Why Ethereum Got Scaling Wrong with Lucian Mincu of MultiversX

Transcript

It's not the main goal that we will only have lower transaction. It's not only one one particular point but actually the goal is to have the most meaningful experience when interacting with this kind of technology like how can you build something that you basically the technology will become invisible. Now for that to become invisible it needs to have extreme low latency, extreme low cost, extreme efficiency, high bandwidth technology regardless of whatever architecture. On the one hand, I think the first step was for Ethereum to have the beacon chain and then the second to the export the the state to the shards, right? We launched specifically with the peacon chain and with the shards from day zero because we did not have the legacy actually to to maintain and do

the entire [Music] steps. I'm here with Lucian Minku. He is co-founder of Multiverse X. It is the oldest layer 1 blockchain with full state sharding. In fact, they were trying to and have implemented the initial road map of Ethereum 2.0. So, we're going to dive deep into Multiverse X today. Uh it's one of the oldest L1s in the space that's still active, still building, still relevant I think technically, but also as an ecosystem.

Uh so, yeah, happy to dive into all that with you today. Hey, Lucian, how's it going? Hey, Sebastian. Pleasure to be here. I'm very good. Very excited actually to to have the discussion. How are you? I'm good, man. Um, it's been an exciting time recently. Um, we've been traveling quite a bit. Saw you a couple of times the last couple of months, but yeah, certainly like I feel like there's a disconnect between the the narrative and the markets right now. And I'm trying to make the best of that and wrap my head around what the hell is going on. I think like a lot of people in the space. How about you? 100% agree. So I think uh indeed we've been both traveling quite a lot I think the last

couple of weeks and I think especially in the last uh it events and especially the side events I think the institutional side of of of the industry was more packed than ever actually regardless of the market prices or market sentiment. I think it's crazy and it's kind of feels to be in two worlds at the same time. On the one hand we have almost all the most bullish news with regulation with everything that we wished actually with institution with governments tapping finally stepping in and on the same time I think the markets are just bleeding but I mean it's just a temporary thing and uh we've been we've been we all been through this I think uh several times until now and the m the

market just are just doing their thing. They're all gonna Yeah. Actually, I was I was looking at uh so we're recording this on March 13th. I was looking at um uh kind of prices this morning and I was it kind of surprised me. I mean, I guess I knew this, but I hadn't like looked at the data properly, but uh if you like the only assets if you've held for the for the last year that would have made you money are Bitcoin, XRP, and Tron. Um even you'd be down even if you had been holding Salana a year ago. definitely be down if you're holding ETH or any of the other top 10 assets. So, it's pretty weird what's happening right now given just the bullish news to come out of,

you know, the US with more regulatory clarity on the way implementing this US strategic reserve with crypto having crypto friendly folks at the SEC and different regulatory agencies. I guess it was priced in, but it just feels like given all of this really bullish news that somehow the markets are not reacting to it positively. So there could be, you know, looming geopolitical realignment and possibly, you know, conflicts that are also contributing to this. But yeah, as as builders and investors in the space, it's definitely been uh difficult to navigate. 100% agree. 100% agree. But on the other hand, maybe to to build on top of that, I'm finally, I think, actually extremely

bullish on on on one particular thing, I think we've been all kind of driven or pointed towards the casinos that we built on top of this kind of technologies. And on the one hand, I think everything that kind of worked until this space or or mainly we we took it as a good uh good experiment, it was built purely or mostly on speculation. If we look at centralized exchanges, decentralized exchanges, pumpf fun like or any kind of supply chain towards that, it was mostly built on top of that. So on the other hand actually I'm finally extremely happy and bullish that I think the market have went back to the fundamentals. We heard it this pretty

much all over the X communities and so forth. Founders are coming back to building real and exploring new frontiers of the industry. I think this is something that I'm very bullish about and uh I think we're going to discuss a bit about that uh during the next minutes. Yeah. Yeah, for sure. So yeah, let's let's start uh with a little bit of background here. So, Multiverse X has been around since I think 2017. You guys did your ICO. It used to be called Eland. People will remember that as one of the larger ICOs and L1 launches of the last dec of the previous decade because it was in the 2010s. So, yeah.

Can can you share a little bit about the history there and how you got to where Multiverse X is today? Definitely. Definitely. So, so I think maybe to to start back to the initial story, we we start looking uh I think prior to starting working on on Elron, we definitely looked at pretty much everything what was relevant in the market especially also got involved in pretty much even Cosmos Polka dot everything that was especially tackling uh scalability infrastructure at that point of time and also challenge a few of the projects actually where I' I've worked closely with to start exploring this kind of path with proof of stake specifically and state charting. Even

Ethereum back then had pretty much nothing about that. But actually it was just in a blog post or some tweets of Italy just mentioned roughly the words, right? No architecture, no nothing around that. So we we pretty much looked and said okay what what's the most challenging on the one hand idea and also what's the most promising architecture that we see that could be relevant even to more than than just a cycle and actually something that will have very strong fundamentals and this actually brought us more to the state charting like the blockchain trlemma that Vitaliki was pointed that means that actually you cannot achieve uh scalability or high high performance without compromising any of of of of this of the three which is scalability,

security and decentralization. Now those three actually the only way to achieve the optimum ratio to any of this with the last discount to any of those values would be through the state charting which I'll build maybe pause here for a second ask you let you give maybe ask something and then otherwise I will just build work with you on No I think that's helpful. Yeah, I think that's helpful. I mean, um, with with regards to state charting, like this was part of the Ethereum 2.0 roadmap. Ethereum went in a different direction uh implementing a more rollup ccentric road map and like this prototank sharding uh which maybe you can compare um with the state sharding idea but um do can you can you

share like point to reasons why um Ethereum ended up not implementing this this state sharding and went down a different direction. Yes. Yes. maybe just defining a bit the state charting then the way I look at at blockchain architecture. So maybe just one one step and then we'll we'll get to that. So for everybody to understand like the easiest way to look at any like we we refer as blockchain as a supercomputer and a supercomputer still it's still a computer in the end which has a computer architecture has mainly three components CPU run disk right CPU is is consensus you work with requests to CPU the CPU worked with a with a very fast memory which is the RAM in practice and then it it once it permutates the values it

writes to the disk how it got there right this is very high level gender their compute architecture. Now if you look at blockchains, we have request translates to transactions which ends up to consensus. Consensus process those transaction. It works with a very fast memory which is application state. Right? Right. This is the RAM mostly of the blockchains and then it writes to the blockchain to the history to the disk how it got there. If we look at general general computer architecture, how did we scale up any meaningful even the smartphones that we have today? They did not end up with one thread that has 100 GHz. Actually, it has a bunch of threads. I would even say what's the most like what what's the main difference between GPUs and CPUs? GPUs

are especially optimized for parallelization and a large set of parallelization. And this is the way I think it I I would build the the position of sharding. It actually permits and allow us to run multiple consensus in the same time. multiple consensus on the same time, multiple CPU threads, each with their own dedicated memory and then each with their own disk allocated disk. Everything orchestrated and coordinated over a beacon chain or so-called meta chain. Okay, pausing here. This is actually just a bit of uh very simple words, layman terms I would say explanation of uh the the initial Ethereum 2.0 road map. So, so basically now we we have a general architecture

where we can say okay so in in normal blockchains like Ethereum at this point you have pretty much one CPU one thread everything is uh executed sequentially. So in the initial road map that would mean actually that the accounts account account storage or account state will be exported and pushed towards sub or or so-al shards right so that would be a a subnet within the same network that will be rotating validators and maintaining the same security primitives but then also for example if I would translate into very simple practice that would mean that for example if Ethereum address range would be zero X00001 to E 0

X000010 half of them in the in the case of two shards half of them will be colloccated in one chart and half of them will be collocated in the other shard. Now what does this bring? So it in in in practice why Ethereum I think did not end up executing on this actually this brings on the one hand complexity engineering complexity when it comes to composability and automicity on cross crossart messaging. So this is still something that we I think almost any kind of messaging system crosschain messaging system fights at this point to optimize and this is also something that we built and embraced from day zero. So on the one hand we have automicity within the same the same colloccated

storage zones which can paralyze the entire execution even smart contract processing and so forth and then also an asynchronous uh composable model messaging system that that actually abstracts the entire uh the entire complexity or or knowledge about about sharding or something that is outside of your subnet. Right. How is that? Yeah. No, that's very helpful. And so just to kind of recap there on a on a shard you have automicity of transactions and application composability but the issue and the the challenge is maintaining that composability across shards but you guys have built a an internal message passing system that will do things asynchronous asynchronously so that applications can interact with each

other when they're not on the same shard. I think also this is just a temporary like this is the hardest way to solve because on the longer term you could for example synchronize the consensus and build it is just a matter of engineering steps to synchronize ahead the execution needed where you could basically have automative even crush art right so but it was just not the the main challenge at this point but I maybe just trying to maybe going back to the Ethereum right how did we ended up actually on on different trajectories So on the one hand I think the first step was for Ethereum to have the beacon chain and then the second to the export the the state to the shards right we launched specifically with the beacon

chain and with the shards from day zero because we did not have the legacy actually to to maintain and do the entire steps but then on the other hand what what they did actually and how it turned out to be with the sequencers and the entire rollup road map it it actually translated to this kind of crotoden and sharding which is on the one hand the blobs which are specifically the slots pre-allocated on the the data availability layer which would guarantee that they don't account any execution on that for the main chain and then also that also ensures that it is a more effective storage system so forth for the L2s so in this case where our our road map would the state will be

still persisted on the the each of the shards in this case actually Ethereum serves as a data availability layer so to say it. If for example Ethereum would have implemented a unified method of reading and writing and interpreting those blobs from within all the rollups they will end up with an asynchronous uh bridgeless communication within all the L2s. I think it kind of translated we are doing the same but without posting the state to the beacon chain. This is like cross notorizing the headers, cross notorizing and finality orchestration across all the rollups or shards or however you would tell them call them but then only specifically would be the part with the with the reading and

interpreting specifically cross data from one rollup to another. I hope it's I I think it's pretty complex. I we're going very fast through a lot of many many many ideas here but uh yeah let me please pause me whenever you feel it's Sure. Sure. So, so the the like Ethereum has gone on a a different architectural direction which is to promote and facilitate applications being built as roll-ups. Now, this offers like a lot of flexibility at the roll-up level where roll-ups can choose whether they want to run optimistically or as like validiums or zk rollups. And you know there's even examples of I guess like side chains

like Nosis chain which maybe is is not necessarily like um like sort of as an outlier. Uh and then this DA u uh component which is the prototank sharding like gives more flexibility I think at the the rollup level. So if we just look at things from a purely architectural and kind of design choice perspective it gives more flexibility at the rollup level. However, what we've seen is that it it um it leads Ethereum mainet to have less control over fees and and how it monetizes access to uh its state. Is that the like is that the right way to look at it? Whereas one% sharding actually sort of imposes uh its fee

structure on the shards. Yes, but not only. I would say the main reason that why we're building chains and trying to do more specialized architecture towards more efficiency and so forth. It's not the main goal that we will only have lower transaction. It's not only one one particular point but actually the goal is to have the most meaningful experience when interacting with this kind of technology. Like how can you build something that you basically the technology will become invisible. Now for that to become invisible it needs to have extreme low latency extreme low cost extreme efficiency high bandwidth technology regardless of whatever architecture right so now the question is how much can we bridge and f can one

particular technology can capture about that and I think this is particularly one that I would push and think that Ethereum would still have a better upside if it if they would deliver a faster execution environment for them as L1. There's maybe yeah, it will be competitive with all the L2s and L3s and so forth, but I think at the end this will translate to even more efficiency, even better experience for all the L2s, faster cross communication within all of the the L2s and so forth. But I I'll go back more to multiverse. We're more than happy actually to to do another one only on on Ethereum. Yeah. So like um you know with regards to to multiverse X

like what what does that look like like how many shards are there practically and how do applications decide like on which shard to build? Is it just a question of which other applications they want to be composable with or are there other considerations there? Very good question. Very good question indeed. So on the one hand given that the the basically the the default mechanism is like the the address the less bytes of every address get gives specific location on within one shard. Basically what most mostly or application have application developer have done actually they choose a key that were closer close closer to any kind of whatever they felt more

composable or had more composability needs to. However this actually we we we built it's soal adaptive state charting. So this we just started at the uh back in the days with three shards three execution shards and one beacon chain. And this actually splits specifically into a binary tree model and gives each of the address. But however the the chain and I think we we ran prior to the mainet. So we we inspired a lot of game of stake battle of stakes back in the days after launching everything successfully. We said look let's take the wildest configuration that we can run on this architecture and see what happens. And actually we ran I think 50 shards and tried to generate as much in spam and did those the network where

where we as a foundation did not hold any any node and so forth and then we we actually validated that the network can actually scale and operate prior to mainet to upwards of 50 or 100 charts. Right? So this gives us actually the entire upside. The main challenge was specifically to make sure that there's enough economic incentive for for uh those validators because it is kind to run 3,200 validator nodes. It's kind of a interesting model kind of different than any other network in the space. But then also one I think besides just adding more validator validating nodes or more shards, we figure out that there's a lot of upside on the

engineering side. So even from the prototyping to the mainet code we we brought 50 times improvement and I think now we with the upcoming road map I'm very bullish actually what when it comes to engineering efficiency again what would be possible with the with the network so what's on the road map yeah very good question so so the things that I'm very excited about are two two specific upgrades major upgrades so one we started with sixcond block time and I think it was two two blocks finality and especially for example five blocks for cross art. So uh that that that turned out to be kind of uh very good at back then but uh still given the new

narrative and now we're going back to what what I think the narrative for 2025 will be back to fundamentals. I I strongly believe that there is one strong narrative which is reduce latency increase bandwidth. I think everything that we see succeeding in the last uh last uh months or last uh last uh years, it kind of built up again the entire narrative. Reduce latency, increase bandwidth. Sooie, Aptos, Salana, everything like that. It it had it it kind of built everything towards that.

Now the cool part is that we actually been working on that again for a very long time. And finally, now the casinos are are kind of dialing down. there's more room for uh for engineering for hardcore engineers to shine a bit or come up front right so there are two points that are we are working on on the one hand there's uh so combining both of them is sub-second finality what does it translate to so basically we're increasing the security of the network to one block finality which is the ideal like this is the holy grail of pretty much any network that people want to build at this point and then secondly with the with the July high I think towards July or August we're targeting to have a block time reduction from

first from 6C to 600 milliseconds then end of the year 4 400 milliseconds and then 200 milliseconds now combining this with one block finality and and sharding this kind of position to a most craziest I think architecture that someone can build build again and uh yeah even if it was built seven years ago crazy how far we can uh build on the same fundamentals. Wow. Okay. So, um like what kind of technical architecture or like design changes have you had to make to achieve this kind of speed on the finality level and reducing the block times. Let me just zoom out a bit. Right. So, trying

to to again grasp and compress everything uh to make for everybody understand. So on the one hand each chart has 800 validator nodes 400 active for 400 waiting and there's a random sampling every 24 hour where onethird of the validators are being reallocated towards shark to to avoid collusion or any kind of takeover malicious takeover within that now going back to the 400. So in order to achieve high throughput we had kind of two two models on the one hand on the beacon chain because there's very little heavy execution there that that where we needed ever since genesis the highest security standard that gives the finality and orchestration across

all the shards and that one had a consensus size of 400 or 400. Now, especially if you if you want to run a consensus where you have two or three three rounds of communication with 400 validator nodes, that's kind of challenging, especially processing upwards of tens of thousands of transactions per second because purely because of communication bandwidth, latencies and so forth. Now, we solved that we we this was as I said five years ago when we started this and it worked very good. And then on the shard on the shard level we had uh one uh 73 73 uh plus one. So leader plus 73 random sampled every every round every single round with uh uh yeah unpredict

so-called unpredictable unbiasable by one is the concept. Now this was the yeah this was the initial setup. Now what we're doing actually we're upgrading the the consensus size because the the the the entire network topology improved a lot. We we're upgrading the entire security primitives to 400 or 4 400 or 4 on all shards. So there will be higher like the all the validator set will at any point of time be actually validating and giving signatures on that. And then there's some new concept that are basically so-called inclusion proofs that will gives a receipt basically knowing that in practice. So in consensus basically in order to consider

a transaction or block final or valid many many times you basically don't need to wait for 400 or 400 signatures because the the the cryptographic assumptions says that if 2/3 plus one already gave you the signature you can already that's already 100% valid. There's no concept of 200% valid right. So everything on top is actually just a bonus which count on on the rating mechanism or or economic incentives for validators. But you can pretty much peer to finalizing the or considering the block final or or closing producing the block. You could already emit a receipt that says okay this is already valid.

It's already everybody already has this and everybody can reconstruct specific given the the the data that they all have individually aggregated they all come to the same conclusion as a receipt. Now this is yes this is a bit of the the the security assumptions that we built for for reducing the the latency and then secondly is on the the the reducing the block time. It is I think again an entire discussion. I I won't go specifically only towards that because but yeah it's it's kind of going to take another 10 or 15 minutes at least to build up the entire architecture. But yeah I think there there's a lot to to to share on that as well. So what is this going to

practically speaking from the application standpoint? How is this going to improve the life of users and application developers? um yeah like what's what's the actual tangible benefits? So this is a very very very good question. So I've been thinking about and discussing a bit with many of the guy the people or CTOs from other other projects in the last weeks and then uh I I came up to the conclusion that for many years we tried to replicate and do whatever most of the chains were building with a noncharted database. So specifically for example if you look at the centralized exchange structure data it doesn't require a

parallelized or sharded architecture it doesn't mo most of the dexes doesn't require that most of the specific specul speculation orientated application I would say don't need nec by necessity a sharded database now the question is can blockchains be anything else than just that can we build anything and then I came to the the on the one hand to the to uh to the conclusion that maybe we're looking too close to the tree. Maybe we're just looking too close to copy the next hyperlid to copy just the next uh fancy thing. And then the question is what's the most valuable technology that has been built on a sharded architecture versus non-sharded architecture and this

is not related to blockchains. This is fundamentally speaking on the technology and this is something that I'm working on uh right now and it will be unbuild as a next layer of of application stack on top of that. So but on the other hand still compared to anybody else compared to all other chains is still going to be superior to pretty much all other ones because finality execution time faster settlement and so forth. But then again levering up what's the ultimate application that could be built only with this kind of technology stack. It's still something that we're going to unveil very soon. I cannot uh disclose that part at this point. Okay, cool. Uh well hopefully we'll get to know more about that soon.

um like zooming out a little bit and just kind of like looking at at Multiverse X as as an ecosystem and you know as a chain and L1 that's been around for a long time. Um you know it's it's um there there's like in 2017 when you guys did your ICO and like subsequent launch uh there weren't a lot of other players in the space. Now there's dozens of Ethereum L2s. There's new sovereign L1s. There's like Salana that has captured a lot of the memecoin and uh NFT volume even though that's kind of dead now. There's just like lots more

block space and the industry tends to want to throw its money at shiny new things. How how challenging has it been for you guys to you know like I realize that you guys had like a huge ICO like well capitalized but you know I often think like for for these chains that have been around for a long time how challenging is it to maintain like relevance and attract developers and users and use cases um when there's just like so many more things uh being built um in the space that you know unfortunately whether like those things are technically viable able or or viable at all long term, you know, attract people's attention.

Yeah, more than happy to to to walk you through. So maybe just a small correction. We had a very humble ICO. I think in we were at initial point of time uh at initial release of the white paper overs subscribe at 100 mil but then did not raise anything just to wait a regulatory clearance and then the entire market felt like it is today. So it Ethereum was first time when we released the paper I think first time at 1,200 and then it fell back to $80 just to get the so kind of it's kind of today people are sweating and thinking okay it's very bad when we started it was that was even harder more hardcore now and we raised only 5 million actually

and turn it into the peak I think into 10 billion so that's kind of 2000 time for our investors I think there can be they they can be very uh happy on that. Secondly, so just to clear that we did not had the the the pockets so to say to to fight the the big treasuries but actually were you were very hardcore I would say very hardcore self-made guys to literally penetrate you. Secondly, I think specifically now if you're generally speaking, if you're lucky enough to be relevant and stay focused on your course, there's I think a a very interesting phase that maybe sometimes you're ahead too too early to you're too early with something. You're too early

with technology that you're being you're building as a founder. And if you're lucky, literally lucky enough to get to the right timing, you might actually win big time. But this is actually you could there's a saying you can have the big the best team the best product and the shittiest timing and you're dead. And this is also applies vice versa right you can have the worst team the worst product and the right timing and then so this is kind of a play where I think specifically to that the best analogy how I build it is look looking at CPUs versus GPUs. CPUs are mostly single threaded or were designed mostly single or for many many years single threaded then built a few threads and so forth

and then it came up there was GPUs for many many years we had GPUs how long it took us to build the product market fit for GPUs that will will actually build the application will generate that demand for that kind of technology. Now looking at sharding, sharding was always it was the the very complex part similar to GPUs that you could have some demand but it until this point there was literally very limited architectural applications built to towards that but I think finally we're getting towards that. I think this year specifically with the AI needs with the new narratives that are being built there's a very new interesting landscape of products that are more more than just

transactional transactional interactions within uh within accounts right so this is something that I think we're building towards with the right narrative with reduced latency increased bandwidth and then also making sure that our ecosystem is very much alive and building on top of that so I think those are the verticals that we're we're planning and then also staying even closer to the head of the curve. I think the part with the US narrative is we're going to see each other more often there as Sebastian. I think so. Tell me a little bit about the the ecosystem and like things that people are building there.

And I'm like actually quite interesting quite interested to learn more about the things that you're doing around payments and helping bridge the gap between blockchains and how people pay which is with with cards. So yeah, if you could share more about that too. Yeah. Yeah. Yeah. So more more than happy. So I think fir first the the state of the ecosystem it kind of built just to give so we we talked a lot about protocol on top of that there's a was engine a was engine or was VM that is extremely fast and we're working towards bringing parallelization into was execution on top of that and then the interestingly and very unique I think on top of that is the rust there are two frameworks two specific so it you can generally speaking compile any kind of a language

you can use any any language that compiles to wasn't We are providing a rust uh rust SDK which is called uh spacecraft SDK and then there's uh an a brand new SDK for building applications smart contracts which is a swift base which is very non-intuitive and I think uh or uh I think quite unique in the space nobody else uh in this point has any framework that you can build with swift uh language compar that translates to spark contracts on in any blockchain. This is a new uh experiment built entirely by the community actually and this is something that makes me really smile I would say again looking how how far they they can go with their imagination. Now on top of that there's

AMMs, stable swaps, algorithm. We just launched the first overcolateralized stable coin, algorithmic stable coin. There's multiple lending and borrow protocols. There's uh liquid staking. There's I think pretty much everything if I would compare to any of the most successful ones. I don't see kind of anything missing on the general block building pieces. And now this is now maturing at a very fast pace. Yeah. Now back to to the payments. I think yeah that's that's uh again I would say that it started closely more to to multiverse X but now it actually become a monster in itself each of them. So I'll briefly mention that actually it is one of one of the the spin-offs companies actually

it's called X money. It uh ended to have pretty much most of the licenses in within Europe with EMI similar license what what circle has specifically when it comes to a stable coin issuing. It has a IBN account, card issuing card uh uh processing or acquiring business for both retail businesses and then actually bridging everything delving towards a stripe merchant API with payment split that accepts any currency to any currency underneath in one batch. That's the goal. I think that's something that has not been. So you could actually pay with Ethereum, pay with Eagle, pay with Atom and so forth. And then regardless where the merchant will be in the life

cycle of the transaction, they will they can choose their preferred currency and so forth. Like one thing here that I think is interesting is um as far as I can tell the offering of kind of crypto credit cards where you you have a card that ex that pulls directly from a balance on chain are pretty common in Europe you know so we have no suspay we have holy held you know previously we had things like monolith there's there's a lot of different offerings in Europe uh also the I kind of iband to onchain um with things like manarium uh it it like works pretty well like the licenses have been issues here to do that. Um I I

don't get a sense that there are many of these products in the US or that it's not as developed. Um you know maybe the rel I mean obviously the regulatory position in the US over the last administration might have something to do with that but do you get a sense that this is changing and that we could see more of these products in the US? like do you have anything any alpha there that uh could be interesting for the audience? Yeah, I I think two two points. So generally speaking even ourself even myself uh prior to this administration I've been avoiding to set foot too often in US I think this is generally speaking for web3 and this this has quite obvious I would say to the other to the other founders uh specifically the ones the industry

leaders that kind of took a penalty for for for uh as an example for the industry now with this new administration 100% they they are open business open for business. US is literally inviting hardcore builders and this is the the most brilliant I would say market to to uh play with. However, it will still take a bit of time until we'll have fully fully clearance on all the regulatory framework as you mentioned as well. Where do you see the space going?

like you know we like I was talking earlier about how there's just like tons of L2s on Ethereum and I think a pretty solid ecosystem of application chains and and more chains I think considering going the L1 route you know as as someone who's been in the space for a long time and has seen this architecture evolve and the infrastructure evolved where do you think we're heading you know in a couple years from now looking back in terms of Will will we have more application specific chains or will uh apps uh rely on underlying infrastructure in the sense of like building L2? How do you see that playing out? Yeah. So so maybe two two points.

So very very good questions. First there's uh I I'll I'll build again or point to the point when dipsseek went out. When deepseek went out actually and pointed and proved that it can achieve even higher performance with less resources need the market first took a big crash and said look maybe there's not enough demand there's not going to be enough demand for whatever we're producing now and so forth. However, if we look at the internet history, there was never like the the lower the resources or the cost was, the more demand has been generated over time. And I think specifically toward towards that we're getting and I'm very extremely

bullish on how much demand and how we're going to how things will scale up specifically regardless of the markets. I'm not looking at the markets and never look at the markets. It's good to be aware, but as a builder, that's like shooting yourself every day in the foot. You you build something that you're visionary and you you envision that you'll generate and push with insane force to generate the demand and we will collectively generate the demand for this kind of more compute, lower cost, faster transaction, faster settlement generally speaking. Now also when it comes to specific applications, I'll take one specific use case where for example in payments specifically, we will have a lot of specialized payment

channels. There will be a bunch of maybe uh chains or private chains which will specifically route financial transaction within closed networks or with between closed institutions which will have mostly their own stable coin issuing on top of that on-ramp off-ramp and money clearance and then anything else will still remain on their own settlement chain. On the other hand, looking at the demand that we have generated for the entire space and I think uh on the on the more bullish side, looking specifically on the NASDAQ volume, curious if you have looked specifically on the NASDAQ volumes and market volume, web three market volumes over 24 hours

in a good day and in a bad day because this will also point you towards something very unexpected. In the worst days we had almost half of the market the entire 24 vol 24 hours volumes of NASDAQ on crypto and then on in the good days we have exceeded that volume by six times. So this points actually that generally speaking once the regulatory framework will be there once the the we will spam those networks with memecoins and we'll make sure that they are reliable enough we will have 24-hour open markets worldwide. This is a no-brainer that this will will just and this is just a humble I think use case that I think will will generate at least

for a few of the L1's good good demand. I would say one one thing that I've been thinking a lot about, I think the demand for stable coins will outpace the demand for um for L1's because um the so stable coin demand is going to increase with applications that don't necessarily need a lot of kind of onchain business logic, but that do need stable coins to facilitate payment. and facilitate settlement. So if you look at use cases for instance in like trade finance where you have you know container ships coming from China and like coming to say like

you know Europe or anywhere in the west there's an there's a large number of actors in there paper contracts like a lot of kind of legacy infrastructure to make all that happen and I mean this is true across all of finance but like trade finance and sort of shipping finance is is one pretty complex use case where there's lots of different actors insurance, settlement on delivery, etc. And until we had like until we have globally accepted stable coins, like those you those use cases are just very difficult. But they don't require a whole lot of like onchain logic, right? Like you don't really need a DEX to do that. You might need to do a swap at some point, but it's mostly just like locking assets in a contract and

releasing them when like some event happens. And like that's a multi- trillion dollar market right there. And so I I think that there's a lot of use cases that are going to require stable coins for just faster settlement and you can you know strap on like a bunch of kind of AI front ends that are processing information and relaying that information on chain in a more effect more efficient way. Um but you don't need a lot of like transaction throughput for that. You know it's like kind of large transactions being settled.

Um and so I I feel like a lot of the use cases that people were talking about, you know, around 201617, um when building permission blockchains, which was like sort of a big big part of the industry back then, um are going to get just really facilitated with with stable coins. Um, so yeah, that's a bit of a a long-winded way to say that like I feel like stable coins are really the main use case that are is going to be developed over the next couple years. Fully agree. I mean I I I also happen to to uh yeah be lucky enough to work on some of that. So how how lucky can can someone be? I close with that. I think

yeah was was a very good talk. Very very curious if there are any other anything else. Sebastian, no I think we covered most of it. Uh yeah, and happy to get an update whenever um whenever you guys roll out the new road map and when there's any new exciting updates happening in the multiverse ecosystem. More than happy. Sebastian was a pleasure again as every time we met. Thanks for taking time and thanks for inviting me and uh yeah, see you soon. Cheers. [Music] [Music]