Teemu Paivinen, co-founder of ZKcloud. Welcome to the podcast. Thanks for having me. Yeah. So tell us about, well, before we get into ZK Cloud, we'd to learn about you and your origin story. You've been in the space for, in the crypto space for quite a while and have advised a number of well-known projects, including Dapr Labs. Tell us about how you got in and why you stay. Uh, so that last part is a good question, right? Every, every few years, uh, I have to kind of rethink my justifications and reasoning for staying. Um, so I'll, I'll get to that in a sec, but yeah. So I, I, um, discovered Bitcoin through a friend of mine in college. This was back in like 2011, uh, or maybe even a little bit earlier than that. I don't know, I got really excited about it because I was kind of headed towards a traditional financial career. was sort of, I guess you could say sort of on track for like investment banking or private equity or something like that. And I had started feeling like a lot of the structures that people that have been put in place in the traditional financial system were just not that. They weren't quite right. There were a lot of very twisted incentives. Um, and, and there was really no way to change it. Like you, you know, no one can actually change the financial system fundamentally anymore. I think it's very set in stone. And so I saw Bitcoin as this kind of opportunity to try other things. And the only way you could really do that is, by being decentralized and, and censorship resistant. Um, and so I got. Excited about that. I started one of the early Bitcoin exchanges here in Europe. Um, run that for a little while. Um, then ended up selling that to, to one of our competitors and, and then kind of, and that, well, uh, was sort of bouncing around and then Ethereum happened. And I was like, Holy shit, this is, this is a whole new thing. I, I, It's worth for any sort of like OG Bitcoin people who are listening, it's worth mentioning that when I first read the Ethereum white paper, I was like, this is never going to work. And lo and behold, it did work. And so was totally wrong on that. But once sort of the first application started coming out, you know, we got like these early, like something that kind of looks like a dex. We started getting ICOs for governance and like all kinds of things. Um, I got quite excited and kind of dove, dove into that, um, and spent some time working with, folks like open zeppelin and then dapper labs who back in those days was very Ethereum focused, right? They, like, this was prior to crypto kiddies launching crypto kiddies was on Ethereum. It was, I think you could fairly say the first big, uh, NFT project. Um, Um, and so, um, was, was doing that stuff and doing some angel investing. um, and then in 2018, I, I started a company called equilibrium, which is now equilibrium group. Um, and there, the idea really was that the actual expertise of, of building a blockchain, uh, was at that time not available very freely. They were like some. core teams and foundations that had that expertise. But our hypothesis was that there was just going to be this explosion of all kinds of blockchains and a lot of innovation in different directions. And so there would be a need to have that expertise available outside of those, those couple silos. so equilibrium focused basically on just building the best sort of low level engineering team for blockchains that we could. And so we've done that. And we, you know, work with or have, sorry, have worked with like, basically everyone. So we have all the old school names like Ethereum Foundation, Protocol Labs, Polkadot and so on. And then also sort of the new school or this latest generation. everybody from Starknan, ZKSync to Celestia and Axelar and then things like that. So. ZK cloud is a project that started out being called Gavilog. Cause we didn't really know who we were going to call it. Um, and Gavilog is the development company was spun out of equilibrium. Um, and I spun myself out in the, the process, um, because I think the, uh, you know, we can get into kind of like why and so on, but basically I think the ZK opportunity, um, is very long-term and we'll come to sort of define blockchains as a whole and maybe other stuff as well. But on the blockchain front, I'm fairly confident that it's going to be like the underlying technology for basically everything at some point. It will take some time, but that's, think, where we're headed. I felt like we need to get in there and try to do our best to... to help move the sector forward. Let's definitely get into that. Before we do though, I'm curious about your time at working at Open Zeppelin. They've become kind of like the standard for almost all smart contracts. And they've really set the standard for a lot of the kind of key pieces in smart contract development. What kind of, I guess, what lessons did you learn from your time there? as you built out equilibrium and as you, it sounds like you worked as a kind of a software development shop for other blockchains. Yeah, yeah, exactly. for foundations and projects and stuff, it's like, looks just to sort of clarify, it looks quite a lot like traditional software consulting in some ways, but then in other ways it doesn't. So a lot of projects are like extremely long-term, like we'll be the primary builder and maintainer of like one of the main clients. And so the actual relationship is more... Well, it's very long term and it's much more like a partnership and we're sort of invested in the network success. And so it's quite, quite, quite different in some respects. But so I need to think a little bit about what I should say and what not, but so when I met the Opens up and Founders, Um, I was actually, I had an idea to like build, uh, basically, uh, a custodian. So this was maybe like 2000 end of 2017, 2018 there, basically were no custodians. So this is a time when, even though it's not that long ago, feels, incredibly long ago, but, um, there were like the first kind of token funds, like probably like polychain had just. sort of exploded already, uh, or, was in the process of, exploding. Um, and there were new funds coming in and sort of that was happening, but there were no qualified custodians, just zero. Um, and so my thinking was that, uh, like someone should build one. And so I was kind of exploring that. And then I met the open zeppelin guys. Um, I think I was introduced by the placeholder ventures folks, Joe Monegrohe and Chris Bernisky since they were kind of the leading edge of security on the Ethereum side of things. And so that might've been a natural fit. ended up not pursuing that project and instead just kind of like helping the open zeppelin guys. And so at that time, I, how would I put it? I came in as like a fairly experienced entrepreneur, um, and they were first time entrepreneurs. Um, so I don't know that I like learned that much about how to sort of build companies necessarily. but I did see, I think what was most impactful for me was seeing how companies and how like the culture was different when you have a completely different place in the world. So all of my. company experiences in Europe and partly in the US. And then when you go into Argentina, just like the whole thing is different. Like Open Zeppelin was one company, but they were in a building which had four companies and all of these guys were friends and they were hustling together. Like it was a very different kind of relationship between both the kind of employee and the employer as well as sort of these peer groups. Um, and so I, I found that to be, that to be very inspiring where they had kind of this little mafia going that really supported each other. Like Decentraland came out of basically that same building, Nomic, um, foundation, which is the guys that build hard hat, which is sort of critical. Ethereum tooling is from that same, same building. and, and so I think it was mainly kind of like a cultural. thing that I took away from that and then trying to create sort of non transactional incentive aligned relationships, if that makes sense. Yeah, I asked because in an emerging industry like blockchain, we lack, really for a lack of a better term, common libraries, standards, everything's brand new and we're trying to figure it out. But some of the opens up lens coming out as one of the key standards that almost all smart contracts follow now. And that's pretty cool that it kind of came up organically. And over the next years, we're going to see more and more of that. I was speaking with an interoperability protocol, and they're in the middle of having this kind of a consortium and discussion with other interoperability protocols of how to have a layer of standardization that they all follow, yet have space for more bespoke customization. And so it's happening right now and it's quite interesting. And so I was curious about your time there. Maybe like just quickly. I, I think one of the super interesting things about opens up Lynn is that they kind of made these sort of standardized open source contracts. Um, and they didn't really have a business model. Like they didn't even know how they were going to monetize it or, know, anything. and this is, I'm not sure if this is unique to blockchain, but I think it happens more in blockchain than maybe other places where you just kind of build something useful. uh, surprisingly often monetization kind of works out in some way. Like you, need to kind of figure out eventually. they, you know, came up with a whole bunch of different things. Now they, as I haven't really been following that tightly, but they have like a whole product suite, but really like the first thing they caught like figured out was like, if we've like, because we're the maintainer of these contracts, like people trust us and security. And so then maybe we can do security audits for these people. Um, and sort of built up from there, but it really didn't start with this, like. Intensely commercial mindset, which I think is fascinating. And I think there's something to, to be learned from that. Yeah, I think the lesson is you add value where you can and then eventually you find a way to make a living out of it and that's certainly what they seem to be doing. Let's get into ZK Cloud. I'm curious about the, would love to learn more about it, what you guys are up to, but let's begin with the problem. What is the problem that ZK Cloud aims to be the solution to for? Yeah, so there's kind of two. One is sort of a very general problem that we're trying to attack very head on, which is just the cost of ZK proving. so sort of anything that lowers the cost of ZK proving essentially makes rollups, any ZK applications, et cetera, more viable. as businesses or ecosystems and more competitive with other strategies of approaching this. And so I think that's super, super important. The other thing which is more, it's more sort of like a working backwards from the future kind of thing is decentralization. so ZK proving inherently is a sort of like, It's very easy to just kind of let it be done by a single entity. If you think about it from the perspective of a ZK rollup, you just, you know, you're trying to make this whole thing work. You're trying to build a community. You're trying to do all this other stuff. And like, you're not sure if you should be decentralizing the sequencer or, know, before we announced people weren't really even thinking about decentralizing the prover. et cetera. And I think it just gets kind of forgotten a little bit. and the, the problem with that is that like, is actually a mission critical function. And if that remains, uh, as a purely centralized function, and let's say most of Ethereum transactions in a couple of years time happen on rollups of different kinds, which I don't think is, is outlandish at all. Like I might even be the case if you count base. then you are suddenly in a situation where basically the entire Ethereum ecosystem is not censorship resistant at all. You can shut it down. And so I think long-term this is an important thing to solve and to not let it go the way it goes. like without interference, like we shouldn't just let it be and see what happens. because the end result might be one is very hard to work your way back from, because then if there is a single entity, then you kind of have this situation where that entity is probably making a huge amount of money and doing that. Um, and they have the most ability to compete on cost and they have, um, the most capacity to do this. And so it's very hard to unseat them. It's very similar to like the cloud sort of web to cloud market. dynamics, think we're like, if you want to start a new, you know, cloud provider now you need probably you need like a hundred billion that you're just willing to like spend, uh, with the uncertainty you still might not work. Um, and so we want to kind of avoid that situation. so ZK cloud is essentially designed to be both like super cheap in the here and now to make. as many things viable as possible. And it is generic. we can do proving for whatever use case there is basically. You could run any of these ZKVM, et cetera on the network. And then also offer sort of like a default decentralized option so that if we just all use that, then we don't really run into this issue. point and it sorts itself out. So that's the problems we set out to solve. Along the way, we also realized that actually decentralized systems, when you're generating zero-knowledge proofs and you don't need to re-execute everything that happens, Like Ethereum basically functions by way of every single node in Ethereum doing the exact same thing in parallel, which is kind of insane when you... think about it. So all of Ethereum can be reduced to a single computer in terms of the actual computations it performs. So when you are using ZK, you don't need to do that. You can actually start doing different work in parallel. I kind of lost my chain of thought there, but anyway, that's, yeah. Yeah, it's really crazy when people kind of reduce the blockchain to like a database. Like in a lot of ways, a database is actually more efficient, right? Because you don't have to reproduce every transaction every single time. And that's really the beauty of ZK, that it allows us to kind of move forward rather than having to look back and do the same things that have already been done. exactly. Exactly. And I remembered what I was saying. So like, because you don't need to do that, you can actually sort of find the cheapest ways to do things all over the world. And they're not all going to be the same. Like there's going to be someone, you know, building a data center next to a big dam or something where the energy is free and they have fiber and like, it's just all optimal. And finding those things in a centralized fashion would be very difficult, especially if you need to scale up very quickly. so I think decentralized systems and sort of incentives to bring hardware capacity online are actually a very powerful way here to compete with centralized systems. And I think they can actually be considerably cheaper. That's something we've started to prove out, but like are hoping to prove out a lot more. and that just simply happens because if you're Google, have to, you know, have a bunch of people looking for like some specific location or instance. and if instead you just have this grassroots movement to find, they don't need to be, you know, you don't need to run a whole massive data center or computing cluster. can just be a couple servers, but like the setup is. incredibly cheap, but still reliable. You can find those pockets all over the world and you can find huge amounts of them. So bringing that together, I think is a way to deliver like order of magnitude, lower cost structures. It's kind of that, I don't know if you remember sort of the file coin original vision was basically like everyone at home will just like mine. Filecoin with their hard drive and it obviously didn't work out that way but it's sort of like that same ideology. think like we don't think everyone at home is gonna do this but there's certainly a middle ground between your home PC and a massive data center. And I think there's a lot of that compute that's very underutilized all around the world. And so if you can bring that online. you know the cost structure on that compute is non-existent and And if you make ZK proving basically free That opens up a whole bunch of interesting things we can do both on and off chain in aetherium So let's talk about the, I guess what the market looks like. It sounds like these, you've got a two-sided market of kind of app developers that need to use ZK Cloud for ZK Proving, for proving, but then you've also got node operators, is that correct, that kind of provide that service to developers? Is that correct? Yeah. Yeah. So you have, mean, they're, yeah, they're usually like roll up teams are like on the other side. There's some other use cases, but rollups are like the obvious kind of easy to understand one. And they want to generate, or they need to generate proofs consistently. And they don't really care sort of who does it as long as it happens on time, every time. And is cheap and is sort of redundant in some sense, right? And so their biggest fear is that the prover just disappears or goes down for a while that this causes massive problems all up the stack in their roll up. And on the other side, we have people who are running hardware. so we basically have like the... software that runs all the provers like the ZK cloud node software handles everything. It utilizes your hardware resources and sort of deals with payments, does all that stuff. You kind of just spin up your machine with our software and get going. Got it. if I'm a layer two and I need services like data availability, for example, and I would typically use a roll up as a service provider like like alt layer or conduit, would ZK Cloud be one of the options in as I spin up an L2 and I have an option for ZK proofing, it would be I'm thinking in terms of distribution, like how would you be, how would ZK Cloud become part of the stack? Is that, is that typically how the distribution would work? Yeah, so we work with basically all the rollup as a service providers. So I think that is exactly right. There's also the actual sort of main. So that applies to things like polygon CDK chains, elastic chains, scroll SDK, things like that. But then there's also like the main. Chains themselves like scroll and stark net and ZKSync and, and, uh, you know, as tech hopefully soon and so on. and, uh, and then kind of there's a, maybe a third category, which overlaps to some degree, um, which is kind of the ZKVMs and what people end up using them for. And that those might not be like, it seems like they will probably be used for rollups as well. Um, but there might be some other, uh, things people come up with. And so a lot of the, you know, ZKML and, and CK co-processors may these projects that are mainly trying to, there's some like, I don't know, like some reward calculation that just costs a huge amount of money to do on chain. Um, oftentimes they'll then go to these types of providers and be like, okay, can you compute it off chain and then just deliver a proof to lower the cost? Yeah. Yeah. I mean, it's like, there's some spectrum of things that are currently already like that. But then I think those sort of more exciting long-term thing is like, okay, if what if just nothing needs to be designed with cost in mind, like what would people come up with? Because everything in Ethereum is like super cost focused because it's been so expensive. Yeah. No, you mentioned a polygon, myden and ZK or scroll. And also Aztec. I had written on my notes that these were potentially competitors, but it sounds like there's some level of cooperation. So I'm curious how you view your go to market strategy and how you approach competitors or responding to competitors versus kind of cooperating with them. Yeah, so I Like we don't really see ourselves as having competitors as such. So obviously the roll up teams themselves like are, you like some of them started working on like their own solutions to distributing proving workloads. We don't care. Like they can do that. We'll just plug into those then. but it's just easier for them to potentially not do that or do something very simple. So a lot of them started doing this themselves and then at some point they realized, there's gonna be like a whole spectrum of proving service providers. And so like, what's the point? Like I should design my thing, assuming that it's just gonna be outsourced somewhere. And so that's kind of what's happened. And so we don't, I don't think there's any, roll up teams who are trying to build like their own kind of prover networks as such anymore. They just have some logic by which their responsibility for proving is assigned and then the rewards and so on. Then there's, you know, the ZKVM projects are also all kind of like, I don't know, doing something like a proving network. all, they're quite different. A lot of them like, Um, risk zero has sort of like an on-chain auction based system. Um, succinct, what little they've put out about what they're planning on that front is also like an auction based system. Um, and so on. So we, like, if you meant those types of, uh, projects, we don't see them as competitors at all. We're happy to help them as much as we can. Um, we ultimately don't care about anything else besides volume. So we're like, I don't know, like the Toyota of ZK proving where our whole game is to get maximal volume so we can drive down costs and then that's it. Like there may be middleman, there may not be, doesn't really matter to us. And ultimately it doesn't really matter to these people either because they're all sort of relatively isolated demand funnels. Um, and so you can, you can kind of like either try to create your own demand, but that always Like there are a lot of VCs in crypto who are very confident that there's only gonna be one winner or whatever in various categories. In this sort of stuff where you have completely open source technology and you have dozens of different, it's just like saying there's only gonna be one winner in like programming languages. Like it's just not the way it's gonna go. There's too many trade-offs. And so we, just support everything. we don't like specifically don't try to go into that territory. Like we're not trying to build our own ZKVM. We're not trying to build our own roll up. We're not trying to do any of that stuff. We're exclusively trying to support people on the actual approving front. And so I think that actually makes a lot of the relationships easier because if you look, you know, at various other ZK projects, like they'll sort of they're doing something and then they'll like suddenly be doing a ZKVM as well. And then they're doing this and that like they're much more fluid in where they are competing. And sometimes they're competing everywhere and sometimes, you know, not. And with us, it's just quite clear. Like it's much less risky to. tell us what your plans are because we're not gonna try to build a competing ZKVM. We'll instead try to help make proving on ZK Cloud for your ZKVM as fast as possible and as cheap as possible. Now, is it accurate to describe ZK Cloud as, it sounds like it's becoming essential middleware for, that's chain agnostic and for any chain that needs some level of kind of ZK proving in one of their offerings, something like that? Yeah, I mean, the vision is that every single ZK proof would be generated on ZK Cloud. There's some exceptions because some ZK proofs in like Aztec need to be generated locally by the user to get privacy guarantees and things like that. like everything else, we would like to ultimately be generated on ZK Cloud. And we don't care how they end up getting there, but... but that's sort of the, vision. in a world where, you know, like, well, this is maybe a good topic to talk about. Like ZK proving times are getting so low now, uh, that they're the justification for having like a, a fraud proof role, like an optimistic roll up is getting very, very diminished. Um, they have, you know, set multi-day withdrawal windows, which basically prevents interoperability unless it's completely centralized. while with ZK, it just got this like out of the box. Um, and like we just posted, a slightly provocative blog post today called optimistic is dad long lives ZK. And there's some figures on like the actual proving times that we're seeing out of, succinct and then risk zero to prove Ethereum blocks. And we're on the order of like, can prove an Ethereum block within two Ethereum blocks. So in like, you know, 25 seconds or something like that. And this is like, once that just gets low enough, like we could be posting proofs constantly. And so your interactions between different roll-ups or like bridge transactions would... go like at the speed of an Ethereum mainnet block. And I think that just fundamentally changes the calculation for how to do things. it, we can, mean, Justin Drake and Vitalik have talked about this a lot, but we can basically like start merging L2s and the L1. Because if you're an LL2 that generates a ZK proof of a block and you're posting it within an Ethereum block, like there's very little difference besides in just like how quickly you know what has happened kind of like, but finality always just happens on an Ethereum block across everything. So, so I think the ZK proving in general, it's my point here is that like, it's possible that very soon, just all of this is ZK. And that's kind of like what we. are aiming for. One of the key pain points is the seven day withdrawal window. Now, would ZK, because you're able to prove it within two blocks, would that effectively eliminate that? Yeah. Yeah. There's so, mean, ZK relapse already post proofs like settle much, fat, like much more often. so they don't have that, that withdrawal window delay. but, uh, but like, there are all kinds of considerations, I'm sure in terms of just like the cost structures and involved in such, but in terms of like, like in terms of performance and the cost of proving Um, there's no reason why you can, uh, be settling. Well, I guess like every two blocks, um, which is quite a, quite a substantial difference. So I think we're going to end up getting sharding essentially. It's just not going to look like metallic originally thought. Yeah. blog post that you're referring to, does that optimistic is dead, long live ZK? Is that also a kind of you're alluding to some of the key kind of stacks in the layer two space, like optimism, for example? There's no reason why they shouldn't be using ZK Cloud in their stack, right? Yeah, so I think all the optimistic rollups are just gonna switch to being ZK rollups quite soon. And yes, once they do that, then we would love to make sure that they have decentralized, very cheap proving underneath it. But really this is more of like a high level prediction on how things are gonna go. There's just no benefit to them from being optimistic anymore. I'm thinking in terms of the adoption of ZK Cloud and what are some of the ecosystems that have lots and lots of users and applications that could potentially... What's the 80-20 look like in terms of your potential target customer to get maximal usage and adoption of ZK Cloud in the shortest time possible? You can go after every L2, but... you your resources are limited. And so what does that look like? Yeah, we like in some ways we do go after every L2. So we have kind of because like there's like a lot of L2s but most of them run on specific stacks, right? So there's only really like three or yeah, three or four stacks depending on if you calculate or count in the optimism stuff. And so like we have integrations with these different stacks, which just makes it very easy for, for people to just start using it out of the box. and we don't need to do like a crazy amount of these, right. Um, and then after that, it's mostly just like, it needs to be reliable and it needs to be cheap. then. Like, I think that drives adoption. Like we're a small team. obviously don't have infinite resources to spend on this stuff. So we do need to just make it be so compelling that the. the economics kind of like sell themselves if that makes sense. then, but in terms of like categories, think like right now it's the ZK roll-up stacks and the ZK roll-ups themselves that like are doing proving a lot of them are outsourcing proving we're doing proving for scroll and Aztec and ZKSync already. But then. I think this is the year when we start seeing real movement in terms of the optimist stack rollups starting to look at converting into ZK rollups. And like, at least right now, it certainly seems like that's where like the majority of the usage is. What makes you say that, that this is the year you think that you'll see more adoption of ZK in general, but also ZK Cloud? Well, I think a big part of it is that it's just getting made quite easy for them. So we already have kind of like transition paths from both succinct and risk zero into both SP one or risk zero based sort of like optimistic stack compatible ZK roll ups. There was a There was some project, I forget the name now, which was basically doing like, um, optimistic stack, use whatever, uh, ZKVM you want and like, so on. like the, there's both kind of the maturity of the technology. Like it's been tested like ZK in general. We have to remember that no one had done anything ZK related. That was more complicated than was Z cash had done, you know, since just what like a year and a half ago, like the first rollups are not very old. Um, and so it's just been tested more. and, uh, and then you start having these massive, massive UX benefits and like optimism's entire roadmap is interop. Like that super chain is the, the whole thing they're pitching. Uh, and the only way to make it happen is ZK. So You know, it's sort of like, maybe it doesn't happen this year. I'm, you know, not an Oracle. I don't know. But, but certainly it seems like the incentive is, starting to be quite high to start thinking about it. just in terms of immediate, um, UX benefits and then also long-term kind of interoperability plays. How do you think about growth for ZK Cloud? It sounds like a lot of the actual work is more of a business to developer and also business to working with these specific kind of layer two stacks versus individual developers building apps on, know. So I guess how do you think about growth? Yeah, I mean, it's a B2B chain. I like B to D because it sounds sexier. That's true, it does. I should switch to that. I should just use that more. But I guess it could be like blockchain as a service or something like that. I mean, a lot of the very practical work is just talking constantly with these different teams, understanding what they need. Um, like we've always had a very sort of feedback driven, uh, way of going about things, where like this whole thing started from us talking to different rollup teams and trying to figure out what they might need. And so we just kind of keep iterating on that. uh, and, and the further we get, so now we have like a live network. Um, there's a whole bunch of people that are starting to use it. Um, we're starting to get into very like practical level feedback, which is like, you know, some specific feature that they specifically need because they have some special circumstance. Yeah, it is great. It is great. But that's kind of like the, that's sort of like where it is. It's also a fairly finite amount of kind of customers, which is sort of nice. Like it is growing very rapidly, but but like, least like in terms of the different stacks, like we can talk to every single team that's building a stack. That's, that's nice. uh, so a lot of the growth is kind of a pretty traditional BD in a sense. Um, and then growth kind of on the, I guess, community side of things and stuff like that, like that's, think where things get more interesting. It's less straightforward, uh, because it's, you know, there's no DeFi on ZK cloud, right? Like there's no, we're not trying to get you to become a user. Um, what we're actually trying to convince sort of the broader public of. is that this is like a reasonable business or a good business to be in one that is, um, likely to see explosive growth in the next 10, 10 years. Right. Um, and so it's more of like, um, like we're almost more saying like we can handle the actual, you know, running of the chain and then the BD side of things, what we need you to believe is, is believe in it as like an investor. Um, as an asset. And there, it's interesting. There's not that many examples of that kind of thing. You could maybe say Celestia is kind of along those lines, possibly. Sort of the same dynamic. But I think it's like, I think like we're learning slowly the playbook for doing that kind of thing. It's not quite the same as with all other chains. Um, and so it has taken some learning, I think it makes sense. ultimately. ZK cloud is successful if ZK is successful, and, um, and it's not tied to any single ZK project being successful. So it kind of functions as an index for ZK. I think if you, you know, believe in ZK at all, and that makes for an interesting, interesting investment and also proving is something these projects need to do, you know, rain or shine, right? Like bear or bull. So, um, there's a lot of interesting stuff on that front, but yeah. think, I don't know how appropriate the analogy is, but I think of ZK Cloud almost like a necessary piece of middleware, kind of like a chain link, for example. Almost everyone needs a chain link. that's a that's a great example. Actually. Yeah. Yeah, Celestia and chain link are both kind of similar in that way. Yeah, the benefit of, you know, when you think of Celestia, it's like they're able to help, you know, save chains money, right? Or save applications that, you know, with their data availability costs. they're able to save tons of money for these applications. But the question has always been like, how does, that's nice. And as a project that they're adding value, but how does that accrue to the TIA token is kind of the big question. And I think that's a question probably every project kind of, sorry, every project eventually needs to kind of figure out. And you guys will figure that out, especially as you're marching towards main net, whenever that is. Maybe we can talk about that. What are some, I guess, what are your thoughts on main net and some of the work that still needs to get, you know, still needs to be done? Yeah, so I mean, there's, it feels like there's infinite work both ahead and behind us, but right now, so I think, was it last week we announced applications for test net validators. I think that was, I could check, but I don't wanna start typing now. I think it was the 10th of, February was the deadline for anyone that wants to run a validator. There's limited spots because it's a Cosmos chain and it's quite a fast Cosmos chain. that's kind of the next thing that's happening. That tasknet will spin up quite soon. And then it's just onto mainnet. I don't want to give any specific timelines, but we are looking to do it here in sort of the first half of the year. exciting. yeah, yeah. Uh, and like the, it's almost like a separate, so we have fire starter, which is our sort of permission chain that is running and then just doing proving for, people. Um, and then we have main net and there's like, we're kind of, well, I've started realizing, in the, in the past few months that we're kind of like running. many chains at once. But there's a reason for that. it's because like mainnet, when you just have everything be permissionless, there's just a lot of edge cases, which are very hard to predict. And this is like a mission critical function for our customers. And so we need to be really, really sure that mainnet is stable and reliable. like in ways that like traditional networks don't need to think about. Like we need to make sure that if somebody accepts a proving job that they end up doing it or that they get like removed from the network very aggressively and that there's like a backup and like somebody ends up doing it. And we need to do all these things because we're not re-executing. So we don't have this natural redundancy. We need to design all these special mechanisms in. so it's gonna take a while before proving really lands on mainnet. And we're gonna be quite slow and thoughtful about that transition. Having said that though, mainnet. is really the best place to start testing out these things in a way. Like we can never fully test out a decentralized system in a non sort of permissionless real money way. And so we're very excited to start doing that. And I think it's gonna be, it's gonna be a pretty crazy year for our team trying to get all this stuff. Awesome. but we have an incredible team. I'm excited to say we'll probably be announcing more on that soon. Yeah. Exciting. Well, Teemu, thank you so much for taking the time to talk with us. And we look forward to watching ZK Cloud as you guys march towards Mainnet and grow the community and especially grow your amazing set of customers that are using the ZK Cloud service. So thank you for taking the time. Thank you for having me.