Filip Rezabek on the T-Minus Space-Cyber Briefing with Maria Varmazis

Filip Rezabek on the T-Minus Space-Cyber Briefing with Maria Varmazis

TLDR of this podcast episode:

  • SpaceComputer is an early-stage startup focused on building secure, end-to-end infrastructure for the "space internet" across both ground and space segments.
  • Space Fabric Architecture: this is our core payload architecture, which allows multiple tenants to run sensitive workloads on a single satellite using hardware-backed security without requiring tenants to trust each other or the operator.
  • Hardware-Backed Isolation: Space Fabric system leverages Trusted Execution Environments (TEEs) to achieve physical isolation and cryptographic assurance, providing air-gapped security that is difficult to compromise in orbit.
  • On-Board Key Generation: Private keys are managed inside an orbital key management service (KMS), and generated directly inside isolated hardware elements in space rather than on Earth, to ensure secure cryptographic verification and end-to-end software attestation.
  • Post-Quantum Readiness: The platform incorporates post-quantum cryptographic libraries, with an upcoming demo mission in early 2027 to test its hardware and software in space.

Watch the full episode on YouTube:

Welcome. I'm Maria Varmazis, and you're listening to T-Minus Space Cyber Briefing. In this show, we examine the evolution of cybersecurity in the global and orbital infrastructure that powers, protects, and connects our lives.

Hi and welcome everybody. As I said at the top of this show, T-Minus is all about learning how people are building and securing orbital infrastructure. Now usually I prefer to interview experts who can tell us all about the broad landscape, some of the specific challenges and opportunities ahead, but sometimes it is also illuminating to talk directly to the builders, the people who are actually right now building future secure orbital infrastructure and hear about what they envision and how they're making it all happen.

So in light of that, my interview today is with Filip Rezabek, who is the CTO and co-founder of SpaceComputer. His company has the explicit goal of building secure end-to-end infrastructure for the space internet as it's often being called, which will work across both ground and space segments, of course. So to do all that, they are working on a payload architecture called Space Fabric that enables multi-tenant workloads on a single satellite, all without requiring the tenants to actually trust each other or even the operator, all using hardware-backed security.

First, let's hear a bit about Filip.

So great to be here. My name is Filip Rezabek and I'm one of the co-founders of SpaceComputer and also the CTO. So trying to use my previous experience with building secure systems, kind of achieving what we try to do with SpaceComputer, which is to provide secure infrastructure for the space internet.

And as a part of my background, I am currently in the last phase of completing my PhD in the area of computing at the Technical University of Munich. So I'm currently living in Germany, where I originally moved to from the Czech Republic, where I was born. And I have been living here since 2017, doing my masters as a typical historian and continuing with staying in academia, but always interested in the industrial application and always keen to bring the research background, basically, to industry.

This is basically the part where, when we were co-founding SpaceComputer together with my co-founder Daniel, was a great thing, basically, of seeing how we can bring both of these folks together and now being actively building on the activities and expertise I have from the previous activities. Excellent. Well, thank you so much for joining me, Filip.

It's really nice to meet you. And I was telling you before we were starting to record that when I heard about what your company is doing, the technology that you all are building, I said, that is right on for our show. And I really want to learn more about it.

So please tell me about SpaceComputer and also Space Fabric, because the story for this sounds just really fascinating. So tell me more, please. Definitely.

So as a company, as SpaceComputer, we have been around for roughly two years. So we are rather an early-stage startup, and we are bringing to the area of space basically both components, which is both the ground segment and the space segment. And we are trying to provide a holistic end-to-end system, where the motivation for what we try to see is that currently, when looking at the space, we see there is a bit of a friction with respect to communicating to your spacecraft, but also at the same time, the spacecraft is usually built as a purpose-built silo that is operated by a single party.

Sometimes, of course, you can still have multiple payloads on board, but usually you don't want to be sharing the payloads with other people on board, which, of course, means that it's naturally bringing limitations to the scalability of the overall solution. And as a part of the vision that we want to focus on is to provide the infrastructure, the trust layer to enable the space internet infrastructure. And for that, we want to, of course, have the seamlessness of being able to interoperate with different type of parties, make sure that actually you can do it in a trust-minimized fashion without either trusting the operator, the other users on the payload, and also make sure that you can have a seamless integration with respect to the ground station and other components of the overall end-to-end system.

And as a part of the overall story is the space fabric, which is the architecture that we envision to deploy on the space component. And it's also accelerated by the Earth ground segment as well, which is mainly enabled hardware that will be deployed on the spacecraft itself, which we are currently working on as a part of the payload development. And this particular payload enables you to run on a single satellite your solution, your workloads, your sensitive workloads in a multi-tenant solution without needing to trust the other parties that have access to the same physical infrastructure in space.

And this is basically the part that we see like it's very important not to only get the, so to speak, contractual assurances, but actually you're also getting cryptographic assurance that your particular system that you're running is really secure and has clearly defined boundaries between the other users in the system. And you are also getting cryptographic proof as a part of the overall story. So you can have this end-to-end attestation from the spacecraft all the way basically to consuming the data or any type of other interactions as a user on Earth as well.

Yeah. Okay. So I'm going to ask if you can give me sort of a comparison to typically how payloads would be run right now, just so we can contrast with what you all are building to understand.

So at least in the world that we currently see that usually the one payload equals one user. And the reasoning for that, of course, is the sensitivity. Sometimes you don't want to have access or provide access to other parties because inherently maybe you are running your sensitive workloads, your sensitive code that you maybe don't want the other people to have the option to share.

And we basically want to find means on how you can actually share the particular access to the payload by two multiple parties without the need of trusting any of the other tenants on the same spacecraft. So this is the way on which of course can unlock many new use cases as well, because of course the barrier to entry later on becomes lower and you can basically easily also share the cost for having access to such type of infrastructure. And you can do it not only on the, let's say, as we know, for example, ride share options or they try to do this on the same spacecraft, but having multiple payloads that are not easily integrated together and don't really communicate to each other.

We really want to focus on it from the payload level. And of course, afterwards also scale it to the other payloads on the same bus system so to speak as well. Okay.

So something that I was reading that was very important to what you're building is the idea of physically isolated secure elements. Can you tell me a bit more about what that means here? Yeah. So as a part of the crucial and core component of the space fabric architecture is the reliance on the technology that is called Trusted Execution Environment.

And this technology, it's not something new that we are bringing to the table because it has been existing already for some time. We can see it actually in the majority of our phones. It's also currently being actively pushed by major silicon providers like Intel or AMD and NVIDIA as well as a part of their overall story.

And we are relying on this technology to bring it to space. This technology, however, if you look at the on-Earth deployments have several fundamental issues. Either it is the supply chain, it attacks the route of trust that is usually coming to the respective silicon manufacturer.

Or also what we see is a single point of failure with respect to having, for example, only a single reliance on a single route of trust. And the other part is the physical isolation. So in case there were very recently several academic papers, I think as you have physical access to the silicon that is equity based technology, you can actually compromise the respective hardware.

Which is, of course, if you consider sensitive computations and workloads, it's not ideal. And with respect to sending these components, not only the traditional one, but actually we are trying to enhance them by additional challenges that we try to mitigate, such as the supply chain attacks along the way. We are able to achieve the physical isolation provided once it's deployed in space, because it's very hard currently.

At least we see as of now, and I think for the short to medium term, it's going to be the case. It's very hard to go and get your respective satellite and be able to extract and physically compromise the respective infrastructure. So we are relying basically on this aspect of the physical isolation to provide very good security guarantees with respect to having basically an air-gapped system, which is very hard to physically compromise.

Tell me a bit about how the key signing works with your solution. My understanding is this is a sort of unique angle that you all are using. Can you tell me a bit about that? As a part of the process of getting this cryptographic proof, as I was mentioning, of the overall software stack that is actually running in space, you have to generate your respective key pair.

You have the public key that, of course, we are relying on public cryptography, that the public key can be shared to everyone. You want to, of course, share it with everyone. But the private key is the sensitive component that you don't want to leak to anyone, because of course, once you leak it, you have a problem.

And as a part of the overall story, the solution we try to provide, we want to make sure that at no point in time, we are actually generating these private keys on Earth before they are being shipped to space. So once you are actually in space and you have this perfect physical isolation, we actually have the process in place that will generate the private keys only on board of the secure elements that are on board of the satellite. And afterwards, we are using it, for example, for signing whatever information the users aim to provide.

But also, for example, we are using it to sign the information about the software as well. So, for example, what I mean by that, you can sign what is the operating system you are using, and not only that you are signing it for the aspect of you as an owner of the infrastructure, you want to have the confidence that you can actually create an attestation that can be shared with any other third party that, hey, this is actually the respective operating system that I can go and publicly verify that this is the respective software stack that I'm interacting with. So that way, it kind of provides to the user the assurance that they actually can see that this is the respective system where I'm later on comfortable to, for example, deploy my sensitive application as well, because you are building the attestation from the ground up from the basic components of the operating system, the software stack, all the way basically to the application layer as well.

Okay, how different is that from sort of the standard operating procedure right now? So the main difference with respect to that is that if you look at the standard operating procedure, you can start your operating system, but you are actually not getting any aspect of the hardware enforced attestation. So maybe you can measure certain aspects of the operating system by software, but it means that the software itself has to be trusted as well. So it's a little bit like a chicken and egg problem, right? Because I am measuring the operating system, which I have to trust.

And since we are shifting the layer of the trust, one layer basically below to the hardware, we are able to not only attest to the hardware state, but also to the software state. Let's take a quick break now. We'll be back in a few moments with Filip Rezebek of SpaceComputer.

And we're back now with my interview with Filip Rezebek of SpaceComputer. As we're talking about hardware, the question that is for anything related to space is, so what happens when we need to upgrade that hardware? Given how much cryptography is evolving right now, seemingly with quantum in the mix, how does your solution sort of stand up to that with, again, air-gapped hardware? Great question. I think the answer to it is actually we rely on a mix of software and hardware for upgradability.

So for example, what we can do, the security components we are using also support the transition to post-quantum cryptography. So if you assume that, okay, I want to upgrade the firmware that is deployed on the particular hardware component chip, I actually can go and deploy the particular upgraded firmware that is, of course, being signed and it's being verified and so on and so forth, to basically provide the guarantees. And as a part of the overall story, actually, the technology of trust execution environments can also help to ensure the software you are running is also providing additional isolation on the process level, on the software level, between the individual parties.

So you can actually also go and, for example, have your private key for this post-quantum cryptography secure and ready, you can actually run it also during the runtime, because you don't always need to store it on the chip itself, because you are having this additional runtime security assurances as well. But overall, we try to make sure that the hardware that is provided as the fundamental components, so to speak, has the post-quantum cryptography readiness path moving forward, but also we can do additional security, provide additional security guarantees on board of the runtime as well, where we can later on ensure the cryptographic agility as well for the solution as well. Really glad you mentioned post-quantum. 

I feel like that is the phrase that gets mentioned in the same breath as crypto pretty much all the time when we talk in the space realm. Tell me a bit more about that, please, you know, just looking towards, you know, post-Q-Day. Definitely. 

So what we see is, of course, in the design or space will definitely take some time, right? Even if we start to design the satellite, for example, today, it might take a couple of years because before they materialize and before they're really being deployed. And, of course, by then we have to already consider that quantum computers might be around. As an example, what we are, the current timelines we see on Earth deployments, we are usually talking about 2029, 2030 as the part where most of the aspects of the critical infrastructure will be shifting to post-quantum cryptography as well.

And if you consider the lifetime of a satellite plus the design lead time, sometimes, especially for large missions, you definitely want to consider post-quantum cryptography as a part of the solution. And we aim to provide the answer to that by ensuring that actually we already have the post-quantum cryptographic libraries in place that are currently following mainly the NIST standardization for both the exchange mechanism and signing from basically day one. And we are currently actively developing and also providing this accessibility to the users that can later on take these libraries and integrate them.

And what else is something that makes us really excited that these solutions, even though we are a young startup, we will be actually having a space heritage rather soon with our upcoming mission later on this year, where we plan to test all of the aspects of the hardware and the softwares that I'm currently discussing are actually tested and have flown to space. So this is something that we are very, very excited about. I was going to add that was going to be the next thing I was going to ask about.

I'm glad you mentioned it. I'm curious, what are you hoping to learn from this demo mission? So I think there will be a lot of components we want to unfold. So firstly, it will be the first mission that will be actually implementing the space fabric in place.

So this is something that, of course, is the fundamental building that we are relying on. So we really want to see that even though space can be quite adversarial due to radiation and temperature and so on, we can actually do quite deep and sophisticated cryptographic operations on a classic hardware in space, because this is something that is, at least to the best of our knowledge, wasn't done in such a fashion. And we, of course, want to make sure that all components are surviving the whole mission, of course, that there will be no bit flips, because, with respect to cryptography, everything is very sensitive to radiation and others.

So this is basically really just making sure that the hardware is robust enough. And of course, the other part is the whole aspect of the assurance regarding the protocol we are building on top of that, regarding the communication, the endorsements to be actually really attesting that execution is happening in space, and doing also a couple of POC missions that we are currently finalizing with our partners along the way. Makes sense. 

That's fantastic. And I'm zooming out a little bit from, you know, what you've been working on lately and the demo that's coming up. I imagine you all are incredibly busy right now. 

So thank you for taking the time. I can only begin to imagine. Given your company's mission, you know, essentially is securing and helping to build out space internet infrastructure, which is, I mean, just something that is so fascinating to me.

When I think about, you know, the vision that I've heard from many sectors in the space industry about what internet and space is going to look like when, you know, interconnected mesh, you know, networks, orbital networks, that kind of thing. I'm curious about what you're building, how you envision it will scale, especially on a constellation level. What's your vision for that? Great question. 

I think there are a lot of things to unfold. So I will try to keep it at least to the core components. So one part we try to take the barrel and the inspiration from is how the internet on earth has been developed. 

So space started more as, let's say, purpose-built silos, right? We had like a few labs that have certain activities and you had more private sector coming into place, building up your individual components. But eventually kind of the internet emerged by defining set of open protocols where everyone who wants to join the internet is able to communicate in the same fashion and try to find the right methods of abstraction so that they can actually easily integrate all of the complex systems together into more of a coherent system that is able to communicate to each other and provides all of the cool services and applications we see nowadays. So we try to draw a parallel from that and try to do something similar.

So we want to operate heavily in an open source mode. So that way you actually can have open protocols, open libraries where actually the users can go and deploy and use them for the communication. But also we see that the internet itself needs some developments and we definitely want to also, for example, consider some of the modern requirements that we see especially, for example, against images that are being generated by AI or information that basically are being very easily tampered with. 

And we really want to provide this end-to-end attestation and data provenance from day one as a part of the infrastructure we are building. So kind of thinking on how the development will be coming forward where, of course, providing smooth operations on open protocols can unlock many interesting use cases for the future while also considering security and data provenance as a unique feature. So that way you can have different parties easily consuming data without the need of trusting all of the individual parties involved in the process. 

That's fascinating. Filip, this has been super neat to hear about. So thank you so much for taking the time and what I can only begin to imagine must be a very busy time for you, Filip. 

All the best success to you and I look forward to hearing how the demo mission went. If you're interested, we'll be in the Silicon Valley Space Week later on this year, so we'll be definitely happy to catch up with some of you during the time. It's planned for early 2027, and in general, if you're interested in what we are building, follow our social media and hit us up. 

I'm looking forward to it. Thank you so much. 

And that's T-Minus Space Cyber Briefing brought to you by N2K CyberWire.



Follow along on X and LinkedIn as we get closer to launch.

SpaceComputer is building sovereign compute and secure communications infrastructure for the Space Internet.

Space is evolving from isolated missions into shared infrastructure. We're bringing technologies established in terrestrial computing, confidential execution, hardware-rooted security, cryptographic attestation, and key management, into space. Our goal is to make orbital compute and communications accessible as secure, interoperable services, without requiring every customer to build and operate their own satellite infrastructure.

Visit our website for more information.