Off-Nominal - 219 - ITAR-Certified Auction Houses (with Ed Ge)

Episode Date: November 21, 2025

Jake and Anthony are joined by Ed Ge, CEO and co-founder of Aethero, a space computing startup focused on building the best integrated hardware and software in space. Also we randomly go eBay shopping for old Russian space hardware.TopicsOff-Nominal - YouTubeEpisode 219 - ITAR-Certified Auction Houses (with Ed Ge) - YouTubeAethero - Space Data, Re-ImaginedCosmic Shielding works with Aethero to protect Nvidia Jetson Orin NX GPU - SpaceNewsROSCOSMOS COSMONAUT SERGEY RYZHIKOV SPARE WINTER JACKET SOYUZ MS-27 ISS EXP. 73 | eBayPREFLIGHT THE CHALLENGE MOVIE COSMONAUT A. SHKAPLEROV JACKET SOYUZ MS-19 | eBayCosmonaut Oleg Novitsky Jersey T-shirt Soyuz MS-18 ISS Expedition 65 Gagarin :) | eBayLOK Lunar Lander LK Salyut Space Station Periscope Extremely Rare Artifact | eBayActual Console Panel Soyuz TM/Mir Space Station Extremely Rare | eBayFollow EdEdward (@somefoundersalt) / XFollow Off-NominalSubscribe to the show! - Off-NominalSupport the show, join the DiscordOff-Nominal (@offnom) / TwitterOff-Nominal (@[e-mail]) - Spacey SpaceFollow JakeWeMartians Podcast - Follow Humanity's Journey to MarsWeMartians Podcast (@We_Martians) | TwitterJake Robins (@JakeOnOrbit) | TwitterJake Robins (@[e-mail]) - Spacey SpaceFollow AnthonyMain Engine Cut OffMain Engine Cut Off (@WeHaveMECO) | TwitterMain Engine Cut Off (@[e-mail]) - Spacey SpaceAnthony Colangelo (@acolangelo) | TwitterAnthony Colangelo (@[e-mail]) - jawns.club 🐘Off-Nominal MerchandiseOff-Nominal Logo TeeWeMartians Shop | MECO Shop

TLS and go for main engines, start. Jake, happy Thursday in the era of so many New Glens. Since the last time we talked, there's a whole new one. Somewhere between nine and four New Glens we've had lately. So it's just, it's out of control, right? Yeah, the falconization of rockets continues. How you doing? I'm doing good.

We've got Ed, let's see if I say this right, Ed. Ed Gee, I'm just going to go for a, for a, guess on that one. Hard? Good enough. Good enough. Okay. You said good, which is also soft. Should we stick with the soft G? And it's not good enough? It's not. Are you just roll with it? Just go with it. All right. Okay. Edgy. All right. Cool. We're going to talk about we've been

trying to get this episode on the books for a while. We've had the schedule gods have not been kind to us trying to get this interview going. So I'm glad that we finally were able to make this happen because it was like one thing on this side and one thing on that side and we just couldn't get together. But we're here now and we're excited to talk about AI powered satellites and satellite capabilities and I guess compute in space. You're going to tell us what we're going to talk about because it sounds like it sounds like pretty fun. But maybe just introduce yourself. Who are you and what are you doing here?

Got it. Yeah. So I'm Ed. I'm the co-founder and CEO of Athrow Space, a company building radiation hardens, spacegrade AI computers for satellites. enable variety of missions from advanced Earth observation all the way to missile defense on orbit servicing in space autonomy, and of course the famous orbital data centers that SpaceX and Google seem to be going after. So we essentially have a full suite of products ranging from space rate GPUs in a box, all the way to a fully integrated software suite that enables autonomous like users to train and automate and build new.

computer vision models off of the satellite data that the satellite likes collect when they're on orbit, consolidation management over the air updates software to enable you to manage your AI-lit modules onboard space raft, as well as multiple module generations. So we actually have one existence like heritage computer that's based on the NVIDIA or NX GPU. We have more coming out based on the AGX and one on the IGX, in addition to products that enable satellites to run multiple computers and the distributed sector like the true data center would instead of taking one large server strapping into satellite and telling them a dad down center. Yeah, you don't want a GPU satellite.

You do want an actual data center. An actual data center gives you reliability and resiliency. That data center is actually had. That's a point of a data center, right? right right right okay cool that's a lot of stuff i think we've got a good that's a good slate of uh conversation anthony and i've been talking about AI uh just before this show so i'm sure we'll have lots of interesting who hasn't different yeah ed did you bring a drink today do you have anything prepared to show us i got a bottle of water that's uh i'm pragmatic and efficient i suppose

it fits with the vibe okay i love it love it's it's cooling you down Yeah, it's cooling you down. Water cooled. What do I got? I got a victory hop devil, Jake. Shout out to the Jersey devil. This one's delicious. If you like an IPA, boy, will you like the victory hop devil?

It's a great one. I dig it. Some people would say it their best. Yeah. That's what I got. A great thing about an IPA is that as long as you make the IPA, you can't screw it up because it's just like it completely blows out all the taste. a mead elitist down there.

What did you bring anything you made or you just, whatever? No, I got a cocktail today. It's coming, don't worry. But I have this cocktail today. This is a rum and tune. So this is, I've had this type of liqueur on before. This is Ishtabintun, which is like a honey based, like liquor here with a star an ease.

So it's got like that black licorish thing. But it's kind of a Mayan drink. and I saw this new brand, this Yumbub, so I thought I'd give it a good. It's very smooth. I like it. It's really good. A little bit of rum, a little bit of lime juice in here. Perfect.

Up on that new brands of the Yucatan situation down there. Look at you. New brands of the Yucatan. It's like honey-based rum. Is that what you said? Yeah, it's like fermented honey and they put star-knee, so it's like black licorice is like the dominant flavor in there. So it's kind of it's pretty like it's very has like a syrupy consistency.

So it's like very like smooth and sweet and yeah. Well, it's good stuff. Good stuff. Yeah. Well, okay. You start, Jack. Where are we going? Okay.

Well, I mean, the first thing that I think that I want to kind of talk about is just, um, uh, I think there's like some obvious things about like AI on satellites that most people will kind of get. You know, I'm thinking about how these things can, like, work together and, like, collect a lot of data and then, like, make decisions, right? Which is, like, I think an important part of why you'd want to do this. But I do also want to get on to this whole data center and space thing,

because that is, like, one where I haven't quite, I haven't quite gotten my brain to a place where I understand why anyone would want to do that. And so I'm really hoping that you can kind of talk a little bit about what the business case is, because it just seems, it seems kind of weird to me to put all your compute up in space, where it's like far away from us and has latency, right? So I don't know if the, if that makes sense to me.

But maybe I'll do that. Like, what's the business case? Like, why do I, why do I care if there is AI on my satellites? Like, what is the need there? Well, let's define data centers in space from first principles, right? Like when people talk about data centers in space, usually everyone's thinking about training GPT6, whatever, in space, doing like AI hyper-shelene in space.

And ultimately, that's less for the day for it, like, actually AI in space, more for today on space-space solar power. That power's going to be cheaper in space personally. I don't feel too technically, that's not within my area of expertise. I don't think that's going to be the case within the next, within like the 2030s that it will make more sense to put like a massive AI data center in space, set in on Earth. And that problem at the 2040s is still very far away.

So God knows what the hell we're going to be doing then, if the AI theme is. So I'm going to be your net top trend. But for AI on satellites, it makes sense quite a bit, right? Because a lot of these satellites collect a ton of data, and sending all this data back down to Earth is quite inefficient, particularly if, let's say, you're capturing, like, International Space Station produces terabytes of data every single day. And they produce so much raw research data that they have to send hard drives down on the vacuum capsules.

And it's like when you put, When you're producing so much data and the ground station periods are so limited, and now data can send down during each ground season contact period is limited. It makes sense to process data in space. And then you go into more advanced use cases, right? You look at onboard servicing. When you're doing autonomous RPO, we look at missile defense, when you're trying to detect an object moving out Mach 25 across the atmosphere,

an object moving like Mach 30 in space, you can't really wait to send this data down to the ground and then have someone looked at the data and then send it back up again, you need this data to be processed instantaneously, you need a satellite to react instantaneously, and that ultimate is where the biggest, biggest use case of AI in space comes from.

I'm laughing at hard drives in the dragon. Yeah, the bandwidth of a dragon capsule. What is it? Yeah, what is that famous? Was that the next case? Wait, who wrote the, no, I forget, somebody wrote what is the bandwidth of the 747? It was just like years ago, right?

like how much data could you send across the world and the time it takes to fly there. And now this is even better with like loading the dragon up with a bunch of hard. But the guys have a third of an even fun. There's some, there's a protocol called IPV, avian, where you try to essentially have internet data packets being sent through carrier pigeon.

And apparently it's a very high packet loss rate, affordable latency, highly not recommended. Severe packet loss at all time. Carrier pigeon packet loss. Yeah. Here comes another. Yeah.

No, someone, this is like someone to do. For that. He can't be on TCP. I think I've told the story before, but someone in Canada did that once as an experiment to protest our high bandwidth prices because the ISPs that were pretty bad for a while. Still kind of are. But someone did the same thing where they like, they tried to upload a large file from Halifax to Vancouver and then they erased it by putting a hard drive on an airplane and flew it over there. And the airplane won, like, you know, like very, very easily. so it was not a good situation.

The other aspect, though, real quick, you're saying the ISS generates so much data, right? You think about, like, planet imaging. What percentage of their images are for, like, any particular use case? Now, that might be a bad case because they sell to a lot of their deal, but, like, if you had a particular use case in mind, what percentage of their images are just junk to you and, like, noise, and you don't actually care about them if you're looking for,

maybe one of the better examples would be, like, Umbra or somebody that's doing sensing of like ships at sea, right? And they're doing, or even Sentinel. That just launched Sentinel 1D that's doing like radar imaging and then tracking ship beacons. And they're cross-referencing to find a ship that is not broadcasting, but is sailing. It's like so much percentage of the stuff they collect is just noise.

But you only care about the one little subset. So just send me that. Like don't send me all this other stuff and make me deal with it. You deal with it and let me know when I should go look. That's definitely the point. And these companies like Planet, like Sentinel, are actively integrating AI onto the satellites, right? Like, Planet right after us, they start sending up GPUs onto their own spacecraft.

A planet's among the main partners now from Google's TPU and Space Program. Like, these EO companies badly want AI on the satellites, because, like you said, so much of what they collect is just noise. No one is paying millions of dollars for images of the empty ocean, for images of a desert that, like, like, the Sahara's probably important. But no one's, I'm not paying $10,000. Or to like a sound like 16 photos Out of a day

Like one every once in a while Probably good Do you need it every You know 90 minutes? No Yeah Exactly

And not to mention Cloud cover as well Yeah Which Yeah Oh you're muted Jake or something I can't hear you Can you hear Amid

I can't I know He's gone I don't know That's because they heard you He told all this smack talk About AI in space and your satellite internet is out.

You can hear us. This is great. We control them. At least you're not doing that thing. I was just laughing. The internet broke down to Yucatan. Maybe he should say him a curio-tigian. He's on a Starlink.

So I think it's more that he's trashing their upcoming business model. And that's what did it do. You might need to reboot and come back. Yeah, you're coming back in. Oh, man. The other week, he had a real big incident where his mic got stuck repeating the same noise over. over again and it drove everybody nuts. So this is at least better than that if we can hear.

Can you talk about what kind of hardware situation is right now with the company? What have you have out there flying or what are you working on? What's the status overall? Absolutely. So our first generation flight computer based on the Oren Nano

and OrenNX as flight heritage we're actively flying it. We have customers actively flying it as well by the end of this year. We already have a second generation module based on the AegeX Oren coming out by the end of this year of O'HFlight Heritage as soon as the first half of next year.

In addition to that, we already have begun working on building our own satellite platform. So one thing people have asked me a lot is, why do you guys use companies like an off-orbital? Why do you guys insist on doing all of the demonstrator missions in-house? And my

reply to that is, I hate the thought of being a component's company. I would rather shoot myself and run a component's company. God help me. And so I have a strong bias against being a component's, they pure, the components company. And that's because,

like, the department of war does require companies to, uh, have their own platform, right, the wrong solutions. If you've got to contract to be a prime, ultimately in space,

that is where the bulk of the money is, at least for the American space industry. So if you are a component, if you are a component company, you are essentially stuck as a subcontractor, uh, for most of your business. life cycle. Some people might be like, that's fine. I'm fine building a 50 million dollar business. I'd be happy with 10 million. I'm not. I want billions. I am not here to make 10 million. I want to make 10 million. I would have gone to big touch, read that. But that really does. So that's really the sentiment I'm moving ahead in, right? And we have a second satellite launching in February of next year. We have a third satellite, a full S-Fodfast launching in the summer.

as well. And these missions are giving us slight heritage on our next generation systems. These missions are giving us heritage on a full software ecosystem. They're giving us heritage on our ecosystem of expansion modules. I add additional functional functionalities to our computing prowess. I add functionalities for the software-fine radios, functionalities for our onboard computers, or high-speed data payloads. And this really enables us to tackle the market, not just as a space hedge computer, components provider, but as a fully fledged prime contractor that's capable of working with bus manufacturers, white labeling the buses, and offering full solutions to the end customer.

Are you back, Jake? I think so. You're back. Yeah. Yeah, yeah. You call most of that, I think. I think my USB hub died. The chip on there was not rad-hardened.

It's probably because you bought it from a shitty component manufacturer, Jake. I could have gone for the full-down hardware manufacturer. Yeah, I should have gone for the full service. Yeah, you should have bought it from somebody that makes the whole damn package, Jake. Okay. Love it. Before I was rudely interrupted by my own hardware, I was going to ask, you know, like, this is not where all the money is, but the planetary imaging situation, I think is one where it could really benefit from this. But I'm kind of curious to hear your take on it because, like, if you look at something like the Curiosity Rover, you know, even the person,

We're in averance rover, they are sucking up a ton of data from their instrumentation. And they're leaving it on the surface, right? Like the deep space network does not have enough bandwidth to bring every bit home. And some of that data is going to be like required for onboard processing. You know, like their navigation cameras that are just like it can like auto drive, right? It's like, you know, taking a whole bunch pictures as it drives and changing its courses and doing things like that, which feels like a very good like use case for this kind of technology. but then sometimes it's like they got to take a picture of a rock because there's something interesting on it

and they got to take 1,000 photos of it from every angle and then send it all up and then like you know we got to go through that the hard way to try and figure out where the where the useful stuff is in it there so like we could we could help with the bandwidth by picking the you know the best 5% of those pictures or like even doing something interesting like kind of combining them into like more derivative data products that are that are like you know more cost-efficient bandwidth-wise, but still preserve all of the actual information that we need to do science on it, right? But where I'm going to go with this, though, is that I know that these rovers require so much redundancy and, like, reliability that the computers on board are 10, 20 years behind, right? And so they have to be rad-hardened, obviously, for, like, somewhere like the surface of Mars. But how do we get this kind of technology into those things without just having to, like, wait for the, 2045 Rover that's going to have today's technology that you're talking about, right? So that's a good question. So for deep space computing, GPUs ultimately do run into some limitations based on the fact

they're designed for terrestrial use cases, right? They're designed for performance, aren't really designed for resiliency or redundancy. One thing that we've seen, some of our customers do for six-noon missions or missions to near-Earth asteroids is you simply have multiple GPUs on board. And when one dies, you just switch it off, switch on the next one. And that works really well for, like, clustering them together, since that means tasks can be carried over seamlessly. But ultimately, we do believe that if you really, really want to,

like, send missions on, like, 30-year missions, 40-year interplanetary missions, you do really need to look at Radpart by design chips. But ultimately, there isn't that much. big of a market deal just yet, right? For a radiation hardened by design GPU chip that can last for 40 years in deep space, because let's be honest, like if you go

to Nvidia, how much AI chips do you think Nvidia sells in a week, right? 100,000. How many satellites do you think will be in orbit in the whole decade, like get launched in a decade? Even with starting, assuming all the Chinese

mega constellations all happen and Kuiper as well, probably 100,000. So like we look at the decade versus the week or a status a month, it doesn't really make sense for any of these AI or chip design companies to be designing like a specifically rad hard chip because you'll never have the production volume to form a chip fab to justify unless we're willing to diversify the use case. But then you just end up the chip that's massively overkill for any environment that you have on Earth. Yeah. So you're saying no. Yeah, I was going to say, unless you're ban. on the nuclear testing situation.

You know, we might need rad hard and chips more often on the surface these days. Well, the way I'd phrase it as it's, there are, the reasons of obstacles blocking us in doing this aren't just on the technical side. There are also obstacles on the market side, right? Because ultimately, the market is rational. The market wants to make money. If it doesn't make as much money as pursuing an alternative option, there simply isn't a market for it.

And with our distant customers, we do believe that, like, the alternative of, hey, well, we'll start targeting, like, Rad-hard MCUs. We'll start targeting radiation hardening, all the individual ICs and peripherals around the chip. And eventually we'll target the chip itself when the market is mature, when the market makes sense. But for now, adopting the strategy of, hey, we'll have multiple backup computers on board. That fell, switch it off, that fell, switch it off, that fell switch it off. That makes more sense. and particularly for these large missions, like they're talking to Curiosity Rover.

These missions from JPL, the flagship missions, they have billions of dollars, the prestige of the whole agency attached. No engineer wants to be that engineer that made the call of saying, hey, we'll only use that brand new GPU from Nvidia, that GPU fails, and all of a sudden you become the most hated man

in the entire space industry. Yeah. It's pretty argument. You pretty much summed it up there Who bought these components Somebody buy components instead of a whole spacecraft Get them out of here I was trying to talk about NASA missions

Not making money Geez okay all right So So when you're looking at your own hardware And putting these platforms out there into the world What are the markets that you're looking at in terms of like the buyers, the customers for these platforms specifically, and do you hand those over for them to operate, or are you also trying to run operations for the satellites

that you launch as well? So that's an interesting question. So one of the cases that we've gone after is running operations for our own satellites that we launch, right? So our upcoming missions, we actually have right customers, ride sharing, software payloads, and fodder payloads with us, where they have censures, they have software that need a ton of too, whether the tests doing LOM research on edge environments in space, what tests are new imaging sensors,

and these customers have these sensors and software on board of satellites and run them, and we essentially contact them not only for the space onboard a salad, but for accessing the compute resources from all of our GPUs such as it on board. So that essentially is a semi-orbital data center model that actually works very well, actually, like, an AWS of space. But another thing that we've also noticed, the Space Force and the Department of War,

it's an increasing enthusiasm for a model where commercial operators launched a DOW satellite into space, still the W satellites and stuff. They launched into space, they operated themselves, and essentially they're held in reserve by the Department of War and called upon when needed, right? Because let's be honest, most of the time the satellite's up there, it's not really needed as an active asset.

It's much better to have flexible assets, flexible reserves that can call upon in times of need. And that's what we think the direct is going to be, particularly for ISR and SBA assets. I just watched that movie House of Dynamite, and it's making me think of that. It's like a nuclear war scenario with detection. Yeah, it's good. It's brand new. You should check it out.

Yeah, but it's got kind of the same thing, the early detection, and then, you know, the ground-based interceptors. all that kind of stuff. It's been a it's on my mind right now. But one of the themes like one of the themes of it, of course, is like, you know, there's all these resources that are just kind of sitting around waiting to see if something happens. And it takes a little bit to spin up that machine because they're kind of like, oh, yeah, we get these false alarms all the time. And then it's like, oh, this one. Okay, well, this will escalate this one to level two. Yeah, it's probably just this. Oh, I know it's level three. Okay. We should start calling people on this. and like, you know,

all of a sudden. But that calls it, sort of unlines your point about the flexible assets, right? Because there's like a lot of just kind of sitting around with this stuff.

I don't know, but. You just hope there's one Stanislav Petrov in there. You hope. Ready to call the false alarm when he sees it, you know? Flock of geese.

Flok of geese. So it's for the, the space development agency, the stuff that you're talking about there, with, you know, or the resilient kind of constellation from the Department of Defense angle, Department of War angle. Have you noticed the way that they talk about that shift over the years? Because I feel like the idea for what now is, is it, is the proliferated, what is it,

PWSA, proliferated? Perliferated warfighter space architecture. There it is. Terrible name. And there's also another SBA too. They have space domain awareness, which I was referring to. And then you have the space development agency. and then just face that association

and it's a whole this nightmare of branding I feel like the talk about this started a couple of years before the moment we're in now technologically began and I wonder how when you go out and talk about this stuff about the capabilities that we actually need

on these satellites to run this kind of stuff has the understanding of what is needed there made its way up to the level of decision makers in, you know, let's just look at the defense angle. Like, do they have a good understanding that where the hardware was when they started talking about this isn't where it is today and where it is today is what we actually need for this? And, like, how are they kind of designing the program to actually take advantage of the stuff that newer entrants are actually working on these days rather than let's do a new architecture with the old hardware, which is kind of

how it feels like it started? That's a good question. And I would say the understanding, the has improved quite significantly over the years, right? You have programs like SBA Halo. It's explicitly calling for new small satellite manufacturers within the United States. You step up and take a shot at the Gulf.

You have the SDA explicitly sending money aside for new hardware manufacturers, new satellite companies and new satellite bus companies in the U.S. You have new semi-primes appearing like companies like K2 space, Apex Space,

that are now actively bidding on and competitive with the old primes like Lockheed and Raytheon on these large-based programs and you also have what's starting

to become more noticeable which is a retreat of the larger primes from these programs are oftentimes fixed price the older primes they want to do cost plus you don't want to do fixed price

like when NASA said we're going to do fixed price like bowling straight up to stop bidding on the NASA contract and remember the reception in the space community in the space industry at the time was I hope you do I hope you stop

dating and yeah you're getting the point yeah you're getting the point right you're getting the point but I do think as we start to move towards more fixed price contracts that gives a massive advantage two smaller startups that have the ability to move fast move on technology like develop technology at low cost and scaled it up effectively versus larger primes or loaded with just bureaucratic red tape. Even the mindset, too, of what you're talking about, of like, you know,

don't spend the money to rad hard in a GPU, but ship five or six of them and know that you're going to have a bunch of failovers happen. I feel like many people in the industry would have shuddered at that thought 10 or 15, 20 years ago. And, you know, that's,

that even harkens back to, like, the original outset of Google, which was like, we could spend a bunch of money on expensive servers that don't fail, or we could buy the shitty, servers on the market and just buy way more of them than we think we need. And that eventually did win out as a strategy play versus their competitors that were buying exquisite units that

would never fail. So, um, and, and I think like if you read a bunch of space news, literally like space news.com, right? There's always somebody high up in one of the space defense agencies talking about how we've got to embrace the new era of resiliency and buying cheaper satellites, than really expensive targets. And they've sort of done it a little bit, but I don't really feel like they've fully adopted that mechanism. Well, it's difficult, right? Because imagine that you're a program manager in the SDA

and you bought 10 satellites, and half of them failed because you bought them using relatively unproven cheap cost technology. They were cheap, right? But they were cheap. It's not probably a good excuse to give to Congress or to give to your business. boss with half of your sandwich.

I mean, but the excuse of the I bought five more than anyone else did. Like, you know, that, right? The outcome-driven approach is, I think, how you went out. Yes, but there is a line that has to be drawn. And ultimately, it is that there's a line that has to be drawn, right? Just for example, I know a lot of space companies now are using automolive grade parts or industrial rate parts, subspace rate parts. And because for short permissions, all the mold of industrial grade are just good enough

to work in lower surveys. And that would have been completely unheard of in the 1980s and 1990s. So, hey, you can take the same resistors and capacitors. I used my car and put them on a satellite and send that satellite into space. And it worked because the risk culture just was inferred. He launched prices were so high that we sent up a satellite and it failed. It would be a year or two before you could send up a satellite again. But now with Falcon 9, with New Zealand, with all of these launch options coming up,

If a salad like Bell, screw it. Wait three months, six months, sent another one. Yeah, this is the whole iPhone project at Ames, right? Same idea. The iPhone project, Ames. Pour one out for Ames. Yeah. I also wanted to point out, Anthony, that Curiosity rover has two computers.

That was more than 15 years ago. Look at this. That was more than 15 years ago. So maybe check your dates a little bit. Are you saying I'm old? They're both broken, by the way. Yeah, how many wheels did opportunity have, Jake? That really worked out.

Outcome driven, though, Anthony. It's still driving. So outcome driven. It's more wheels than most people sent to Mars. You know what I mean? Yeah, maybe three that are broken, but you know what? Three more than you're upset. You know what I'm saying?

Exactly. All right. So you are, first of all, Ed, I fucking love how opinionated you are about shit. It's my favorite. I love it. I think most people are scared to take opinions, and I appreciate that you are not. So convince me on, you went from, I don't want to be a component manufacturer. I'm building the whole thing.

What is the special sauce? Why are your buses better than everyone else is out there? It's about integration. We don't just do, you. So a lot of people go after just a hard tool, just go after the software, right? they just see people doing just mission software, mission control software, or just like satellite with my software,

or see if we're just trying to go out to just pure components, let me just give you the hardware and you fuck off. I don't think that's really a play that works out well. I think that is the tagline of Rocket Lab space systems, by the way. I'm pretty sure. I'll check the website, but I think that might be. But ultimately, you need to give both the hard run software.

The hardware and software ultimately are the same product. You look at the iPhone, right? Why is the iPhone such a beautiful product? Why is Apple such a well-known company? Why do your products? Why do people like Apple products? You're singing Anthony's language here. The products of anyone else.

Because Apple products have the hardware and software integrated with each other. The software in itself of the Mac OS, the iOS, is a product in itself, right? And this is what makes Apple devices so sticky. It's the feature, the seamless integration and intercompatibility between all products within the Apple ecosystem. If Apple didn't have that, I don't think people would love Apple as much. If Apple didn't have that, I don't think Apple would be one-tenth of the size it is today. And if you look at that approach, that's the same approach we've taken with our product. When we build this hardware and the software, we build them at the same time.

We integrate everything together and we make sure that everything works very, very well together. Because the whole point of an ego system is seamless to the compatibility. And I look at all these companies just doing hardware, just doing software. I think that's what they're really missing. I dig it He's got me, Jake Yeah When you're over

With the Apple love Yeah What's the quote? Everybody company That's serious about Software makes their own hardware That's what's the You got it right enough

It's wrong But you got it right enough Okay good I'm just a software It's the other way around Yeah It's the other way around Okay

That's right You got the vibe Not the details Yeah You basically LLM to that Your way through that one Jake You can use me as a source of idea.

It's just not a source. Yeah. So what is the earth infrastructure like for the company then? Like, you know, factories, production lines, what's the idea there? What do you have set up? Where are you going with it? So we started off set up once a design house where we handled the final integration and assembly in-house. Now as the company is

grown, we've done more and more to insource functionalities like testing, small-scale production, and eventually, as we grow, we do want to integrate, like, everything from board assembly to final testing and final delivery in-house, right? Because the more it can bring on in-house,

the easier it is to maintain strict quality control, the easier it is to trace any flaw to a single step in the production process. It's the same thing that Elon Musk did when he approached starting to the same thing that Tesla does where they tightly control every single step with the process to understand if something goes wrong, they understand exactly where and why it went wrong. And that ultimately,

I think, is something that any serious argument company should do. Yeah. So where were the first things built from, if you were design shop originally, like, where were the first round of hardware built out? And how did you put that all together? We actually worked with some PCB assembly shops down in South Bay. We put them together. We got the enclosures from a local CNC shop, and that's how we actually launched them, the fraud that we launched into space. From there, we've expanded working with more PCB assembly shops

in the South Bay, as well as more services for CNC. We've begun sourcing from CNC shops from all over the country. We've had some issues and hiccups with the quality of some of our suppliers. So we've now, we've Now we've learned several painful lessons around having backup suppliers and backup suppliers and backup suppliers and this also has enabled us to get stuff like our AS9 100 certification, gain all these industry that mean industry standards or quality mean industry standards

that enable us to get through to fly on our customer spacecraft. I feel like you're circling around the vertical integration discussion there, right? That's like backup suppliers eventually to get to a. point where you're like, we're just going to build the CNC machine in our warehouse and do it ourselves, right? Ultimately, that is the goal. CNC has proven to be trickier than one, my essence. Yeah, I bet it is.

This is the living in Mexico vibe, totally, because it's just like, you want to, you want to get some sort of, you know, a supplier for stuff or you want, like, service to do stuff for you. And it's, like, really inconsistent and hard to find. And then eventually, you're just like, well, I'm, I'm, I'm, I'm a supplier for stuff. electrician now. That's the situation. So I get it.

I feel that vibe. I feel that energy. Do these people know what they were working on when they were running this through the CNC process? Do they know like this is? I think I think they do did. They knew it was so like we told them this has to be ITAR like we went to the AS9100 compliant and ITAR compliant shops. right we can't ship this out to China

but they did know and ultimately I do think it does come down to a bit of if it's not your product you don't really care about the quality as a month so the more than vertically integrate does it also brings a stake like the huge answerable

if this thing fails to within the scope and responsibility of the company yeah yeah that is like I'm gonna keep going on this analogy but that is like a very much a DIY advantage right because it's like yeah maybe you're not like good at you know lumber or carpentry

and electrician stuff but like nobody cares about your house the way you care about your house and there's like there's an advantage to like getting the work done for that or like you won't you won't you know cut corners or do things like that it's a thing man it's like a real real thing that I believe in very strongly if you knew you lord Jake there was another Steve jobs quote you could have used here but oh is there there's a back of the cabinet quote that you could have used. Give it to me. Let's learn a Steve Jobs quote today on off-nob.

If I said there's a Steve Jobs quote about the back of the cabinet, if you rift this right now, I bet you would nail it. Something, something, you can tell how good it is if you open up the cupboard doors and look at the back of the...

I don't know. Yeah, you nailed it, yeah. Basically, TLDR, the back of the cabinet, as good as the front of the cabinet because this guy cares about it work. Yeah, yeah. You L-LM did, but it's good, yeah.

Mm-hmm. Okay. That's a longer one, admittedly, than the one that you screwed up earlier. That was one sentence at most. Okay, so when you set up these fully vertically integrated situations, where are you going to do this at? You stay in California? Is it going to be nearby where you're sitting right now? Do you know yet?

I heard the vibe has moved to Texas. I don't know. Or Long Beach. You could buy a failed space company's place in Long Beach at some point. I'm sure, another one will come up for auction. I'm sure. I'm sure. There seems to be equal.

I'm sure that given once the next bubble crashes, there'll be quite a few failed space companies with pricey real estate that we can... And data centers, too, yeah. And data centers, yeah. It's true. I've actually, in our office, we have some relics that were acquired

from a failed space company, spake-flip the auction. And I believe that failed space company acquired for another failed space company. Come on, tell us what they are. What is it? What'd you get? What'd you buy?

I can't have comment on the specifics, but I do assume that it's like the office cheers in San Francisco, right? Like, everyone has an urban miller chair, but the urban literature should buy them like $1,000. You buy them off of like Facebook, a marketplace of like $300. So no one in San Francisco actually buys like the brand new urban miniature. There's like a 10,000 urban millerchairs slowing around the city and they all like rotate through startups. right and i'm assuming it's going to be the same like office reddicts like you have like paintings and posters and just keep rotating through bank club startups yeah you have one of the inspirational posters that's like you know the corporate art with like that's a bit too old

but we don't have some page in the sky is not the limit yeah the sky is not the limit yeah evidentia was for something yeah couple of them. Wow. Man. Okay. Jake, we should, we've been thinking of dumb games we could play.

Remember we thought of, what was the fantasy football style game that we thought of a couple weeks ago? Oh, like rocket, rocket fantasy football, right? We like draft which rockets. Right. The other one I have is like, let's do a power ranking of like next space companies to give up their office relics and Herman Miller chairs.

Let's see if we could get it right. I would be interesting seeing that Because there's quite a bit of The auctions are always fun I'm assuming right I have friends So man they love those auctions They're like we got a whole factory out of it

Yeah I have friends who have fought everything From TVAC chambers to actual full blown shells of rocket engines From these banked leapsy auctions And I'm like Wow you pretty much have everything you need You start a full blown launch company

right here. Are these ITAR certified auction houses? Yeah, you're out, Jake.

You can't come. You could definitely not come. Just be an ed hanging out. I would assume so. I mean, I don't even know. How do you even get

IPAR certified as an auction house? How do you auction anything I tar covered? Like, that seems like a no-no. I mean, it's a shell,

right? So you don't actually have the plumbing and valve inside and just get like the rough dimension. I suppose that's not it's not ITAR.

It would depends if they've drilled the throat out. That's the really the injector plate and the throat that's going to get you. I assume there's probably some auction house somewhere that is

maybe like there's a federal option, right? Like when in the U.S. Navy sells off and some of the old chips, like can China or Russia send an agent saying, hey, no, I'm Vladimir Putinov. I'm no relation to Vladimir Putin. I'm here to buy kills of frigates. But can we do the same thing?

the Russian suit, like when the Russian, when the Ross Cosmos, like, like Russia Gossom, everything is for sale. I've one third of a Soyuz capsule being sold in Russia, right? And I was like, how they get my hands on the fucking Soyuz capsule? I want that Soyuz capsule.

When they auction off, when they auctioned off the old hardware to put the light on. You guys won't believe this. I've seen a toy that's from the International Space Station full-blown Russian space suits like worn on EVAs. For shell on eBay.

And one of them was quite a fun listening because I remember looking at the listing and could see the space suits have the Viger, right? Vigar's gold. It's reflective. And you could see like the surroundings of the space suit. And it was some guys living room. Was it Tim Dodd? Some guy has,

some guy in Canada. Tim bought his on eBay, right? Where did Tim buy his? Yeah, I think he bought it on eBay. Yeah. It's like, but his is like a high altitude flight. though, isn't it? I don't care. It counts enough for me.

I mean, this one was a full blown of Russian Orlon space suit. It's some guy's living room and the same guy has a toilet from the ISS on also on sale. I don't know. I assume that happened together. Some Russians stole the ISS toilet and brought it back and then sold the Soyuz, the space suit, and the toilet all in one go. another part I remember seeing to use an unidentified electronic part and from a

Russian station. What is it? It's an unidentified electronic part. Sure, all right. That's CIA Honeypot for sure. It's the NACA IMU is what it was. You see all sorts of shit on eBay now,

right? We've got to buy one at some point, Jake. We got to do this. Every once in a while, one of these auctions flows across the Discord. and we're just like, oh, what is? Oh, it's like, oh, it's like a tire from, you know, from the bus that Yuri Gagarin peed on or something.

And you're like, well, maybe I'll buy this. Like maybe. I would buy that for sure. It's $8,000, but like, maybe I'll buy it. If that was available, I would make that a toilet. And I would, I would be great. I don't think you get good luck if you don't go into space after. I think that's all the works. I guess that's right.

You kind of need to go into space for the good luck to happen. that would be awesome yeah we got to buy something at some point there's a lot of these good ones do a Kickstarter help off nominal buy uriguergen's pee tire yeah is it still the same tire do we know because they all pee on the tire right so listen the story is much more important than the rubber okay how do you know it's the same tire like did they keep the tire did they remove the tire and say uriga geron peed on this tire we got playing the warehouse from future posteri I mean the way the warehouse that

that just demolishes itself over the years Yeah, the warehouse whose ceiling is not compromised but yeah At the center of a land dispute The way that they treat you Gagarin I assume that tire

is preserved somewhere in a vault I mean to be fair The Russ is still do half of the King's space shuttle inside a warehouse And I've seen Siberia because it's Kazakhstan We got to save that

We got to do something You could probably buy it I'm not kidding could probably buy it and get them to ship it to you. You could probably steal it easier. What do you think happened to the one-web satellites that got stolen at the beginning of the Ukraine war? I think they still sit inside the warehouse

if some Russian guy hasn't sold it and now suspiciously has a new dot shot outside of Moscow. They just have 36 one-web satellites stacked up somewhere. In his living room? It'll show up on eBay eventually. We'll show up on eBay. Us three will go in on buying a one-web satellite at some point and we'll be great.

Is it before or after the ISS sign it? Oh man, that's a good item. I know exactly who Jake and I would send that to if we bought that. Yeah. Yeah. That would go right to Brendan Byrne. We'd put it in his house. Yeah.

Best birthday present I could ever give them. Would you put it in your house? Would you set up like a bathroom that looks like a space station bathroom in your house? I don't think the ISS flight works in 1G. That's a problem. That sucks. it out no matter what, man. It's just better in 1G. You know,

it's like a vacuum thing, Jake. There's airflow involved. You don't know about this? This is the the main reason that I would be hesitant to go to space. I don't want to... This is the one? You know what the one for me is?

It's cruise ships. I look at cruise ships. I'm like, God, everyone's got neurovirus and there's like strange murders that happen a lot. And that's like, they have fresh air also. Like there's no way this is better with not fresh air And I take cruise ships all the time And they don't bother me either

Yeah it makes me nervous It makes me nervous to apply that To close air I'm not into that man I'm not into that I mean to be sure It's not enough And it's also no

You also tap really fall off The spaceship and die I mean You watch the expanse man You can get They space so many people in that Although to your point though it would be easier to use a space toilet

if you did have neurovirus. That would be a better full package, Jake. Okay? In no way am I married to a gastroenterologist, by the way, in which I am fully comfortable talking about this, and you are not. That was interesting is that they have personal flight suits of the Soyuz astronauts, right?

So I'm going to have seen a personal flight suit. Of this one Russian Soyuz astronaut, you have to wind up on several missions. ISS and I was thinking, how the hell did they get that thing? And then you also look into it and just like that food package, a food package at that expiration date in 2027. They're like, wow, I don't know how the hell they're getting it, but they obviously have someone in Rockin.

Like current was sending them free shit. Yeah. That's bleak. That's a bleakest report of the Russian space industry I've ever heard. As someone who lives in a developing country, let me assure you, corruption in the government level is a heck of a thing, man. You can do all the times of shit. Yeah, there's some...

So, yeah, the... The Rush Cosmos is currently for sale on eBay, so if anyone is interested, you can probably go in and find good shit that you want. It's like a bankruptcy option of a company, except instead of a company, it's what used to be the second largest space agency in the world that sent the first man into space.

Maybe the commercial space station companies can pick up on this. And they will like, all of their food, all of VAST's food for the first year will be like, you know, former Ross Cosmos packets. I found a Russian Herman Miller chair. Install the ISS toilet on the space station provides, saves him like, probably saves him like a couple million, too. Honestly, that's a great idea for them. I wonder if they've looked into that. That would be so funny. Man, I think they're selling me.

That feels like more than that. They're selling the light bulbs in the factory at this point, so I think there's... Yeah. It's not looking good. I mean, giving enough time, and you can probably go by the blueprints, and actually a full-blowing soil-blowing so-use from the Russian Space Agency for probably, I don't know, less than 10 million. But there's definitely opportunities here for any enterprising young American entrepreneurs I wanted to break into space tax.

This flag to Russia slip on this ancient to go bribes in Rostasian to give you all the blueprints and IP forever. You'll probably do it. You'll probably do it. Work for Shenzhou, then might as well work for you, too. This is very true. The show took a great turn. I'll say, Jake.

This took the best left turn of any show we've done. It makes me want to spend the last time of minutes pulling up eBay and trying to buy something right now on air. You probably could. I'm going to search for Rock. I am very curious. Is the spacesuit still for sale? Well, we know it's on your watch list. Log in, watch list.

Space toilet, yeah. There's a spare winter jacket. Oh, wow, this is a spare winter jacket from Soyuz MS-21, which is very recent for $500. Oh, shit. I want the Ukraine jacket. That's the one I want. Used. Good condition.

It's like, it's all right. You know, he wore it a couple of times. Oh, you want the Ukraine jacket. That would be amazing. Oh, my God, Jake. I would spend so much money on that as an art jacket. Oh, shit. You know that those were immediately burned.

This is the most expensive off nominal we've ever done, Jake. This is so good. Ed, you've ruined our life now. I'm buying one of these things for sure. For $350,000 in locating a U. was free delivery, by the way. You can buy a Russian Cretchen space suit

for unisurface operations. That one's not even completed. No way that one even works yet. I know it's three companies that would buy that tomorrow, so if it worked. This is good. This is a good section of eBay, man.

I can drop it. I can shoot you guys. You guys for which time of the internet. I can shoot you. I might try to buy this MS-21 jacket, Jake. It's located in Canada. It's already in Canada, man. We're going to tariff you on that.

Should I buy this jacket, Jake? You're going to get in trouble with the tariffs. Is this a legitimate jacket? Zerritsky. Cosmonaut Zzerbritsky. They haven't even flown this one yet. It's a spare winter jacket. Yeah, I mean, who doesn't need one of those?

You don't. I do. No. is MS-27. This didn't even fly yet. Right? Oh, no, this is April, April 2025. Roscosmos is so corrupt.

Ross Cosmos is so corrupt to studying stuff from missions in the future. What more can I say? They built a time machine and they're using the time machine to go from future Ros Cosmo, steal with your shit and sell it on eBay. That's a great reason.

I think it's still on the station. They're selling the future, basically, right? So, like, they put up, you know, the main purpose the jacket is merch and they sell it on eBay and then if it doesn't sell, they actually give it to the cosmonaut and then they fly up to space and use it.

Wait, they're selling, I found even better one guys. This one is a Sergei Rizakov jacket who's currently on the space station and they're selling his spare winter jacket. He doesn't need it, obviously. It's the same person too, the same seller. So what's going on? Click on that seller.

What's going on here? Colbolt 2000s? What's going on with Cobalt 2000s? I think that's a guy that was selling the ISS Floyd. Let's see if it's still over there. Look at this. He's got so many things. This is the guy, man. Oh, is that a signed photograph?

Ooh. Yeah, this is the MS-8. On the same guy, is Britsky. Signed. Look, this. Whoever Cobalt 2000 is my jam. In North York, Ontario, Canada, Jake. All right.

I'm about to buy some stuff from this guy's shop and see how it pans out. Look at this shirt. This is an amazing eBay store 10 pages of this Holy shit That is like the Gagarin Cosmonaut photo version of the department store

portrait you get with your cats like hovering above you You know like that one Can I do sort by most expensive? Oh, price and shipping here we go This is a $47,000 There we go Soyuz-TM, mere space station, extremely

rare, actual console in Ontario. This has to be like, is this like a museum that got liquidated or something? It's got to be or something, hey? Like, how does this guy have all this stuff? Here's a Lunar Lander Salute Space Station Periscope. Holy smokes.

God. That's so awesome. Man, I, oh, that one that one speaks to me. I would buy that if I had. an extra. Half a Bitcoin, Jake. Yeah, this gets expensive. These are pretty awesome, though.

All right. God, Ed's spending us so much of our money today. This is so problematic. It's good. Do they ship to Mexico? I'm just wondering. I don't know. Or do we have to go home and get it?

Yeah, you got to go home. Just don't tell them on the way back in. Yeah, do you think the Mexican customs are going to appreciate that? What is this? Listen, don't worry about it. It's just a lightly used salute space station, space telescope that was used to spy on the Americans in the Cold War from space. Don't worry about it.

It's only in good condition. Yeah. It's used, so it's not subject to import Texas. Jake, I got even better. I got the best item so far. The pre-flight jacket for Skaplorov's jacket for the movie The Challenge. The Offenominal favorite movie, The Challenge. God, this is good.

this is good eBay here. Ed, we owe you at least two minutes to plug some shit because of how good this episode has been. Are you hiring people? Do you want to have people contact you? Tell us what's up. What should we watch for? Got it. Yeah, of course.

So we are very interested in exploring Orbital Data Center applications. While we do understand that a lot of the out application for Orbital Dataster's are out in the future, we do want to talk about the kind of applications of space and computing today, particularly for large programs like Golden Dome, what we can do for poems like BWSA, and for the Department of War objectives. And another thing that I also want to shout out is that Madman on eBay, he was currently setting the Russian lunar spacesuit, because I do think if Aksion or who of the, who's on another guy that got kicked out of the...

Collins. Collins. Biles. Acyon. Okay. Acyon or Collins. If you guys hear this and you guys are listening to me, buy that Russian spacesuit, putting American flag on it and just sell it to NASA.

We need the art in the space suit. We need American sky on the moon. Good luck. Also an amazing troll. If we flew a Russian space suit to the moon and walked around in it via a starship, that would be amazing. Yeah, that would turn some heads, yes. We would turn some heads.

They'd probably be further along in NASA and McConnell Cours, to be honest. Yeah, I mean. Well, definitely further along in college. Put a nice black covering over it. Give it to George Santos. We'll be in good shape there. In the space suit. In fairness,

they've had some progress posted on their suit lately, so that's been good. It's true. We'll see. They're doing testing. It's not just a mysterious made-up thing now. I don't know. I'm still suspect.

I'm suspicious because it's big JSC, but we'll see how this. Where do you want people to reach out, Ed? I don't know if we got to that part. You got an email or something? Yeah, sure. My email, Ed at Atherro.com is my best email for reaching out to me. I check it.

Probably should check it more. Or if you're selling any high-quality Russian space artifacts, Ed at Athero.com. I can also refer you to whoever is, if you're sending the Soyuz capsule, reach out to me as well. I need a new office. Working in that would be amazing.

I would be fine, but I think, yeah. Jake would not. No, I couldn't do it. No. Well, Jake, I think we're off next week because it's Thanksgiving. It's Thanksgiving. It probably was yours recently, so happy Thanksgiving. You still celebrate that?

A month ago. Yeah, that was a month ago. Do you do that still in Mexico, or do you just... Sometimes. Sometimes I feel very inspired and I do a Mexican Thanksgiving fusion meal and I cook like mashed potatoes with habanero in it. It's really good. But I didn't this year.

Maybe next year's the next year. I'll do it when you come visit me. That's what I'm going to, how's that? I told you. It's in the, it's in the Calangelo family plans here. You can come down for Canadian Mexican Thanksgiving. How's that? Actually, sounds great.

I'm into that. I'm a pretty good cook, you know. I know. And the habanero thing and your weird honey stuff that you're drinking right now. How was that? Check in on that thing. Fantastic. Yeah?

Fantastic. Okay. Yum Bob. Yum Bob. Come on down. Yum Bob. Awesome. We don't know who's hanging out with us on the other side of Thanksgiving, but I bet it'll be good. And then we're only two weeks from the off nominees.

Yes, it's coming up fast. So we've got to... We have some legwork to do with the Australians before the off nominees, Jake. We've gotten no radio contact from Gilmore Space since we thought we might have the inside edge on that. They have an instant win available to them if they just follow some few short steps. Yep. They sure do. Man, Ed, thanks for hanging out, putting up with our bullshit.

Absolutely. We appreciate it. I've got this done, so it's good. See, everybody. Bye. Bye. Bye, everyone. Bye.

Ciao. Bye, two, three, four, five, four, three, two, one, end of death.