Welcome back to the show. Today I'm thrilled to have Nick White with us. Nick is the vice president of Celestia Labs, the team behind Celestia, a groundbreaking modular blockchain network that's redefining data availability for blockchains. Nick, thanks for joining us. Peter, I've known you for so long and I've been watching this podcast as you've been developing it, so it's a pleasure to be on. you know, always had a lot of respect for you, so, and you have a lot of insights. So I'm excited to dig into everything. I am too. Yeah, as you mentioned, we've known each other for a long time and have watched you guys from the sidelines and Celestia is doing a great job. And so congrats to every, know, for everything you guys are doing as an entire team, you guys, a team, you guys are killing it. um Let's start off with the, the problem that Celestia was initially kind of formed around. Maybe you can describe for us that problem and in like maybe give us a sense of how big of a problem it was and then Celestia as a solution to it. Totally. So when I first got into crypto, which was in 2017, 2018 era, everyone was super optimistic about the future of blockchains. And there's so many ideas, not just bringing finance on chain or NFTs or there's just like decentralized social media. Like there was just so many things. But then very quickly we hit the wall of reality, which was that the infrastructure just could not support even like uh a fraction, even like a hundredth or a thousandth of like the vision and ideas that people had. So I've kind of been fascinated from the very beginning about how do we solve building the infrastructure that can actually support crypto applications at scale and at maturity. And so that's What I've been working on, that's me. We worked together at Harmony, which was trying to solve that problem. I then moved on to join Celestia and Celestia is solving that same problem, which is like, how can we build the underlying blockchain? That's the right fit for the end game of how applications will be built and how they'll be scaled. we need to have, you know, there's a few problems to solve within that. And I would just say, focus on two that Celestia really is targeted towards. The first is um just scalability. How do you build a blockchain that can increase its throughput in a way that still preserves the underlying verifiability and decentralization? uh so Celestia is solving that problem using a technique called data availability sampling, which was a genuine breakthrough in crypto scaling technology. kind of for the first time solved the big block problem, which raged in Bitcoin, then moved to Ethereum and uh frankly, you know, was never really solved. It was always just kind of kicked down the can down the road with solutions like Solana that just require more like higher node requirements and do make the nodes more efficient, but overall don't solve the underlying same core issue. um And then the second problem that Celestia is designed to solve is how do we build crypto applications in a way that they can be even more powerful and provide more features and better UX? um And that's where this idea of application-specific chains comes from, which originated in the Cosmos ecosystem. Celestia borrows a lot of inspiration from Cosmos. The idea there is that if you're building an application as a smart contract, contract... That's a very similar model to, for example, shared web hosting. If you're building a web app where you're building your application and running it on someone else's server, where they have defined a bunch of constraints for you and you fundamentally just can't actually, there's only a certain limited set of things that you can customize for yourself versus building in a cloud computing paradigm. where you can show up to AWS and say, this is how much compute I need. This is the exact sort of specifications that I want for my virtual server. And it gives you just so much more customizability. And so that's what an application specific chain is, is basically your own dedicated server where you can define the execution environment. can define, can, you know, you make it, you can make it private. You can make it, um, just tailor-made to a very specific thing like on-chain trading, like a club, like we're seeing a lot of that, and we can probably talk about that a little bit later. You can make the economics and the gas fees and fees in general better. You can use different cryptography that allows you to integrate more natively with a phone so that you can sign using things like FaceID and biometric stuff or integrate natively into the... the secure enclave on a mobile device. So all these things that get unlocked when you are building your own blockchain. uh so Celestia is, you know, the problem with Cosmos is that if you had to deploy your own L1 completely from scratch, that's a huge amount of overhead. But thanks to this other innovation, which I would pair as very complimentary with this data availability sampling that I talked about, is this innovation called roll-ups, which is essentially the equivalent of the innovation of virtualization that enables cloud revolution. That's a roll-up. Roll-ups are basically virtualization. where you can run a virtual chain without, so that means you don't have to build out your own data center or your own, you know, physical servers in your office. You can rely on something like AWS or in the case of blockchains, you can rely on something like Celestia. So that's, that's like the high level problem that Celestia solving. Scalability. and building really customized vertically integrated powerful applications. That makes sense. Let's go back to data availability sampling. Could you describe, maybe walk us through what that looks like, what's involved in that, and why is it a breakthrough? Yeah. So it sounds, so one of the problems with the term itself is just data availability sounds like data storage, which sounds very unimportant and kind of like a, whatever, you know, that's just a backend thing. And it probably is not very critical to the functioning of, of any application. Like it won't, won't really matter, but actually what data availability is, is central to what a, what blockchains are. and the guarantees that a blockchain gives you. So blockchains are essentially verifiable computers. So, and what you need to ensure that the whatever computation is happening is verifiable is everyone needs to agree on and have access to the, a shared set of transactions or instructions that are basically being fed into that computer. And That what that means is that everyone needs to have a shared view and public access to this. You know, set of transactions or instructions. That's what data availability solves. It basically is like a, place where you can post transaction data and get guarantees that it's been published and is available to anyone. Uh, that's part of the protocol and that has, you know, it's connected to it. So. Uh, you can think of it as like this magic billboard, essentially, if like, there's data that you need to make sure gets out into the world so that your application is verifiable. And Celestia is a service that does that. And so it's essentially the same thing as what any L1 does, except that most L1s do that, but they also actually execute the transactions on top of that. So they will Not just post and publish the transaction data, but they will interpret it and say, okay, these are the transactions and this is the output. Celestia's case, we don't handle that second part. It's just a raw log of posting data and anyone can layer on top of however they want to interpret that data. And that's what gives people a flexibility versus a standard L1 like Ethereum or Solana. It's like. We only speak this one language, which is the EVM or the Solana C-level VM. We don't, we're not able to speak anything else. So you send it another transaction in a different language or, you know, a different model and we don't know what to do with that. And Celestius basically says, we don't care what you want to post. You can post anything that you want. And it's up to you how you want to interpret it. uh Yeah, that's basically what data availability is. It ensures a verifiability and security of applications. And it's what NEL1 does. It's just think of it as block space. If we want to transact over decentralized rails, we need block space. And Celestia sells just raw block space. Now, you guys, Celestia started this, I don't know if it's a meme, but it's really a of a rebranding of sorts where there's the modular chain versus the monolithic. And I don't know if you're, I remember, I'm sure you remember, but there was a lot of Twitter spat back and forth between you and other kind of chains that would be considered monolithic. Mm-hmm. and this is all related to what you're describing as data availability, maybe describe for us what that means, monolithic versus modular, and how it's related to this idea of L2s and roll-ups and kind Celestia's role in all of that. Because there's really kind of two movements, right? There's the continuum on the monolithic side, chains like Solana, et cetera. And then on the Ethereum side, it's becoming more modular. Mm-hmm. this idea of interoperability becomes really, important. But it also has some trade-offs. So maybe we can talk about all of that. Yeah. So modularity is central to the Celestia design and thesis. And it connects to what I was sort of describing before, which is that Celestia doesn't execute the transactions. So what Celestia modularizes is basically something that's been bundled by default from Bitcoin to Ethereum, every Solana, every single L1 that's been built up until Celestia. bundles the data availability and consensus part, which again is that publishing and ordering the transaction data with the execution of those transactions. So all the nodes in the network have to do both of those things at the same time. And that's how the network functions. In Celestia's case, one of the insights of Celestia was we can unbundle those two things and uh just have a have the L one only do consensus and data availability. And the execution part can happen on quote layer twos or rollups essentially. And so that's, what you're, what's happening there is you are imagine it's like, you could have the, you know, think of an analogy, like a car. It's like someone was manufacturing cars in a way where it's like, okay, you have the chassis and the engine and they come together and they're bundled. And that's it. Like you just, you don't want to change the engine too bad. It's like already, you know, been integrated and that's how the car gets shipped. you know, sorry, there's no, there's no ability to change that piece versus Leslie is like, you know what we can just sell the base chassis and you can choose to bring your own engine to install on top of it. um And so you're kind of modularizing these two components. And that turns out to be, again, very powerful because it's what enables an application developer to have significantly more customizability over that engine component, whether that's choosing a new kind of execution environment they want to ship, changing economics, changing the cryptography, changing the sequencing rules and capturing MEV or like what HyperLiquid's done with adding cancel prioritizations. There's all these different, it just opens up a massive new design space where you can innovate in as a builder and make something that just frankly is not possible on a generalized monolithic blockchain. that's the other thing is like these, most of these monolithic blockchains are generalized. So they're just like the base model. And it's like a one size fits all, but one size does not fit all, especially Um, I mean, if you look around in technology, like there, there's always uh a desire to vertically integrate and customize. And so that's something that Celestia feels very strongly about. Now, the important thing is the other part of modularity is that not only, so like in, let's say in Cosmos, right? You can customize the engine, but in Cosmos, everyone also has to build their own chassis, meaning they have to build their own consensus and DA layer. in combination with that from scratch. And that also means that there's a, that you don't share this common infrastructure. So it's a lot of uh increased overhead to launch your own application specific chain. And also you lose out on the ability to have this shared security, which I think is a really crucial like feature for blockchains, which is that blockchains applications need to be able to verify each other. And so Celestia, that's the other flip side of modularity is not just that, okay, you get to customize the execution part, but also you can still have the benefits of this shared L1 that you get in Ethereum or Solana where it's easier to deploy your app and your app gets privileged access to the other apps in that ecosystem. What are some of the trade-offs? I mean, I get the difference between modular and monolithic. For an application developer, they have increased flexibility um and customizability on the modular side. But what are some, I guess, could be perceived as negative trade-offs from their perspective? Yeah, I would say the main trade off is that when you're launching your own chain, like a roll up is its own chain. um It's difficult for to move assets to and from that chain. And it's also difficult for that chain, the application to compose with other applications. And the reason for that is that blockchains uh like interoperability is still a relatively new field in blockchains. Like we know how to do it. That's kind of one of the problems that Rollup solve is they enable you to have this bridging that's extremely secure in theory, but it's it's proven to be much more difficult to build that and practice. And we do have ways of, that other issue with interoperability between blockchains currently is that it's slower, right? So, you know, if I, I, I have a, I'm doing some kind of DeFi thing on one chain and then I want that to connect to another chain uh and maybe be able to borrow against that position or I'm not sure. Some, some kind of cross chain action. It will probably like, I can do that action on this chain, but it might take a while for that then to be able to get over to the other chain to be received, whether that's moving funds or, you know, just verifying that I did some action on the source chain for the destination chain to be able to verify that and then take action. So then from a user experience perspective, things slow down versus the really nice thing about monolithic chains or these shared just execution L1s is that you get that composability for free in a very low friction way. And uh So that's kind of the benefit. But again, the cost there is you lose out on the customizability. And also there's scaling issues, which is like, if you're co-locating every application on one single computer, it's sort of like your laptop where if you have too many apps running at the same time on one computer, you run out of resources and things start to freeze up. And we see that happen on Ethereum. We've historically seen that happen on Solana. um those are sort of the trade-offs. And I will say, um What's very bullish on the modular front and the application specific chain and roll up front is that all the interoperability problems are known how to solve. It's known how to solve them and the technology to solve it continues to get better. Um, like, like every month. and so to me, it's only a matter of time before we get there to the point where the interoperability is good enough where as a user. you don't notice, you would never know that there's, these applications you're interacting with are on different chains, because it all feels like one unified experience. And that's something that Celestia is working on and something that a bunch of different teams across the ecosystem are working on. You know, specifically one big unlock is going to be ah real time proving, which is like ZK teams continue to make just insane uh leaps and bounds of progress. And so we're getting closer and closer every day. The intent-based protocols also enable a lot of very smooth UX. So that's what I would say is the main trade-off currently of Modular, but it's a trade-off that I feel like is trending in only one direction, which is it's getting better and better. And at some point we will, I think, hit future parity or UX parity with monolithic chains, at which point there won't be a trade-off. So it's question of how long will it take for us to get there. So far it's taken longer than I think most people expected. uh When I met with Jill Gunter a couple of months ago, she gave the analogy of the cathedral and the bazaar and made the comparison that monolithic chains are, it's the cathedral and then the bazaar would be the modular side. Do you agree or disagree with that analogy? I strongly agree with that analogy. Yeah, it's something that we've like thought a lot about at Celestia. And because like things, I think it's kind like a free market for infrastructure is one way to think about what modularity enables where it's like, so, think about like in history of Ethereum, right? What are the issues with a monolithic chain like Ethereum or, even now we're seeing this with Solana is that you have so many stakeholders and they all have different needs. They all want different features to get added into this shared computer, right? That some people want to, you know, add something with account abstraction. Other people want to, you know, change the gas fee for a specific Operation or remove an operation altogether or whatever. There's all these competing or some people want to make the consensus faster or whatever. There's so many different like kind of oftentimes opposing agendas. And so then it becomes very difficult to actually move forward. And in fact, you can't please everyone. Again, this is a, this is the issue of a one size fits all model is that whichever direction you're going to go in, you're probably going to screw over someone else. You can't please everyone. And then that's also what leads to the things just moving very, very slowly. And so that's why also why people have launched alternative L ones from since the beginning of time, because they're like, well, this will never get shipped in Ethereum. So we need to, we need to do this ourselves. Now, the beautiful thing is with rollups, you can launch your own L one. can change, you can take the EVM and modify it and add your, you know, whatever specific change you wanted. And you don't, you're not gate, gate kept by the core Ethereum dev like pipeline, right. Or the politics. So you get to innovate permissionlessly at a layer that you previously couldn't, and you can iterate faster. And all of a sudden you create, like I was saying, this sort of like, um, sort of free market for infrastructure where it's like, well, if this is better, we'll try it out on a rollup. And then if it wins. That's going to proliferate. but, like in a monolithic chain, you're, you're limited to only one running one experiment at a time. You can't try both ways and then see which one is good and stick with it. You can't like AB test. Like you have to do everything in theory and be like, well, we think that this is the right change and that's what we're going to go with. Um, so just like so much fundamentally better. And then the other beautiful thing too, the flip side of this. bizarre versus cathedral thing is that L1s, like Bitcoin, is one of the reasons why people love it so much is because it's ossified to a large degree. That's kind of bearish in some ways because they won't innovate, but it's also very bullish in that you know that it's a stable foundation, right? You don't have politics influencing it. um And the beautiful thing is when you remove a lot of the things that developers want control over out of the L1, like what Celestia has done, and you just focus on something that's very pure and well-defined, like data availability and consensus, you remove that risk and the governance and the politics from the L1. And the L1 can be very stable and ossified in the sense of what it does doesn't change. Thunderlight technology will continue to better. We can use scale of the blockchain. But like Celeste can be this very stable foundation that everyone can build on and trust that it's going to be around in the future and not going to rug them or change things from under their feet, which you can't really say for an Ethereum L1 or Solana L1 or any kind of monolithic chain. They're going to continue to change. uh I want to talk about customers and product market fit in a second. um I'd like to discuss the positioning of Celestia because I believe Celestia was positioned initially as very complimentary and helpful to Ethereum L1. um But I think things are transitioning to where it might even be competitive at this point. Am I mistakenly thinking that? I mean, I think that there's an argument to be made there um in the sense that I do think in an abstract way, every L1 blockchain is going after the same ultimate goal, which is to be this, what I call a security zone, which is essentially a place where it provides this base level of security to applications and this notion of shared security between those applications. so they can interrupt with each other in a privileged way. um So in that sense, every L1 is competing with each other. Now, but then like, it's a very big universe of different designs and is also extremely early in the entire arc of crypto's development. And so I think that it's not that useful, at least in this current situation to think of like, Celeste as a competitor to Ethereum, because I think they're going after, Ethereum started as a very monolithic chain and then they realized that modular is the right way to do things in the future. And they're slowly kind of like moving more towards modular, but they're always kind of, I think be kind of like modulithic. And I think just move slightly slower because of the fact that they're just bigger. They have a lot more. you know, stakeholders a lot more like things to worry about. And Celestia starting from the almost maximally modular point of view, building towards what we think the end game is based on the innovations of DAS and roll-ups and uh is just able to move faster and have significantly more scale than Ethereum. But it's like, there's still different value props. Like Ethereum is a really good settlement layer and like Selenium is not really something that Celestia is going after. yeah, so I feel like they're choosing different points in the design space um in a way that doesn't really make them directly competitive. And frankly, the other thing is a lot of these, so there are a lot of Ethereum L2s that are using Celestia for data availability because they don't have any other option. And in a sense, it's like, um so Celestia is not really, taking market share from Ethereum, it's more so an off, like a outlet for the excess demand that Ethereum can't meet because it currently doesn't have the scale to meet. But a lot of these L2s are still very much aligned with Ethereum and branded alongside Ethereum and using Celestia for DA because we're the best alternative. But it's not like, I don't consider that as like stealing market share from Ethereum. Yeah, I think that's a good framing that, you know, Ethereum, there's so much demand that Ethereum can't meet. And so Celestia, you know, handles that and is able to meet the excess demand. um Let's talk about that. um Like, how do you think about customers? You know, you've mentioned rollups and L2s using Celestia data availability and maybe share with us kind of like what, guess, where things stand in terms of your, of Celestia's customers. um and what product market fit looks like and has Celestia reached product market fit? Yeah, I think that um Celestia has a very early version of product market fit in the sense that there is a lot of demand for people to deploy application specific roll ups or just roll ups in general. And Celestia is the best data availability on the market, the best solution for that on the market. And so we are kind of the default place where if you're launching a new rollup, you choose Celestia as your data availability option. Now the issue is that for Celestia, think to have genuine product market fit or for it to really feel like, okay, this is, this is really going to work. We need an application on top of Celestia that itself has product market fit to some degree. Right. And I mean, ironically, the entire space is. There's a very limited set of apps that have really hit that so far. And so that's, I think that's one of the challenges is like whenever you're building infrastructure, it can feel like you're dependent on, you know, the actual adoption of applications for your infrastructure to have a real, real adoption. But Celeste is really a bet on the fact that we will see massive adoption of blockchain applications. And it's only like a matter of time. when that happens, Celestia will be there to provide the block space for all that verifiable compute. where we are today, we have a lot of, we have, I think, over 50 roll-ups deployed on Celestia, which is pretty good, given that we've been live for a little over a year and a half. uh Some of the names that people might recognize are teams like Abstract, uh which is a consumer-oriented L2 uh built by the team. behind Pudgey Penguins. uh And also, uh Athena is deploying this really exciting RWA and so stable coin institutional DeFi chain called Converge uh in partnership with Securitize. um So those are a couple of the larger chains that we're really excited about going live. We have, uh aside from that, a bunch of DeFi and RWA chains alongside that things like Derive, which is absolutely crushing it in the on-chain options area. I think they're kind of the hyper liquid of at least options trading. Things like Plume Network, which is RWA focused chain. have like 200 million TVL. um And then we have a bunch of a whole like list of there's a stable coin payments app. There's Noble, which is its own. So like stable coin, yield bearing, stable coin issuance layer that's building a application layer on top of Celestia. So there's like a whole, it kind of like goes out very far and wide from there. We're really, I should also mention Hibachi, which is actually trying to compete with hyperliquid. So one of the, one of the application categories that has reached product market fit that Celestia is very bullish on. And we think Celestia has a chance of playing a very big role in is this uh sort of on-chain which was central limit order books, is essentially you were now able, the infrastructure has now gotten to a point where you can build a decentralized exchange that gives you feature parity and like basically the same UX as a uh centralized exchange like finance. so hyperliquids proven this out. their volume is just growing massively. And now there's a bunch of teams that are. following that and trying to build and innovate even more. um And so we have a, we're working with a handful of teams there, Hibachi being one of them where they, one of the things that they've been invented on, not only is there faster, lower latency than hyper liquid, but also they add a layer of privacy. Again, that's something that you can do when you have application specific control, you can be like, Oh, I want to make sure that the, um the user data trades are private. rather than being published on chain because that's something that happened with hyper liquid where this trader James Wynn claims that he got, you know, his, his position got hunted and that's why he got liquidated. But clubs are something that we're very bullish on as like one of the first. Like crypto apps that have really hit meaningful product market fit are, and are going to grow significantly from there. And they're, they need a lot of DA because every trade. that's like a hundred bytes or 150 bytes or whatever needs to be posted on chain for some of these exchanges the way they've designed. So that adds up to like just thousands and thousands of TPS. That's exciting. When I think of product market fit for Celestia, I've worked with a number of L2 teams and every single one of them use Celestia for data availability. And I think that speaks pretty highly of Celestia as a service. um And I think there is, I think, really meaningful product market fit from that perspective. That L2s roll ups, they use Celestia underneath. And I think that's huge. um to your point of Celestia itself creating an application that uses Celestia, I think that's quite interesting and I'd love to hear more about that. um And I think Hibachi is very exciting and I believe they're using Celestia underneath as well as Sysync to help with the privacy stuff, right? um And so that's actually quite exciting. um Tell us about, Celestia has saved these rollups a lot of money. m and which is very meaningful. Maybe give us some numbers around that and why that's important. Yeah, well, especially in the early days before Ethereum sort of shipped EIP 4844, which created the sort of blob market for Ethereum blob space. So it expanded Ethereum's DA capacity. A lot of the teams that we were talking to were spending millions of dollars annualized or even monthly on just paying for data availability on Ethereum. So that was when there was only call data and there was just not enough of it to go around. you know, the roll-ups had to pay that much to Ethereum. And so it was very difficult if sometimes, well, first of all, it eats into their profit margin and also in some cases made us those like unprofitable or like a, just a massive cost to just run their infrastructure. And so a lot of teams migrated to Celestia around then and immediately were saving, you know, millions of dollars that they could either pass back to their users as lower fees or through some kind of incentives or that they could keep for themselves um and, you know, reinvest into further development of their product. I, know, eventually Ethereum shipped for a four four and the cost savings compressed, but they're still there. So I think Ethereum, think Celestia is now about 40 X cheaper than Ethereum even post-4844. So that's still very meaningful. Now the thing is, this is so then the funny thing is there's other DA layers who in the past had been saying like, well, we're gonna do DA that's another 50 times cheaper than Celestia, right? And a lot of people were mistaking this idea of like, well, you should always, like anything that's cheaper is always gonna be better, right? But. That is only true up until a point, right? And so like even right now, I would say that Ethereum for Ethereum blob space, it's cheap enough that I don't think the incentive is to necessarily move to Celestia, even though it's 4DX cheaper. For some use cases it is, and we have actually quite a few builders who do move over. But if you're a base or an arbitrum, you're already making enough money that the cost that you pay to Ethereum for DA It's such a small fraction of like your overall operating expenses that it doesn't really matter. It's like, what's, what's really, what's the reason for us to migrate to something that's cheaper? Not now a lot of smaller teams. It is a very meaningful amount. So those teams are moving to Celeste cause it matters. But again, like after a certain point, once, once the cost of using the DA is a small enough fraction of your overall spend, feel like it's, it no longer matters so much as these other things like. How reliable is this DA layer? How easy is it to build on? Is it integrated with the roll-up framework that we want? ah Is it integrated with the roll-up as a service team that we're working with? um There's like a whole set of other considerations that become significantly more important than just raw cost. And that's also where Celestia shines, is we have, I think, the best developer experience. We have the most integrations. And we have uh just a really like strong brand and, and reliable, like highly reliable, like service. that's, that's what people care about. Like reliability is a big thing because it's like, you don't want your application to crash because like the underlying DA layer went down. uh And so that's another thing that Celestia has as a first mover kind of being one of the longest running DA platforms. Yeah, it's been said that the first in the category is going to be the winner and they will own a really meaningful share of the pie. And I think Selesha is one of the shining examples of that in crypto is that it was the category creator of data availability and its brand is super strong. think distribution is really great. Maybe let's talk about distribution. I believe uh one of the main ways that Celestia is kind of distributed to, uh or is available to developers is to roll up as a service uh providers like RAS. um Is that correct? And what are other ways that developers are learning about Celestia? Yeah. So roll up as a service. for people who don't know, roll up as a service provider is sort of like, you could think of them as a sort of system integration layer where like, or like a white glove service that makes it so that if you, if your expertise is not in deploying infrastructure or even choosing what infrastructure to use for your application, you can go to these teams. they know the full suite of products. They have built out uh the technology needed to deploy these things in a very robust way. And you pay them, there's different costs, sort of like structures for this, but you pay them for this service and it just makes it very like turnkey and easy. uh And it kind of does feel like if you go to, for example, Conduit or... and one of these like Gelato or Caldera, a handful of the names of the teams that we work with, you go to their pages, it looks very similar to AWS in the sense of um you have this interface and it's like you can click a few buttons of like here's the execution, like roll up framework I want to use, here's the DA layer I want to use, some other customizations and then click deploy. And so it's like a no code tool for deploying your own chain, which is pretty crazy. And that's always been the sort of vision for modular blockchains is like, make it as easy to deploy a blockchain as it is to deploy a website. And that's how we unlock a lot of innovation that is currently held back. anyway, role of as a service teams are a big way that we get distribution because they're out there kind of like selling on behalf of Celestia to teams. But, you know, I've been surprised that a lot of the teams that are, especially the really high value teams, they're fine with running their own infrastructure and hiring people or, you know, getting the right, you know, talent to deploy their own infrastructure. Now they still don't want to deploy their own L1. That's like way too much work, but they're very happy to build things. themselves a little bit more directly. um so, and even there are a lot of teams where they don't even want to use a roll-up framework. They want to do a completely custom thing from scratch. And that's also what gives them even an even more higher degree of customizability than like you even using like a pre-existing roll-up framework like Arbitrum or what have you. And so I'm actually seeing a lot of of highest quality, what I'm saying is a lot of the highest quality teams coming directly to us. with their own thing. And that's actually another way that Celestia shines is that we have an extremely good solutions engineering team that has done a lot of the, has the deepest expertise in this technology stack, not just Celestia, but all these different roll-ups, all the ways in which uh You know, we can build these things and we're extremely good at giving like one-on-one engineering help to make integration successful. So we work, so for example, um, abstract, we're putting a lot of effort there into completing an integration. Um, as an, as an example. Um, and so like, we have a lot of, I would say like edge there. And I think that that matters a ton because these high quality teams, want to. They want to work with the best too. They want to work with the best engineers. They want to work on the, with, you know, want to use the best product. Um, and it's not about cost so much as it is like, will this platform and this team that I'm working with help me to succeed? Which really speaks to the strength of the brand and why having a strong brand matters so much because it acts like a magnet for new customers. And that's really what I'm seeing from the outside. When I speak with L2s and roll ups, inevitably they end up choosing Celestia because it's such a strong brand. um Let's kind of pivot to a couple of spicy takes that you've said on X. And I'd love to hear your thoughts on that. um So you tweeted that EVM and SVM were just detours on the way to ZKVMs. um Can you maybe, you mentioned it a little bit, but I'd love for you to expand on that. Yeah, sure. um I think so EVM, SVM, that's Ethereum virtual machine and the Solana virtual machine, which are currently the two most popular ways of building applications. um And essentially those are, as I said, of like constrained environments in which to build. Like they make a lot of decisions for you that are just baked in and That means that you're, yeah, you're as a developer, you're, pretty limited. And there also could be a lot of stuff that you don't need. It's like bundled in there and you're just, it's just there because it's there. That's like the standard package that you get. and what ZKVMs are is essentially a way to, um, you can, you can write an application just the way you would write a normal. Like computer program, which is like just in. Like uh a language that everyone else would use like Rust, for example. And then that can be fed into a ZKVM, which makes it verifiable and proven. And so I think it's just a, frankly, a much more natural fit to how developers like meeting developers where they are rather than trying to force them like, Hey, learn this new language. Uh, like learn solidity, learn this whole new thing. And then, and then you can build applications instead of it's like, Oh yeah, well just, mean, whatever language you want to write it in, do that. And then, you know, the ZKVM takes care of the rest. So it's just like a much more straightforward developer experience, I think, and it will just lead to people building these verifiable applications in much simpler, lower overhead ways. Um, and like, the thing is like the, effort that went, the reason we couldn't get there right off the bat is that like building these generalized ZKVMs is like an extremely hard engineering problem. Um, and making them flexible, making them super performant. Um, there was just a huge amount of innovation that had to happen. and that's frankly happened much faster than I think anyone would have imagined. Like if you rewind five years to 2020, even the people who are the most bullish about ZK. I don't think they would have thought that we would get to where we are today where we have these very generalized EKVMs that, you know, you can like almost verify anything inside of, and that they're extremely performant at the same time. um And like the trajectory of how fast they're improving is like very, very exciting. Yeah, I remember working with a couple of ZK teams and I mean, they were still at a point where they had to create their own circuits, right? And then teams like Succinct and ZK Cloud and RIS0. Now it's like really easy for developers. Like they've made it really easy for developers. They've abstracted a lot of the complexity away. And now it's like ZK is very... um it's uh like down to earth. Like they've made it kind of like actually kind of sexy. Which is great. and the thing is too, by the way, that this is extremely bullish for the modular worldview because like, now it's not just, it just makes it so that the amount of creativity and flexibility that the developer has at the disposal is like that much more massive. And it solves this interoperability problem where if you can, if every app is just written as a ZKVM app. then it's provable by default. And so it becomes very easy for you to take this application and then plug it into another application uh because they can verify each other very easily with ZK proofs. it's just unbelievably bullish for modular. Awesome. Well, Nick, this has been a really enlightening conversation. ah Thanks for breaking everything down and then speaking to us from first principles. um You know, before we get off, what are some things maybe you could share with the community that you guys are focused on, uh very, you know, prioritized in building? ah What can you share with the community? So there's a lot of really exciting things happening in the Celestia world. So I'll try to just quickly summarize them. One is the whole Clobs on Blobs thing, which we talked about, which is we're working with a lot of teams that are building these Clobs, these next-gen Clobs. And from that, we foresee a lot of demand for block space. And so we think we're at an inflection point for Celestia, for the demand of Celestia's block space. And so it's going to be very exciting to see how that all plays out. So people should stay tuned on that. uh We're also working on some big economic overhauls. So uh very soon we'll be shipping uh an upgrade called Lotus, which decreases the inflation of Celestia. It also makes it so that Celestia, the TIA token can be bridged much more easily um to other chains and roll-ups in the ecosystem. like the TIA token so far has kind of been trapped on chain because it only has IBC support. So. That's another big one that's coming out soon. have a massive, the next upgrade that's coming out is going to be a massive upgrade to our scalability and our throughput. it should get us within the range of, um, 20 megabytes per second. Um, eventually first five megabytes, but then with further, like on, on chain governance, we can upgrade to 20, which will be huge. and we're also exploring this idea called proof of governance, which people should check out where we think we can actually. reduce the issuance of TIA from where it is today, which is currently around 7 % to uh roughly like 0.25%. So what that means is like, if you think of these protocols as these economic engines, the issuance is sort of like a drag on the economy. It's like the deficit, like, you know, people are paying attention to politics. It's like, uh You can have a surplus or a deficit and the amount of spending is linked to how much inflation that you're, you know, putting out there of the token. if, you decrease inflation significantly, like down to 0.25%, then the, uh the amount of revenue you need to generate a surplus, um, is significantly lower. So we actually feel like this really increases Celestia's chances of hitting profitability in the near term. So that's going to be really exciting thing to pay attention to. Awesome. Well, Nick, thank you so much. Those are exciting developments and look forward to you guys executing uh and you've got my support. Thank you. Thanks Peter, it great chatting with you as always.