Modular Blockchains and Limitless Scaling of Ethereum | Anthony Rose - Matter Labs, Zack Guzman

[Music] all right ladies and gents we get started here at Denver very excited for this little chat little fireside chat we've got going on here with Anthony Rose matter lab's SVP of technology to chat about ZK sync scaling ethereum how we can do this limitlessly I believe is what really the title was all about here Anthony um I'm the co-founder of trustless media I'm excited to be digging into this because we're all at Denver to learn a little bit more about this technology and what it unlocks and Anthony I figured we'd start just with your background because you came from a big tech company that people might know in SpaceX and have a bunch of you know I think cool insights on why crypto matters so why don't we just start about why you're working on ZK sync right now to scale ethereum and and why you think this is such a huge opportunity and moment in time to be building on E yeah it's uh cool yeah good place to startop um so in terms of like why ZK sync why work on this um so for me personally I've been following the space since 2013 so uh I was doing a postto at the time became aware of Bitcoin through you know other people I was hanging around with doing physics uh and it was one of you know like many people right immediately read the Bitcoin white paper and it resonates um and I'd followed the space off and on over the next however many years uh probably with my level of Interest kind of like uh peing and waning maybe as a function of you know everybody else's um but I kept coming back to it and uh yeah was at spasic for like four years leading up to um working on mat mat Labs on ZK sync and you know you'd be uh working 12 plus hours uh on something completely unrelated and then go home and read about D5 protocols or like look at nfts or you know dig into you know all of the technology that you know all of us are here to kind of dig into together today um and you know like when you feel that compulsion uh you should explore it seriously um so I thought about the problems in the space and for me like ethereum was the only chain that um really offers the promise of what we think public blockchain's going to can deliver at least for me personally so looking at the problems for ethereum you know scalability is the only one if we don't solve that I would argue that like why do any of it um and to solve scalability we have ZK and inside ZK I spoke to the matabs team and was kind just blown away by the quality of the team like the vision the commitment to the mission it was just obvious from my first phone call and as I spoke to the engineers you know I hadn't necessarily expected to leave SpaceX at the time yep um and then I was uh you know handing my notic in shortly after I I wonder sometimes too and you're much more coming at this from a technical lens than uh you know I am maybe perhaps um in terms of you know how this technology can be used and now as we've arrived here right you know it was hard before to kind of the gas fees on ethereum were one thing but now the solutions there are so many and so maybe get into a little bit of what makes ZK sync different from the other solutions to scale ethereum and and what the big unlock is so yeah huge number of different scaling Solutions I would argue the uh you know if we look over the last couple of years the ones that really hold the most promised are the rollups um they don't offer as many compromises as you know previous attempts at scaling ethereum and they really you know broadly fall out into two categories we have the optimistic systems and we have the ZK systems um the ZK systems are you know coming into maturity now and optimistic systems I think were you know fantastic technology to relieve the pressure on layer one but you have you know game theoretic guarantees to correctness with ZK you have mathematical guarantees right and you have the ability to not require this like weekl long fraud challenge window you can get to the point of like instant uh guarantees to the state transition and you know fast finality on layer one for ZK syn in particular because even inside ZK there's you know many um great teams building everyone with different design decisions I would argue that we've optimized for the future right so we've optimized for a system that is super high performance in ZK we've optimized for an architecture that we think gives us the ability to make the system extremely cheap and you know really trying to build an execution environment that we imagine Builders being able to like build really interesting um you know fantastic user experiences on that move us Beyond kind of wether spaces today to a world where you've got like the internet value on chain um yeah so it's like performance and price and uh a really amazing user experience yeah I think the the trade-offs in each technical team and and how you want to scale ethereum are always interesting to get into because there are trade-offs at the end of the day right uh and one of them I suppose is kind of how you build out in this idea of why everyone's trying to scale ethereum is because there's Community there and there's Builders here at this conference that you might not have in other ecosystems uh and getting into that technically like evm compatibility versus equivalency things like that talk to me about the trade-offs that your guys' team looked at in how you wanted to attack this I guess differently than than others yeah this is probably like the maybe most fundamental design Choice um but at least as we were thinking about this um the evm was designed prior to anybody thinking about using ZK right so the design of the system isn't trivially uh provable when it comes to you know building this cryptographic proof systems and these systems are complex and you know what we're building is a virtual machine the virtual machine will be you know running a bun of compute but to do the corresponding you know computations in zero knowledge can be orders of magnitude more expensive than doing them in a regular virtual machine so we designed the virtual machine to be efficiently provable but this meant that you actually end up having to to implement uh you know the system in a very different way so that you can make the user experience feel like the evm so users can still you know write their smart contracts in lity or Viper compile them deploy them it should behave as it uh does on layer one but the system itself is very different and it's designed for that performance and I think you know we're starting to see the benefits of some of these decisions right we've I think demonstrated the most performance of all rollups ZK or optimistic on mainnet um and it's the cheapest so it's like the fastest and the cheapest which for the execution environment like obviously critical characteristics yeah yeah I was talking to one one project today and they were talking about some of those trade-offs and the idea of maybe the edge cases of copying over in terms of evm compatibility and and what that looks like and obviously again there are always trade-offs when it when it comes to that and the incentive structure I'm sure you get asked all the time as any project in crypto does when token what does this look like in terms of the incentives to build uh with your team versus others I mean you know it's interesting to think about the incentives and the token you know everything in the space right now the excitement's back around that how have your how have you guys looked at it in terms of I guess a snapshot airdrops what does that look like and and and is it maybe asked too often uh I mean too often is subjective right um but I probably get off the order of 50 Twitter DMS a day on on the topic that seems lower than I would have expected well you know um yeah don't read the you know other folder um but at least nothing to announce right like for us we are building we haven't needed a token so we don't have a token I think um you know at some point in the future if it changed like we would be the first to be talking about it but at least for now we don't you know have a token to talk about um but when it comes to incentives like you know if I think about putting my like getting in the mindset of like the builders who are building on these systems today like if I was building an application I would want to design it and build it on a platform that I actually believed could get to the point where like millions of users can use this application right like if you're looking at systems that are limited by the performance or they're fundamentally expensive you're I mean just like with layer one right that you can't build applications that huge scale on a system that doesn't scale so the incentives for me are like if you're building you want to choose the execution environment that you can believe in and you can be successful with I think what's interesting too is you know when we think about this conference maybe a lot lot of explanations don't need to happen here when you're talking about you know zero knowledge proofs when we remove ourselves from E Denver maybe there is a bit more right and and right now we're seeing a lot of excitement in the Bitcoin ecosystem because of a very simple Narrative of hey Wall Street just figured this out and now there's a lot of money coming in and number go up when it comes to like your excitement as someone who is deep at SpaceX and in physics and you know working on telescopes and like what do you see as kind of the demand side of what your Tech technology unlocks and maybe not now but in 5 10 years time what it unlocks I mean you know the thing that first appealed to me when I was really thinking seriously about the move was it feels like we're on the cusp of a phase transition right where we can really see a world with much much better Financial infrastructure and you know as I've been deeper into the space over the last few years like we can see now the path to an architecture that can basically provide the backbone for an internet value right like a world in which uh you have the properties that we like about public blockchains right you have self- custody um but they're accessible to everybody and they're accessible without the security compromises that we've seen on you know 01s um and you you know you can really imagine a much fairer and more Equitable world and um some of the things I think in the west that we take for granted can be made much more accessible um and then beyond that just you know the level of creativity in terms of the application layer is something that excites me right like nfts a thing for sure would not have predicted and the unpredictability of it as well I also just find very compelling um and you know people building things on ZK sync that surprise me is you know always a source of pleasure yeah I mean you know what we've built at coinage in terms of being a community-owned show might not be as complex as maybe some of the other things that are built on chain in terms of nft equals membership equals ownership of what we're building but for the other pieces there like talk to me about the technical unlocks of the early builders in your guys' ecosystem for kind of what you have to point to cuz the activity is up right when you look at kind of what you guys have been able to show in the early days even without a token so I I can't imagine what it might look like in the future but what has you most excited right now um I probably wouldn't want to pick one project over another because at least for me like you know I really want our system to be a platform on which uh independent of like our personal preferences we attempt to like really engage with all Builders right so um there's obviously defi there's games there's nft projects to First approximation you know all of the things that are being built on ethereum we see like versions of them or or like you know the L2 version exactly uh being deployed and and starting to thrive on ZK syc um yeah you know like I have the things I like the most but also I try to use everything on on ZK sync I really want to to build the best system that we can like we have to be users of it so I play with everything so yeah like uh you know some I like well than others but it's kind of just cool to get a flavor of everything that everyone's building and um and also to try and engage with the building Community it's self directly yeah when when you think about I guess obviously it becomes a little bit easier maybe to explain to other people who might look at your decision to leave SpaceX versus now when you know the excitement is back in the space um but for you I mean I guess how do you contextualize you know building day in day out in this ecosystem and and why you do it um on a very personal level because I could answer that and I think it's really cool to see like ownership like you said in the media space as creators working with their fans and and people who are interested in crypto news but for you like what is the unlock in terms of why you're here I mean I just deeply and genuinely want this system to exist but I think a world in which these systems exist is strictly better than a world in which they don't and you know the motivation uh every day is to keep pushing the system forward you know we spent the first year that I was here sort of building towards that main net launch uh which was like an interesting and exciting time because you have this amazing thing and you want to bring it to life so you push every day to get it there and then it's live and it's a very different world right um You Never Know What To Expect When these systems turn on it's a permissionless system and people can do things with it that you don't predict and you know I didn't know how much usage to expect right like we see a lot of variety and it was very hard to predict which systems would get a lot of activity um but you know as you mentioned right transactions are up tbll is up addresses are up people are building people are deploying um but it still feels like we're just getting started so feels as exciting as day one what would you say I guess is one of the things that people get wrong when you look at your technology and you mentioned kind of the differentiation between a zero knowledge proof uh and the other existing implementations of scaling ethereum I mean what what's kind of one thing that you would clear up at a conference like this um there's a bunch and honestly we only have six minutes left so yeah the the first one that I think is a difficult message not just for the ZK sync but for teams building ZK systems in the space already is to really highlight the advantages that ZK gives you over optimistic systems and I think in the short term these systems don't feel very different right like to a first approximation use any of them today and they're like okay like the user experience is roughly similar in the long term it's going be completely different how I mean thinking about where we're going right if you think about these systems the internet of value cannot run on one L2 right so all of us to a first approximation are thinking about the real scaling problem the real scaling problem isn't you know maximize the hell out of this L2 it's think about how these things scale horizontally and to scale horizontally you can do this today with the framework we've built so we open source the codas for era called the ZK stack so you can deploy hyper chains into this ecosystem and it's through ZK that you can do uh permissionless asynchronous message and like value passing so you know with an optimistic system where you've got a 7day challenge period I mean imagine if you like tried to open a hyperlink on the internet and it took s days to respond it's like and this is a fundamental limit of that system Y and like ZK lets you go fast right we're going to get ZK to the point where it's basically real time so you can build these composable systems that can scale to the like what we need for the internet value um so this is a fundamental misunderstanding and I think people building on optimistic systems you know if you look at l2b which is an amazing tool for understanding the space it is phenomenally complex to understand the properties of each of these the risks Associated to each of them and let alone like what this will look like in 5 10 however many years yeah um so I think on the ZK side we need to do a better job of like helping Builders understand this because they can they just need to focus on building right um and the infrastructure yeah they they shouldn't worry about well you mentioned I think putting it all in one chain the internet today like in 2024 and seeing right now what's happening with AI I feel like is a good example to point to in terms of how big this can all get uh as it expands right and it's really hard to predict and I think a lot of this stuff takes longer than it does you know people have been talking about self-driving car and you know I feel like for forever but now that AI is here in a way that this all kind of can really start to get out of hand very fast I mean how much does that accelerate the need for Technologies like the One You're Building I would say it's not obvious to me that AI is going to be the thing that um we really focus on in the near term and I could be completely wrong and this is like maybe even people that I work closely with would disagree with me um but for me like the first and most important use case that we've seen for these systems is around Financial assets right like around payments around things of that form and I think today these systems could solve problems that uh I imagine in the web 2 world if people understood what these solutions could do already they would be embracing them in a way that they uh aren't yet so um yeah you know once you are able to build this like posable system that can scale horizontally people also can configure their environments in a way that makes sense for their use case right you can build private rollups essentially so that you can start to build really interesting Technical Solutions to you know problems that normal people have um and Abstract the technology away and just like really start to build a modern Financial ecosystem before we popped up here on stage we were talking about your path here your road here you've been working on this now for a little over two years right and your position uh as SVP of tech um I mean two years in the bear ages some people I certainly felt aged and now it's you know the tides are turning you personally I mean how do how was it building in the bear and how different maybe it feels now is I hope we've turned the corner in the in the excitement being back uh I still don't feel like we're in the bll properly yet which is maybe wild given the price of Bitcoin this week um but Building without the noise I think is really helpful right like the system there's just like a lot of work to do and it lets you focus in a way that you know the prices can be distracting prices can be distracting yes but actually like to a first approximation I haven't seen any difference in the last few months relative to the two years before right like we're Engineers building the system that we want to build and like what's going on around us is less interesting um we have a very long-term view well let's look ahead to the long-term view because we've got 2 minutes left and and I feel like looking ahead to what comes next obviously you said you get asked about tokens every day so I'm not going to ask you again but for the road map and how you start to use the technology uh what should people keep their eyes out for for how ZK sync starts to actually make it into the real world in a new way I think that like coming back to the two properties if I think about era like the system that we have running now on on Main net we've been pushing performance beyond what everybody else is doing and we're going to push it way further like it's going to surprise people when it when it's live um and then on price right we have 4844 coming in a few weeks we're going to be ready on day one um so the way that we've designed the system lets us really build an execution environment that's going to be killer and I think it's going to surprise people um but the really exciting thing for me is actually this hyper chain ecosystem we have a bunch of Partners now building they have test Nets running main Nets coming live and we're going to start to see like real interoperability through ZK and real z uh like scaling through ZK and you know we're going to go from hundreds of TPS to thousands of TPS and of course the hyper ecosystem you know yeah like an environment of huge amounts of execution super cheap transactions and um yeah Builders will finally have an like an ecosystem on which you can um build applications that really scale I think it's it's a great point to end on because you're right we are we are so close to that upgrade Denon looming in the near future um but it's been fascinating to hear kind of your your thoughts on this so I appreciate you sitting down here at e Denver with us Anthony Rose matter Labs SVP of Technology thank you again man thank you appreciate it all right thanks everybody