Let's go to space. Let's go to space data centers. Our first guest of the show is Philip from Starcloud. Burning up the timeline. Philip, how are you doing? Good to see you.
The current thing himself. Did this all kick off with Nvidia posting about you? Was that when the kerfuffle?
I think that was the catalyst. Yeah.
Thanks so much for having me on, by the way. I've been a huge fan since day one.
I am glad to have you. This is why the show exists. We address every timeline that is in turmoil. So break it down for me. When did you start the company? What's the pitch? And then we'll go into some of the hows. But just give me the recent backstory.
Yeah. So we started about a year and a half ago, at the beginning of last year. We basically in that time built a satellite. We've got the first launch coming up in a week, and that's why Nvidia just posted their first thing, because we're going to be launching the first H100 in space. So about 100 times more powerful GPU compute than has been in space before. And yeah, that's the sort of brief history.
What were you doing before this?
So I started my career as an engineer for five years. Studied applied math and theoretical physics undergrad and masters, and then I spent a few years at McKinsey working with the space agencies of the UAE and Saudi government. And then I founded and sold another company not related to this, and then moved my way into this.
Here you are. Harvard, Wharton, Columbia. He's been to every academic institution, collecting them all. What is the strategy for this first launch, this first satellite? Like what are you trying to figure out with this launch? What is the value of putting this H100 into space? And then kind of how do you think about the timeline? Because I think the overarching criticism of data centers in space has been like people, I don't think people genuinely believe that this will never happen. It's more of a debate around the timeline on which it happens.
Yeah, which is a fair debate to have, I think. So the first satellite, the reason we're launching this, lots of people believe that you can't run terrestrial data center grade GPUs in space because of the high radiation environment and you need to dissipate the heat in a vacuum. So this is really a demonstrator. It's a 60 kg small sat. It's launching on a Falcon 9, a quarter plate there. And so we'll be running H100, we'll do a whole bunch of firsts. We'll be the first to run H100 inference in space on imagery from working with various DoD customers. We'll announce that soon. We'll be the first to train a model in space. We're training MiniGPT from Cerebras. We'll be the first to do, we're running a version of Gemini called Gemma, which is like a cut-down version, working with Google Cloud on that. But so we'll do some fun demos, like for example, you can SSH into it and then just ask it things like where are you now and how does it feel to be a satellite, and it will give you a kind of coherent answer. But yeah, this is really, really a demo. The next satellite launching in October next year is 10 times the power, will have by far the largest commercial radiator in space, a 30 m wingspan solar panel, and then we'll have very, very sophisticated optical and connectivity options on there. So customers including DoD and other satellites will be able to send us raw imagery, we process it on the edge, and then we just downlink the insight, and that's the roadmap.
How far are you? Like there's a launch that's happening very soon, and then there's the renders which are like, it looks like sci-fi Star Wars. Like there's like 25 shipping containers that are descending. What's the delta there? And what was the decision between like kind of showing more of a vision document? Because I think that's what triggered a lot of people, was like you were showing an amount of mass in orbit that feels like beyond the ISS. It looks huge.
Yeah. No, to be fair, it's extremely difficult to imagine what's about to come down the line with Starship. So like the launch cost might come down by between 10 and 100x. Launch capacity is set to go up by a thousand x or more. And the reason is if you build a new Starship every day for a year, which is like a very conservative estimate for what they're trying to do, at the end of the year you have 365 Starships. It has five times payload capacity. You can launch 10 times as frequently. So we're really talking about a thousand times like more tonnage per year that we can get to space. And that's coming like 3 to 5 years, which is really not that far away. So the idea of the render, like you can launch the whole ISS in two Starship launches. So the render is to show, look, there is going to be a lot more mass in space.
Yeah, I almost feel like you should have done the Anduril thing where it's like anime, if it's like sci-fi where the vision's going, and then it's like render for like what you're actually sending. Because like the render that I saw of like the multiple shipping containers putting together, like that's not what's actually launching next week. So I think that's where like a lot of the cognitive dissonance is coming from, right? But yeah, I mean you'll iterate all this. It's fine.
Yeah. I recently saw Palmer talking about that and I was like, man, that's actually pretty cool.
Yeah. It's nice to have because there is a value for like the vision document, like where do you want to go in 10 years? But it just hits different when it's like the official Nvidia account is sharing this render and it's like, okay, that feels like it's happening soon. It's not like they don't do their research. Like they have a whole team of PhDs that go and look at this, and like they have a board that approves that this can get shown, that it's like feasibly possible.
So talk about, I guess talk about, so this first launch is meant to be like a demo basically, correct me if I'm wrong, but effectively a science project to start to show capabilities in space, to start to prove that you guys can actually put GPUs in space that can run workloads. Like what is the actual timeline? And then it sounds like maybe defense is an early place where there might be some commercial applications, but then what's your timeline on if or when I would be prompting a model like ChatGPT or Gemini or whatever, and running it, it's just cheaper to do it in space? Because that's the long-term vision, right? It's just cheaper. It's not anything special or like, oh, we need to be there for some geopolitical or defense reason. It's just like the energy is free and the heat dissipation is free, so it's just cheaper. So that's where you go even though the launch costs are expensive.
Yes. So maybe I'll start with the second part. So when will it be cheaper? That's coming with Starship. And so we're looking at the first commercial launches of Starship in 18 months. But we're probably going to be seeing mass volume production of Starship on a 3 to 5 year time frame. So yeah, the roadmap is next year in October we're launching our second satellite. That one is about 10 times the compute of the first one, but much more capable in terms of continuous power generation and solar panels. That one is the first commercial service that will produce more cash than it costs to design, build, and launch. I mean that satellite pays for itself just on hosted payloads because that satellite has such high throughput and processing capabilities. There's a whole bunch of people that just want to stick sensors on that. It's not the end-state business model, but that's a nice revenue stream in the meantime. Then we're launching a satellite after that which will be, I mean, yeah, I can't talk too much about it, but you can see this PEZ dispenser form factor that's coming out of Starship. Probably shouldn't say too much.
And that was successfully tested on the last Starship launch, right? Where like there was a whole bunch of debris blowing around. It was a crazy, crazy launch, but it did come back. It landed and the PEZ dispenser thing shot out the fake Starlink or like the dummy Starlink.
Yeah, exactly. So that's about a 4 ton, at least 100 kilowatt satellite. Second one's 10 kW. First one's 1 kilowatt. So we're going up an order of magnitude each time. That one could feasibly launch end of 2027, and then from there we basically will launch lots of continuous versions of that because that will have very low launch cost, and it will be a while before we start launching the full Starship payload bay size. I mean for one we need this clamshell door to be ready before we can do that. And yeah, but to be frank, it's quite a while until we start docking things in space. The only reason you would dock things in space like in the concept video is if you want to train a large model, because then you need the whole data center to be internally. And so that we're probably looking at early 2030s before we start docking.
I mean, I don't have a problem with thinking in decades like that. That's like what we say in tech we want people to do. And so I don't think there's any problem with saying like, yeah, there's some crazy stuff that needs to happen with launch costs and it's going to take a while until this really works out. That's actually exactly what people want from technology founders ideally. I guess there is the question of how you frame it, you don't want people to think... If you just needed to generate a lot of revenue quickly in space, you could do the first space casino where people could beam up, tap into it, all ages.
Oh really? Yeah. I mean, a space token, you got to find it in space. What, how do you think about some of the... I think we had asked Dylan Patel at one point about GPUs in space, and I don't think he mentioned the challenges around heat and heat dissipation, but he was talking about how sometimes you just need to unplug the server and plug it back in again or switch out individual GPUs. But what's the solve on that long term? Are you thinking about these being fully robotic systems that allow you to make changes on the fly? Like what's the solution there?
Yeah, I think probably two points on that. I mean, the first is for these massive data centers where you've got like a kilometer long or a mile long row of racks, the people I speak to now are saying, look, if an H100 breaks in that, you just leave it in there. Nobody's going down there and unplugging and putting that back in. They're just writing software around it and they'll just leave it basically. So that's number one. But secondly, for the first few iterations of the satellite, it'll be exactly like Starlink in that we won't be able to touch it, and so we'll have to have redundancy on the critical systems, and then we overprovision things like solar panels and radiators which degrade over time. Over time the whole space industry in general is moving towards robotic maintenance, but also the terrestrial data center industry is moving towards robotic maintenance, and that's where on a sort of five-year time frame, some of the people we speak to in data centers are saying we are fully expecting data centers to be maintained by humanoid robots on a five-year time frame. And this is where it starts to sound a little bit wacky and sci-fi, but humanoid robots do not require too much modification to work in the sense that a human spacesuit would take care of radiation and thermals. You would need to retrain it slightly, but that maybe sounds a little bit wacky, but robotic.
Yeah, it doesn't have to be a humanoid, right? Like there's a bunch of other form factors.
It could be any form factor, but humanoids will be mass-produced, so probably the cheapest.
It's interesting. I want you to react to Andrew Mallip's viral post. Get 7,000 likes, breaking down a bunch of different things. The core claims I see in here are questions about what if demand for payload to orbit doesn't decline. He does talk about the difficulties of heat dissipation. He says you just finished watching a Scott Manley video on radiative heat transfer and now you think you're going to disrupt AWS with a few solar panels and a rideshare slot. You're going to believe that right up until the next month when you crack open Wikipedia. And he goes into a lot of the technical detail, but I was wondering if there were any key points in here that you think were easily debunked from actually being in the trenches. Obviously he does work in space and does something similar and does seem like an expert who'd be equipped to know this stuff, but you've obviously focused on this case specifically. So what was your reaction when you saw his comedic breakdown?
Yeah, I mean, I like the post and I'm a fan of Andrew. I've been following his stuff and he's a great engineer. Same with Dilithium, by the way. Very impressive what they're doing.
So on the thermals, the criticism usually is in order to dissipate that heat, you need a large surface area. And they think for some reason that that's super impractical. I mean my co-founder Israel is a PhD in engineering. Spent 10 years designing and building large deployable structures, solar panels and radiators. And I mean you just have to build a large surface area and that's what we're doing. So half our engineering team is building a very large, low-cost and low-mass deployable radiator. So that is the core IP of our company is that.
So that's number one. On the launch cost side of things, sometimes people say if there's no competitors to Starship, which it doesn't look like there'll be for at least five years, then will there be any incentive for SpaceX to drop the pricing? Now, I have a bit of a different opinion than most people. Most people say, well, no, they'll just keep it high. I don't think so. And it's the same reason that Tesla is sold at the price they sell at. They could potentially sell Tesla at a higher price, but in order to reach a much larger market, you need to sell it at a lower price. So, for example, they're going to have all of these Starships sitting there. They can only launch to Mars once every two years. Their marginal cost of launch on each Starship is $2 or $3 million. Now, either they can leave it there because there's no demand at the $200 million launch price. Or they can say to people like us, hey, you can launch for 20 or $30 million. They still make a lot of money. That's still very profitable for us. And so it's like a win-win. So my expectation is that actually the launch price will come down even if they don't have a huge number of competitors.
