this piece of the climate that like climate people probably don't even really want to touch. Will you give us money to go build this business? Yeah, it's funny. So I was actually our first investor. I really sort of viewed it through this lens of like, well, if I invest this money and we spend money on CapEx and build this thing, I'll probably be able to get my money out because we sort of have this strategic cost advantage around the energy. And we're in a worst case scenario. Hopefully I'll recover some meaningful percentage of the capital I invest. But you're right. It's controversial in certain ways because certainly during that era, when we started the company in 2018, going into 2019 was when we went to go raise our first external capital.

33:20And the climate at the time was, there was a lot of focus around ESG. Oil companies were the devil. And especially from sort of the venture technology community within Silicon Valley. And the fact that we were working with oil companies felt uncomfortable for certain investors. There was a very blue chip focused VC, or a blue chip VC that early on wouldn't even take the conversation with us because we touched oil. The fact that we worked with oil at all was like a black mark. That investor has since come back to me when we were raising our Series C and said, Chase, look, I really actually want to apologize because I didn't even give you the time of day during that early phase.

34:06And I didn't really understand your business in terms of recognizing that you're actually reducing emissions from the oil production process. And I think when you reflect on the events, the geopolitical events that have happened over the last 18 months, the war in Ukraine, for all the horrible things that have resulted from it, I think one thing that's beneficial from this war is really it's shown people the importance of energy in the world and the importance of oil specifically. And really people recognize that the reality is that the world still really runs on oil. We require a lot of oil to maintain existing quality of life.

34:44And a lot of this gets taken for granted by many, many broad scale consumers from the materials that go into the clothing we wear, the packaging, the transportation, the ease of being able to get on a plane and go on a weekend trip somewhere across the country. I mean, all of these things that we sort of take for granted as part of the modern existence would not fundamentally be possible at all without oil. We sort of took this very pragmatic approach of wanting to recognize that oil is important and it is important to modern existence, but it does have a cost associated with it. And if we could find a way to actually minimize that environmental impact by getting rid of this negative externality and this problem associated with flaring, it could be a net positive for everyone.

35:30But we did. So early on in those investor conversations, I think there were certainly, it was a polarizing topic in terms of, okay, it's not a pure software business. We're not a SaaS company for whatever business enterprise vertical that hadn't been tackled yet. It was a capital intensive business that touched oil and gas. And it took folks that really understood and were willing to do the deep work around what was actually happening to want to take a bet on Crusoe. And we were fortunate to have some great early supporters that were willing to really think outside the box and take that risk on us.

36:06A handful of angels that were very, very helpful in the process, like Lee Jacobs, who runs a fund called Long Journey Ventures at this point. And the leads of the round were Founders Fund, who notoriously is willing to invest in contrarian or counter-positioned concepts. And I think this was a great, great fit for them. And we're very, very thankful for their early support in terms of taking a risk and really recognizing this opportunity. And they co-led the round with Bain Capital Ventures. We worked with a partner that historically hadn't done things like this, but was recognized really that arbitrage opportunity that we sort of discussed, but that partner's named Salil Deshpande.

36:45And he worked closely with a guy who was a principal at the time and now has a partner, Stephan Cohen. And the two of them really were willing to do the work and recognize the opportunity that we were building towards and make a bet on us and really take risk with us. And we were very just grateful and thankful to those early supporters because there's a lot of ideas out there. Ideas can be cheap at times and it requires both a great idea with capital and a relentless focus on execution. And that's really what's got us to where we're at today. I wanted to just share a story about that early fundraise that I think is kind of fun.

37:21So like Chase mentioned, he put up the Capitol to build our first prototype. It was like a 20-foot long, basically shipping container-sized data center, 264 servers, ASICs in it. And we set up satellite internet connections. In the middle of this, we worked out a deal with a private oil and gas producer in Northeast Wyoming, very remote spot. It was like a multi-hour drive from Denver up to northeastern Wyoming and then another hour and a half plus on a dirt road from a town called Lusk, Wyoming. You'd kind of go west from there. How did you find this guy? It was through kind of relationships in the oil and gas business.

38:00And actually, someone who works and has worked on our team since the very beginning, Andrew Likens, was somebody that put me in touch with this company before Andrew even worked for us. He was just kind of like hearing about this crazy idea and said, hey, maybe this could help. I heard they have a flair. and anyway we worked out a deal with this producer single well in the northeast very remote Wyoming and there are these it almost looked like old home videos now where we had our iPhone and we're filming this setup there's sort of like reddit power generation systems it looks like it might be held together with duct tape and bailing wire and a lot of the like plumbing work that we did with the piping was we're still learning a lot of stuff right so you can kind of actually layer that look.

38:42And we're all out there leaking our hard hats and fireproof clothes and filming this thing. And we fire it up and all 264 servers start blinking green. And then over the satellite internet connection, there's no cell service. We're way out there. We're like linked into our wallet and we're seeing Bitcoin accumulating in this wallet. And we basically shared this video with the venture capitalists. Here's a ball of fire. We pushed this button, it got smaller and then over here there's like bitcoin coming it's a book this is the business model yeah once they saw it then the seed round came i think until you can actually until chase took the risk to put up the money to make that real and a picture's worth a thousand words concept it was just too abstract it sounded totally insane i'm sure and then we showed them it really could work and okay here's a seed check we took now we did more bigger scale new clients showed that we can operate a little bit.

39:41Here's a series A. We went bigger, got bigger clients, series Bs. Now we've raised over half a billion dollars of equity and we're operating something approaching 200 megawatts of data centers powered by stranded, wasted, clean energy source. It's really been an amazing journey in five years from that sort of selfie video in the Wyoming whale field to where we are today. It feels like a long way away. Yeah, two other things that I'll just add there. One is that Lee, when we showed him this video, Lee Jacobs, he coined this phrase. He was sort of putting our business as succinctly as possible. He's like, oh, your business is gas to cash, gas to cash.

40:20And that was like his catchphrase around like his investment in Crusoe, which was funny. And then the second thing I'll say is that when we were driving out to this area of Wyoming, like Coley and I drove up from Denver, this area, he said it's outside Lusk. I mean, that's like Lusk being nearby is sort of a generous statement, but like driving through this area of Wyoming, like I didn't actually realize there were areas that were this remote, like in America. You're driving for hours without seeing a single building or car. And you're just like, what? Like no one lives. There's nothing out here.

40:54It's just like a very empty space. But it was a great place for us to build our first flare gas powered Bitcoin mining data center. That is an incredible story. And to think that some people just raise a seed around on a deck. I want to see more companies set up a remote data center in the middle of Wyoming to unlock that seed round. When we raised our seed, we had our first project and the unit economics weren't nearly as good as they are today because we hadn't done all of the optimizations and things that we've done to drive our overall costs and expand our margins over the course of time. But at the time we were raising our seed round from this single location, we had, I think it was about 250K in annualized revenue from the single location and a payback on the CapEx targeting in the sub two year range.

41:48So it was kind of a, it was a good setup and it was sort of like, hey, look, we can make money with this. And you also have this call option on Bitcoin in terms of like if Bitcoin actually becomes more widely adopted, this is going to become even more valuable. And that was what ended up playing out. But it didn't necessarily have to be that case in order for us to be successful. It was kind of like creating these great unit economics based on tapping into this waste energy supply. Yeah, I was going to ask about the unit economics and just about the kind of capital structure of the business generally and how you've thought about financing it.

42:21For everybody listening, you should go listen to the ACQ2 podcast that Chase did with the guys over at Acquired. And you talk about the different ways that you finance the different pieces of the business. Getting really tactical on just, again, like early stage as more companies are going out and tackling these big capital intensive problems. and I want to back a bunch of them and I hope we see a lot more of them. Like how much of that did you have to have planned out at the beginning? How much did you have to convince investors that you had ways to fund the growth of the business? It wasn't just equity for things that maybe shouldn't have been financed with equity.

42:56Like how much of the plan has to be baked in upfront when you are going to be doing something so capital intensive? At the seed stage, there was a lot of like, well, we could probably get debt on these things. We're buying generators. Probably someone's going to give us some debt on that. I don't know. We really didn't have a great plan at the time. But what's interesting about being an entrepreneur and kind of building a company is that you're inspired to build a company to solve problems that you feel like need to be addressed. I mean, at the end of the day, that's really what's driven Coley and myself is that we want to create environmentally aligned computing solutions.

43:32We're inspired about just sort of solving these energy infrastructure challenges, enabling the future of computing to drive innovation and transform the world and the economy. But when you look at sort of the history of venture capital, I think like there's a bit of a recency bias that people have in like the post-internet area and like the Web2 era where it's like the best thing to fund was all of these software startups that have high gross margins when they reach scale and don't require a lot of capex, just require a lot of OPEX in terms of GNA to hire highly talented people. And that's like the primary expense line item.

44:12When you look at actually what I feel like are the most important problems, probably of our generation, it's they're going to be around climate. Like how are we actually going to solve these problems to make the world a livable place for generations? I mean, I have two young kids. Koli has one kid and another one on the way. You have kids. I think we're all sort of thinking about it through this lens of being parents and just saying like, okay, what is the future going to look like for our kids? And energy infrastructure problems are not SaaS problems, right? They require tons of capital to really deploy these things at scale.

44:46I mean, you're talking about like the infrastructure the world runs on. When I think through that, I think there's emerging a lot of interesting ways to grow and fund businesses. And I think founders, particularly in the climate space are thinking through it through this lens of like, okay, I'm running a startup, but that doesn't mean I have to fund it indefinitely with equity. In fact, I probably shouldn't do that. There's like cheaper sources of capital that are going to be way cheaper than my equity cost of capital that I can leverage to basically bootstrap and grow my business in a meaningful way.

45:21And Kolei and I were very fortunate to early on, around the time of our Series A, we got the support from a credit fund that was willing to take this risk with us and think very uniquely about what are the assets they're actually underwriting and what are the revenue streams tied to those assets. And independent of the fact that we're this small startup with a small team and short operating history, no bank was going to take a bet on us, right? No traditional bank, no traditional lender is going to take a bet on us. But we're fortunate that this this creative credit fund called Upper 90 was willing to work with us.

46:00And we actually raised a$40 million project financing facility with them early on that was in conjunction with our Series A. And really this enabled us to scale up the business and prove out that the unit economics would work well in a way that didn't cost us tremendous amounts of dilution. It really enabled us to grow our footprint, spend CapEx, prove out the business model. They earned a great return, especially when you think about it from a debt perspective. The way it was structured was it had equity characteristics and debt characteristics, so the structured project financing facility. But it worked really well for us because it was way, way, way cheaper than equity capital and really, really enabled us to scale.

46:43When we think about it today, I think we've sort of achieved a much larger scale operation. And we have different buckets of CapEx that are really mostly tied to various revenue streams between our power generation equipment, our mining hardware for our Bitcoin business, our high performance computing servers that are mostly the CapEx cost is really primarily driven by the GPUs for our cloud business, and then our data center infrastructure. All four of these pieces of CapEx, we're able to we're sort of pushing this direction of basically having independent CapEx facilities that are tied to the asset.

47:24In the same way, when you buy a house or you buy a car, oftentimes you'll get a car loan or a mortgage and you're not putting up 100 % of the value to buy that house. There's a physical asset that you're actually able to get leverage from and someone's willing to lend against as collateral. So if you default on your car payments, you end up just giving the keys over to the lender. So I think that's a great structure for us kind of going forward in terms of leveraging our equity capital stack to continue to build out the technology and continue to sort of invest in future product development and then leveraging asset-backed financing facilities to really scale the physical assets and operations of the business.

48:07I would just make one other comment about the fundraising and finance side of this, which is, I think just more relevant for like climate technology and hardware and capital intensive businesses in general, there is an interesting interaction between the like preferred equity model and a company that has been able to accumulate an asset base, like a hard asset base, which is that the next investor in, if they're taking preferred equity, if you're investing into a SaaS company, you're essentially kind of, especially in the first few rounds, you're buying a lot of option value. There's not a lot of intrinsic inherent value in that equity.

48:46And you don't have a lot of downside protection on that option because it's really just the team and the idea. And if it doesn't work, there's not much there. If you're preferred equity in a company that has a meaningful amount of hard assets, it's actually quite well downside protected and you have the upside exposure of this option. And I think that for climate technology and companies in general, if you're building biogas things and carbon capture and sequestration things and flare mitigation things and all these hard asset-based things. There is some, in many cases, actual asset value there that would de-risk and give comfort to the next preferred equity investor such that they can provide equity into something that's capital intensive, knowing that in the worst case scenario, they become debt and they are sort of first dollars out.

49:36And actually, I think that the preferred equity model works really well for these hard asset-based, CapEx-based businesses for that next investor. Yeah, that makes a lot of sense. Switching gears to one topic that is also kind of investment related, it's founders-run related, the idea of being contrarian and right. It feels like within kind of the climate space, you guys have been proven to be a little bit contrarian and right on this idea. Kali, I'd love to hear your roommate from Middlebury or the other people that are in that community. How is their view shifted from the beginning to now, the broader kind of climate and environmental community?

50:14What do they think about the fact that you work with oil and gas now? Are people getting more rational about things or is it still oil and gas is icky? I think it's been really fascinating. People, it's like Chase said, since the Ukraine war, there used to be economics and environment. And that was the debate in oil and energy. And now there's this third term, which is security. And you hear it, especially when all that Russian interference and disruptions of energy supplies between Russia, Ukraine, and Europe was in the headlines. People got very smart and very realistic very quickly about the fact that the number one primary source of energy for humanity today is still oil.

50:52It's number one. And renewables are still a single digit percentage of total primary energy. They're intermittent. They are not baseload. And relative to hydrocarbons, they're still small. So we need this stuff. And that's a hopeful starting blind for the conversation. And the next conversation has to be, well, we need to produce it as cleanly as possible. And that's what we're doing. And I would say that the conversations with environmentalists, even like very hardcore environmentalists, which I'm friends with a lot of them. I think they recognize that reality and they like this idea of, okay, we're making incremental progress.

51:29We're marginally decarbonizing this thing that we still depend on with our lives and we probably will for the rest of my lifetime. So let's just produce it as cleanly as possible, given that that's the case and that's just real. I think that that is a reasonable conversation that most people can can't get on board with. It's not the perfect answer that we're all just living in a 100 % renewable powered world that I think a lot of people would like to believe is just running a corner. It just, it's not. And we have to deal with the messy truth. And that means let's produce these hydrocarbons as cleanly as possible with as little waste and certainly as little uncombusted methane emissions as possible.

52:08I can't speak for everybody and I can't speak for other people, but I think that I feel like we do connect on that message. And more often than not, somebody who would think of themselves as an environmentalist is going to be supportive of this idea because it creates that economic incentive to do the right thing, like Chase mentioned. Amen. We haven't even talked about what the product is today. Again, I think the acquired episode does a really, really good job of diving deep in there. But for people listening, what should they think about when they think about Crusoe today? What should they come to you for?

52:40And then what does Caruso look like over the next five to 10 years? So today, when customers come to us, the biggest set of customers that we have are coming from the AI landscape. So people in need of AI cloud computing resources, primarily GPUs. What we've built is a high-performance cloud computing platform that's focused on enabling AI workloads and scientific computing workloads in a way that's both cost-effective as well as environmentally friendly. And so today, that's probably our most primary product that we have a large amount of customers for. When you think about our business, it sort of sits at the center of like, we're solving a problem for energy companies while also delivering computing solutions to folks in need of that computing capacity.

53:33So in a lot of ways, we also view the energy companies, which are actually the supplier of the fuel to power the overall systems as customers as well. And their customers more of just actually were able to deliver to them a service of flare mitigation. Or in the case of the renewable energy companies we work with, they have oftentimes renewable energy production is asynchronous with actual demand. And you have this issue where a windmill may be in an area that has no demand for the power that's actually being produced. And as a result, it gets curtailed or priced at a negative price on the actual power sales a fairly significant percentage of the time.

54:13We can actually help solve that problem by bringing our computing load to the source of this otherwise stranded energy resource. So we view them as customers as well, kind of these broad energy company customers where we feel like we can actually help improve their unit economics by bringing a dense transmission-free energy consumer directly to this supply of energy that they're producing, whether it's waste or primary source of power. So that's broadly how you can think about our business today. I think in the future, our business really is driven by our mission of aligning the future of computing with the future of the climate.

54:55When we think about that, that's driven us to take this very energy first approach to building and offering computing infrastructure to the world, whether that's for digital currencies or high performance computing. Now, when you play out the trends that are unfolding right now with AI, the demand for computing and therefore the energy required to power that demand for computing is enormous. The CEO of DigitalBridge came out recently and said that his perspective and the company's perspective was that the demand for AI was going to drive an additional 38 gigawatts of data center infrastructure over the next 10 years, which is an astounding number.

55:38To put it in perspective, for all of the criticisms of the Bitcoin network, it roughly consumes 8 to 10 gigawatts. You're talking about something that's four to five times larger than the Bitcoin power consumption today, just an incremental demand coming from this one resource in AI. And so a lot of the forecast for the global power consumption of computing is really going to go from, it's around 1 % to 1.5 % of global power consumption today. And most forecasts have that going north of 10 % in the next 10 plus years. And with that, we think it's important that this computing infrastructure is built in a way that is harmonious with the long-term goals of the planet.

56:20And we think we can be a critical infrastructure player in terms of providing both cloud services that are done in a way where you don't have to actually own your own equipment. and we can deliver you a great service and a great platform to basically run your critical infrastructure on. As well as eventually, I think we'll probably contemplate various solutions for co-location or in partnership maybe with some other folks. But at the end of the day, taking this very vertically integrated approach where we can build, operate, and manage the data centers ourselves, that's for our own consumption and our own cloud platform.

56:56But at the end of the day, I think we can offer that excess capacity that we've built in this very highly optimized, high-performance capacity to other folks that actually want to own the physical equipment themselves. Told you this one-on-one, told you this one-on-one. I absolutely love this business. And I love the idea of just being able to get really excited about consuming a lot more energy. It's a really good thing. It's what drives us forward. And I think Crusoe is just like a perfect example of how we can do something that gets us more excited about consuming more energy and making the world better for people.

57:28So thanks both for joining me and for building this. I'm really a big fan. Thank you so much for having us, Paki. This was a fun conversation.