Where Everclear Fits in The Modular Thesis

Transcript

morning welcome back to the rollup today's episode is with our friend Arjun with Ever Clear formerly known as connects and ever CLE is the first rollup clearing layer these guys are building in chain abstraction they've been thinking about chain abstraction and intense for a really long time pioneering the terms back when they were connects and Ever Clear is now a fresh new take on chain abstraction which we believe that the rollup is the solution to the modul fragmentation problem before we get into today's episode first award from our [Music] sponsors cartey is your home for app

specific rollups at the Forefront of developing new tools and platforms that enhance the capabilities of blockchain Technology cartey enables applications that are flexible and built to scale build on a full Linux environment and use traditional software libraries but in blockchain development go to cy. that's cy. to learn more about building your own app chain today welcome back to the rollup today we have our friend Arjun from Ever Clear to discuss the clearing layer chain abstraction chain abstraction stack maybe some cross Shain me some modularity toss in there a little bit as

well Arjun hasn't been sleeping so bear with it him but welcome to the Pod sir it's good to have you back thanks so much I appreciate being here um generally super super excited for this conv s awesome man yeah I mean congrats on uh on the upgrade the launch whatever you want to call it uh no longer connects uh now Ever Clear um generally yeah congrats I mean how how do you think things went yeah um honestly I mean I it like hugely surpass our expectations we've we've this is something we've been like preparing for for quite a while we really wanted to like nail the like messaging around it it's a it's a fundamentally New Concept and this is

not the first time that we've like introduced new ideas to this space we've historically like we've actually very often been the first to do a bunch of bunch of stuff in the space we've like deployed the first L2 we like built the first intent based Bridge we introduced the concept of like modular bridging and stuff like that and chain abstraction but I think historically one thing we haven't done a great job of is like making sure that from day one there's like a very very clear picture of how this works and like like a very very clear understanding of like what exactly does this solve and like how does it solve it and um I think this time we really double down on on making sure that right out of the gate there's like a consistent way to talk about about ever clear and around clearing in

general for this space um and it seems to really have paid off like a lot of people you know like we we went I went into to the launch super nervous because I was like you know this is a really technical topic and like are people going to get it you know I was expecting it to kind of go in the direction of something like reaking where like it took it took a while for people to get reaking right even though it's such an important primitive um but honestly we we've kind of had the opposite experience where like a ton of people just organically reached out to us or just like co-promoted the content or kind of talked about it publicly because um they were all just like able to understand you know that the original tweet thread the original blog post to the point where they were able like able

to talk about it which is kind of cool just before we get into it like where do you keep getting all these like original ideas honestly I think it's just like the kind of person that I am um I you know there's there's good and bad things about this I tend to be the kind of person that like really thinks really far ahead on things and I'm I'm also really oriented towards like trying to find new problems to solve or new ways to solve problems that people haven't thought of before or like new ways to think about problems from like first principles um rather than competing on a head-to-head basis with other projects like I just in my in my my personal philosophy is like okay if there's going to be like 20 projects trying to solve XYZ problem like there probably doesn't need to be 21 and like effectively you

can consider that problem solved unless someone has like a genuinely different way to do things or unless there's like a like a way to kind of rethink the market from first first principles I think a part of it is just that that like we're we're as a team we're like really not afraid to just like kill our babies and we've we have like rebuilt the whole system so many times in like so many different capacities for so many different reasons and like we're not really afraid to do that because each time you do it it gets like so much better you learn so much more about it um but I I also recognize that that's not how those teams operate and there are there are tons of downsides to that too like we've always we've not always but we've often been very early to things like sometimes to the tune of

like years early because you know you're if you're thinking ahead you're not necessarily looking at like where the market signals are today yeah and you guys I I think it was arguably you could say that you guys were years early to chain abstraction and now like that that coming to the Limelight so like yeah I mean take us through like your train of thought from you know having that chain abstraction idea originally coining that term to ultimately killing that baby that connects baby and now like birthing a new baby in Ever Clear yeah so I think it comes down to like where what our background was so like when we first started uh really building stuff in the space the whole thesis was like how can we make how can we make ethereum simple

enough that anyone can use it right like I think we were one of the first teams to start talking about this idea of like how do we scale to a billion users um and and that for for us that meant not just like the technological scalability of the blockchain but also like the user experience of like you it is actually usable right um and um and I think like you know we we were like what when we were doing a bunch of L2 research with with other other folks we we kind of saw like okay well there's a bunch of different hypotheses around how to build l2s but like everyone's philosophy was like okay there is going to be an L2 not like obviously some people were thinking of it as like there will be a project but you know in our minds that was like

there is going to be an L2 ecosystem that needs to be unified it needs to appear as if it's like one single system um and so when the space started going in like 2020 or so it started going towards like okay there's going to be like dozens of different rollups for us it was like okay well there's clearly something missing here which is this experience of like it should feel like a single unified rollup it shouldn't feel like there's so many different options um and uh and I think we were like we started talking about that in 2021 like you know we we kind of had a bridge UI then and that was taking off but that was just something that had like been built on top of the system and it wasn't something that we cared about now this was this was an example of a case where

it was a bit of a mistake where like you know in 2021 what we probably should have done was just like focus on being the best Bridge instead of focusing on the thing that we actually did which was like trying to convince a bunch of applications hey like stop making your users Bridge just like give them a single experience where you deposit into your system from any chain like why do you why are you sending them to another UI you're losing like 80% of your conversion as a result of it but it's just it was the bull market and like people didn't necessarily Care at the time um and like there were other problems bigger problems that were like worth solving then for those ecosystems rather than trying to solve this one um but it's like for that kind of thing it's like it's only a matter of time

right we knew that at some point people we were going to hit this this same hurdle like applications were going to hit the same hurdle of okay I am having to like I'm a and I you know I a user has lent a bunch of of tokens into a pool on arbitrum and like they want to move that to optimism I'm going to go send them out of my app to some completely different application to go do that and they might have a really terrible experience that I just can't control at all and and then I'm going to tell them to come back and like will they come back probably not um you know so we kind of realized like at some point that has to change yeah that's the thing in 2021 the opportunities were so uh fast and everything was going so so

crazy that bridging was just like it was something that that you would do um and and yeah trying to I remember like in Paris even last year the chain obstruction toolkit idea you know it was still it was still something that teams were kind of accepting but like it was still like kind of new but now now you know push towards ever clear so maybe we can talk about how you're visioning the chain abstraction stack and kind of why ever clear is a is a core part of that stack um and kind of one thing that we're starting to realize big time with with conversations from leaders in the space for chain of distraction is it's very much so a collaborative effort in terms of their ideas for how we're gonna

kind of make this happen so like it seems like this with the idea of the clearing layer is like very much so focused on one part um that if if done very well can be perfected and can help the broader space so maybe you can kind of shine light on this on this broad idea here yeah um so to do that I guess let me talk a little bit about like some of the ch Alles that we experienced with the chain attraction toolkit last year when we released that um you know as you mentioned like it was something that was really interesting to a lot of developers but not everybody jumped on it and like we saw that there were two reasons for that the first was the market just wasn't there yet like you know every developer every project we talked to was like yeah this is this is

something that we we like actually need like we you know we we want to be able to expand to more ecosystems we want to allow onboarding for users for more ecosystems but right now we're just struggling to like just keep the lights on because that you know this was still the bare Market people were just it was just like everybody was in the like mv's hierarchy of needs they were just trying to feed themselves um and um and I think that so that was that was kind of like one key problem we were like okay this is this is something where like we just have to kind of wait until these projects grow a little bit and then and then it makes more sense the other thing that was an interesting catalyst is so so something that that really like catalyzed that change was was like what happened with Renzo um I

think Renzo was the first protocol to really just go all in on this idea of you know but if you make it really easy to restake that is such an enormous competitive advantage to be able to do it super easily from every other chain that like that in and of itself can be like a massive differentiator and Driver of growth and that thesis ended up being super correct um obviously in retrospect it makes sense right like and this is something that we like we and many other people have been advocating for for a while but like I think for most application developers there was just no way to test this or validate that like you know it's I'm going to grow more by making it super easy to onboard from anywhere versus like trying to double down on like the arbitrum ecosystem or whatever ecosystem I'm in right now um

the other reason that we we struggled was actually the reason that ever CLE kind of ended up getting created which was um we realized okay if you want to go support applications to onboard from any chain um and you know you're talking about defi applications ideally they want to work with any asset um and there's like a growing number of chains and a growing number of assets this just becomes an extremely complicated problem really quickly and like we realized that there was just there was fundamental limitations to how quickly we as a bridge protocol could get everywhere with liquidity right and so we realized okay this is maybe too big of a problem for one single project to solve like I don't and I and this is kind of where I've ended up personally is like I don't think that there's I really don't think

that there's a world where like there's going to be one in 10 protocol that might just deploy every single chain like it's just too much overhead and like rebalancing especially is like the big piece of that like how do you get liquidity back it's a big piece of that like no one really has an answer to and so we we kind of in the process of thinking about that we we started thinking through like okay like what after we released the chain exraction toolkit and talked with a bunch of people tried to do some Integrations we realized okay we need to get to the point where we can we can support like onboarding from literally every single chain um and to do that we need to solve this like underlying clearing like clearing rebalancing problem um and so the more we kind of like double clicked

on that and started focusing on that the more we realized okay you know what this is something this is basically the same problem we've been thinking about now for three years since since 2021 and that this is such a big hairy problem that like every single intent protocol is just is just also going to run into it and like just the kind of the same thing is like you know I guess this is really how you find like most of the interesting opportunities is like we just kind of double clicked enough to till we got to something where we're like okay there just isn't an answer for this and like it's such a huge problem that everybody's gonna have to deal with it um so I think the second part of your question was like how how does this actually like fit in what problem does

it solve I've kind of alluded to it a little bit already but like in effect when you're when you're making intense happen across chains what's what's really happening behind the scenes basically you're like there's there's semantic debate as to like how you kind of Define this and things like that but like in effect right now what most people mean is that there's a user describe something that they want to do and some third party service provider comes and does it on their behalf so like literally think of task rabbit right or like Uber or like you know bolt or whatever any any other like any other Service app where what you're doing is renting someone else's time and expertise and sometimes their like vehicle or whatever it is that they can provide to do the thing that you want to do um that's that's effectively all that

an intent solver is they're just sort of like providing this stuff they're taking on the complexity and providing liquidity as a service um the challenge for intense and intense solvers is really simply like if I am if I am try trying to go and deposit 100K of liquidity and I I have my liquidity on like nosis chain and I want to go deposit that into a on arbitrum it's super easy for some third party service Rider who has 100K of liquidity to go and do that on my behalf right so they they're just fronting their own Capital they deposit into a that can happen in seconds right we've seen we've all seen how incredible of an experience AC cross is like it's it's just it's such a it's a magical experienced someone else can just do it on your behalf it can happen at web native speeds right we can I'm

certain that we can get this down to like subsecond speeds over time um so what is the core problem well for you as a user there's no real problem at least in this instance but for the solver now the solver has ended up in this position where they have paid 100K of their own liquidity they need to be repaid by the intent protocol how does the intent protocol repay them well the only funds that the intent protocol has are the are my original 100K back on nosis chain so what will any intent protocol do it'll repay the solver back on nosis chain and so now the solver who had 100K of liquidity on arbitrum has 100K of liquidity on nosis chain that is very meaningfully different situation to be in right like there is a ton of inflows coming to arbitrum all the time like

that 100K is going to get used up constantly but who knows what might happen on this sh and like who knows you know if a user is onboarding for from some like totally longtail like L3 this Sol liquidity can just end up somewhere and like they're they're just stuck right there's no way to get out of that um so how is this how is this handled currently for a limited number of chains um really like the top eight chains or so with the top like two to three assets solvers basically just through it they are they like build if what is effectively like an in-house aggregator they'll like connect to a bunch of bridges they'll connect to like OTC desks exchange it centralized exchanges Etc and they'll just try to figure out some way to get their funds back and they they'll bake this into their

pricing where they're like okay you know like based on the pricing that I'm able to to give like this is what I this is how much it's going to cost me to rebound um and this is complex obviously like operationally um every time you add a new chain you have to do all these Integrations you have to think really carefully about pricing stuff like that and it's also it's also like this is where you know the the kind of centralization vector comes in um so actually immediately before recording we started talking a little bit about this but um there is this like economies of scale that you get here where like if you are a larger solver that has you know tons and tons of money say for example you didn't just have this 100K that you're routing from nois to arbitrum but you actually had a100

million do sitting on arbitrum and nois chain well do you care so much about rebalancing right now no not really right like you're just going to keep the money there find some of it's not getting used but it's a small percentage of your toal total Capital so you just kind of leave it there and you'll deal with it later and in theory at some point in the future because you're like less time sensitive about rebalancing at some point in the future someone may make a transfer from arbitrum to NOS chain your funds can end up going back so this is this is called netting where like you have B directional funds that kind of like flows that cancel each other out um and who can net really effectively the big solvers right like winter viw can net really effectively because they have so much money that

like for them it doesn't really matter who can net effectively any IND solver um and and so now you can see why we we have this like world that we live in today where there's like you know not really that many solvers operating very successfully um and they're still only serving like the top few chains um so this was the kind of core problem that we saw we were like how do we how do we get to the point where anyone can run a solver on any chain with any asset and they can do this in a way which is like economically feasible right where it's not just okay we're going to explicitly operate this at a loss because we have because there's no other way to make it work um this is the core problem that ever CLE solves how does it solve it well we

already kind of looked at the solution right like the winter mute has the solution how do they solve it they have tons of capital and they just like net out their flows between between chains and because they're netting out of all of these flows um they're cut dropping their costs of like rebalancing down per transaction like they're advertising it per transaction down to the point where they're able to be super competitive right they can give better pricing than anybody else on these on these intent based Bridges indexes and overall they can kind of like win the market um so the way to kind of emulate that is to allow for a bunch of smaller solvers or really anybody to be able to net off between each other in the same way that n winch mute Nets off their own capital and so what this means is this allows

like anyone who wants to kind of use ever CLE and net off their settlements to have the same economics as a larger player like like winter it effectively fixes the kind of centralization risk of solvers so okay but how like let's let's do an example okay so like I'm you Robin I are we have you know $10,000 and we're participating in the clearing layer of every clear H like are Rob and I interacting with each other like aggregating like a pool of funds and then going to work to like compete with the solver or is it like am I adding mine in with the with the other solver or like kind of how do how how does a small player like us get to part get to participate and also like win the

auction like to beat out a bigger player so let's talk about let's talk about that a little bit because we we alluded to a few different things so you mentioned auctions right auctions are basically the process by which uh a solver gets selected to fill an intent so who who you this is basically like when you click you know find me a ride on Uber there is like it kind of is pending for a while and then it matches you with a driver that's the auction um the process that we're talking about with ever CLE is not how do you get that it's after the driver drops you off how do they get back home right how do they get back to somewhere where they can continue picking up rides um so just just want to make that that piece here now the the interesting thing is the

amount that a driver needs to like drive to get to another place where they can pick up another ride depend determines their cost basis right like if they're having to drive like if you drove you took an Uber out to the middle of the desert for an hour and then they had to drive an hour to get back they're having to pay gas and all this time on the way back where they're not earning any money and so you know in theory their cost basis has gone up now I I don't really know how Uber handles this but if we wanted if we want to handle this in like a like an open decentralized Fair system like then they're in theory the Uber should be charging a higher price for the next ride that they do because it cost them so much more to get back to

where they could even do that ride in the first place this is kind of how intense solvers work it's like the solver has to charge a higher price if they they ended up with fund somewhere totally random they they're going to have to charge a higher price offset that um the how it happens is an interesting question and basically this is this comes down to almost like cow swap like coincidences of wants right so this is this is what netting is netting basically is is a concept where say for example Andy you were doing you you know you were a solver and you made a transaction happen from like you know optimism to arbitrum Robbie did one from arbitrum to polygon I did one from polygon to optimism right so Andy because you did optimism to arbitrum you paid funds on arbitrum you ended up back

on optimism right uh with your funds now ideally like say for example you're a solver you're like I don't actually really want to work with optimism at all I only want to work with arbitrum um I'm gonna have to go and rebalance my funds Robbie kind of has to do the same thing where he's like okay well I you know received this transaction arbitrum I wanted to end up over here and and you know you can see how like we all kind of ended up in the in the same position where we all ended up with funds in a different chain that we initially anticipated what's really interesting here is that Andy you ended up with funds on the the chain where I want funds Robbie ended up in the funds and chain where you want funds you you know and so like it's a it's kind of a loop

um if we all communicated with each other if we had a platform to talk to each other about it where we where we could all just say okay well I have 100K on this chain and I I want it over here and we all just said that and put it in the middle then we could coordinate with each other to just basically just like swap with each other directly on the same chain without ever having to like really go and rebalance funds in the first place and that happens at effectively zero cost the mental model here is if you guys have used splitwise uh that's a really good analogy because what is what is splitwise doing splitwise is an app that like allows you to split a bill or something like that with your friends right and like what you all do all you and your friends you

go on a trip or something everybody just registers the amount of money that they've paid for stuff right and like or the amount of money that they owe to each other and at the end splitwise Aggregates all of that and just gives a final amount so my final amount can be I owe you know Robbie $10 even though you know technically andio is me and IO Andy and and you know it's like it it basically is reducing all of this like tons of transactions down into like a final amount that everybody actually owes each other and as it turns out that this was something that really shocked us when we when we started looking at the dat we had this like hypothesis around netting uh because we were like okay there there has to be like a certain amount of n netted value in the

space right the thing that shocked us was when we started looking at the data that number on a daily basis that number is 80% so 8 80% of the flows between chains like net out based on like public flows this is not even including the private flows where if we think you I think if you include like centralized exchanges otc's things like that that number is probably close to 90 um what that basically means is we're paying like five to 10 times more for bridging today than we need to because that means you know you can see how this kind of trickles up right so it's like if we're bridging five to 10 times more it's because LPS and solvers and market makers and centralized exchanges are balancing 5 to 10 times more right and and if they're rebalancing 5 to 10 times

more that means they're charging 5 to 10 times more for their basic Services than they would have otherwise and then that kind of trickles up the stack till it gets to the user um and that that means the user ends up paying that that way higher amount which is kind of wild if you think about it yeah it is it is wild I mean these the bridges the bridge experience of going from one chain to another when someone else is just doing the same thing back why can't we take care of that on on the chains themselves like that's a major unlock and all the entire time you were describing this like my dad is an accountant and so this is like an accountant's wet dream to like just net everything out and debits and credits just match up it's like yeah

like it'll just it'll just everything will net out and it'll be more efficient and this is B essentially like an accounting thing right like it is so you guys are are managing debits and credits from all these different accounts and then ultimately you're going to have this this method of netting everything out on the on the clearing layer um does this mean that like fundamentally we're going to be taking less Bridges because one view of chain abstraction is that we'll we'll still use Bridges but the users won't use them it'll just happen behind the scenes and so do you kind of tend to agree with like that vision of the world or do you

think like we won't even really need Bridges because these solvers will just fill the thing and we won't really need to take the canonical paths or like the general messaging paths because we'll just have this intense Centric like solver view of cross chain yeah let's talk about chain of production because I think I think that will help clarify this a lot as well because you know the notion I think the notion of bridges is very overloaded in the space and I think a lot of the challenges that we see today with Bridges like unreliability or like um like high costs or like lack of support for a lot of different chains like all of it seems to boil down to a few fundamental issues and like those issues are why chain abstraction and why intents need to exist so I guess let's

talk about bridging in general right so in 2021 when we when the space started conceiving of like Bridges as like a a thing that everybody was using on a day-to-day basis no one really had their arms around like what this stuff should look like or how it should work we had like we had the benefit of experience and that we were coming from like a like an L2 Centric background and so our full philosophy around Building Bridges was always like it should rely on the basic trust assumptions of roll step s and it should not like add to those trust assumptions and so our mental model was like okay the rollup bridge is the bridge and what we are doing is just providing a user experience layer on top of it that makes that experience not have to go through ethereum basically and not have to take seven days uh in

contrast a lot of other projects are coming from a background of like well there's you know two L1 chains and we need to have these L1 chains communic with each other which arguably like was kind of the situation back in 2021 and for that model right everybody's mental model was like okay well we need to have we need you'll need some validator set no matter what happens no matter what model you use you need some validator set to attest that something that happened here some message that was over here is needs to go over here and that it's done correctly so initially that started out in like custodial mechanisms then it was like multi and then you sort of like move higher Fidelity right so you have like more centralized systems like a single custodial part a multisig a posos you know Network like AEL or

something like that and then like clients and like clients are sort of like using the the validator set of the origin chain or the destination chain um now all of these are are effectively doing the same thing which is just like using some validator set for security what's really interesting and this is this is something that we kind of had an intuition for but I'm realizing the rest of the space didn't really because this was this was a specific discussion around l2s and around how L2 development what we realized was like if you are for for this base layer of just let me get some data from here to here which which is the under underpinning thing that that sits behind like fund transfer and like apps and whatever right for that base layer the more you secure the base layer the more

validators are involved in securing a message the more expensive it's going to be that's like if you kind of think about it that way it's not very controversial right it's like this is the fundamental blocker for L1 blockchains as well this is literally the the blockchain scalability trama is like the more decentralized you make the system the more secure you make the system the more it's going to cost the less likely it's going to scale this is where we are today um if you look at like what you know if you look out at like the bridge landscape you're kind of seeing people are starting to specialize So like um you know projects that were in the past sort of targeting I think this is one of the big issues with cross chain back in 2021 and and still kind of persist to this days like projects just

didn't know what to what to Target or what this stuff looked like so they just said we're just going to do everything um and so like projects like like that try to do everything we're just not able to kind of focus on like one specific piece but over time we've we' started seeing that like you know for example hyperlane um has like specialized on message transport which is just let me get data from here to here and then you can plug in whatever security module you need um layer Zero's also kind of gone in this direction um however what you notice when you do this is like okay well if I try to send a layer zero message from one chain to another and it's secured by a like client it's too expensive to be usable right it's at least at least as part of the normal

user flow and I think that this is one of the like key challenges that that like limits the ability for transport protocols like layer zero to actually be used in applications because and this is why like most of the apps that are running on top of layer zero today are using a much like much less secure approach simply because it's just too expensive to use like polyhedra Li Pine everywhere um this is not really like shade towards them or anything like that I think it is it is the right move to have modular security but it's just a fundamental limitation which is like security costs money and in the same way that like security costs money for L1 blockchains and so we shouldn't have normal day-to-day activity happen on L1 blockchains we should instead have an L2 where things are happening we should

batch up a bunch of transactions and only put proofs on L1 that's exactly what the idea behind intents and chain abstraction is Right chain abstraction is the L2 to bridging effectively um right we've our the whole hypothesis behind chain abstraction is we actually for rollups at least we have very secure Bridges the rollup bridges are like already have already solved this problem of how do you get like canonical assets how do you do the like Baseline secure message passing but in order to make them usable you need an L2 you need something that allows users to like do very fast interactions between chains that that is UN that is that is like Limitless in terms of scale right that scales out horizontally infinitely in the same way that like rollups scale out horizontally infinitely and then you just settle back to the underlying

Bridges you settle using the underlying messaging protocol that allows like an intense solver to get repaid um this is kind of the way that we think about this space and when you when you say when you kind of when you start seeing it from this perspective and I I think like most of the chain I think really all of the chain abstraction ecosystem is starting to see it from this perspective when you see it from this perspective of like chain abstractions the L2 to the transport layers L1 um you realize that something like clearing must exist right you're there's no otherwise no way to kind of provide this experience like you know we we're not going to be able to provide a bridge experience that's actually satisfactory until you can separate out like the lower level security pieces from the user experience welcome aboard on our

journey into the singularity this is the endgame road map for fracks and fractal the new L2 out of the fra ecosystem that is totally based on fractal scaling fraxel is a modular L2 blockchain built by fra with modular and fractal scaling in mind fractal scaling is based on the mandal BR set of fractal images and it explains how l2s become l3s and the entire app chain life cycle goes through its Evolution the frax ecosystem is made of some of the top builders in the space and has stemmed from its OG days of the

frax stable coin it is now in the LST space the LRT space and this frax L2 joins the existing Suite of fra products including Frack Swap and frax lend you can see frax fairy the bridge to go from any chain arbitrum optimism or main net into the fraxel main net Bridge your tokens and bring your FRA assets into the fractal ecosystem join today to get the best out of fraxo you are still early join today take part in our Expedition into the singularity makes total sense and you know when we're

thinking about like this this clearing layer and how it's really built for l2s and this like has almost has to be done and then these General messaging layers or and transport layers more built for L1 I'm curious like just to double click like even more on like the how like we have this this clearing layer and I think Coincidence of want is like a great mental model for like how these things get netted out and so I'm curious like what the integration process looks like if it's like an account at like a user level Ops in to this clearing layer and then they can start to solve or like and and maybe it's not or but it's an and the solvers also opt into this clearing layer and then get to get

netted out against this these uh these other solvers like when when you're thinking about where to bring in the accounts that live on all these different chains you know you start to get this like accounting landscape or Ledger I guess right which is made up of all these accounts on different chains and so then you're starting to pull in the state of the blockchain or at least the state of the accounts on these different chains I mean that is like a problem that I think lots of people are trying like the shared sequencers the composers like all of these chain abstraction Protocols are really trying to tie together and get a comprehensive look at the state of multiple blockchains at the

same time on a on at like one vantage point so when you're thinking of pulling in state or pulling in account state from multiple different chains how are you how are you doing that and like how are you able to sync up the state of multiple different chains when some of them have different block times some of them might get reorg right like how are you how are you able to reconcile the state of all these chains yeah it's a it's a really good question specifically because this gets into the exact problem I talked about where it's like if you want to avoid these hurdles you have to add latency right you just have you have to like you want to avoid reorgs you have to wait time so you can have like pre-c confirmations and things like that

but you can't have like an entire protocol relying on that because it could just be easily attacked um the answer to your question for us specifically is that we don't we actually our and this is this is this was was an explicit decision it was a hard decision because you know it's like when you're building a protocol in this space like you're you're sort of making bets on like where is the market um and like who is going to be the user right and so like there's a very obvious user group of people like individual users Traders Etc that want to transfer between chains like everybody understands that market um but the market that kind of sits behind that of like okay here are the service providers to that market that market is still relatively newer we are making a bet on that secondary Market which is that like

you know yes there is additional there's downside to like previously connects now Ever Clear to like to to like give up on focusing on like the end users that are making the transactions but our philosophy is that like that needs to be soled higher up in the stack um we think like approaches like one balance or like magic spend that like socket has been talking about like basically mechanisms within a smart account to lock funds within that smart account conditionally based on some intent being solved that is the right way to do it because that's like within the user user layer and then because it's within the user layer the user sort of like permissioning this themselves you can make that very trust minimized you don't need to have like

communication between blockchains you can just read the state of a bunch of blockchains and the user can be like okay I'm I'm okay like locking up my state within the UI because there's something within the UI that's saying like I'm committing to making this transaction and if it doesn't go through within like only if it doesn't go through within a couple of hours will it get reverted back to me um you know that that to me feels like the only way to kind of overcome that complexity um now you've mentioned a bunch of other things like shared sequencing and things like that right so that so it's like the user the flow that we kind of see is like user kind of blocks over their state or something like that to like Express their intent to do something a solver goes and fills that so the user's

interaction at that point is done and like you know we just don't have to think about the user anymore which I think is a really really good outcome um the solver can like batch up a bunch of these interactions and then go and figure out how to settle and they can settle with every how that settlement happens is where things like Jed sequencing plug in so like for example this is one of the like one of the open benefits of like of every clear is that even Beyond The netting which I think is the immediate benefit the the secondary benefit is that the system is effectively like a settlement aggregator not really an aggregator but basically it's a settlement negotiator coordinator the system plugs into like shared sequencers in different ecosystems or things like IBC for Cosmos or things like you know say for example like the the token

that's being settled is an of you can plug into layer zero for that right um the idea is that there's like like an unbounded open design space around how to optimize settlement um for example uh like token is tokens are kind of doing this all the time right now where like a token may work with a bridge protocol to get bridged into certain ecosystems but not others they may use canical bridges for some ecosystems and whatever that just what that's doing is basically creating a really fragmented back end like settlement back end right so like fra is a really interesting example of this where they've just had an evolving strategy and like it completely makes sense I think some chains they've they've like issued fra using canonical bridges in other chains they've done it using fra fairy and then other chains

like few other chains they've used axar now they're looking at other things as well how if you're an intense sver how do you rebalance that it's like a really interesting and open question like how do you rebalance that optimally the answer is you don't really right like you would have to basically go and take your funds Bridge them to ethereum um you know Bridge from like L to ethereum using axar which is the the canonical bridge for frax there and then bridge to using frax fa some other chain and like you're paying e gas fees at that point it's just not worth it at all um the idea with Everclear is to be able to go and like negotiate these things so the the way that it could negotiate that is saying okay this solver made a transaction go from you know arbitrum to lanaa where arbitrum the the canonical

mentor of frax and arbitrum is frax Fury the canonical Mentor On Lena is Axel um but what this what the protocol can do is it can it can it can net off a transaction that's going from like um you know like scroll to l which are both supported by Axel or scr sorry scroll to like like polygon right where similar situation scroll is supported by Axel polygon is supported by by uh frax fairy it can basically net off those two things against one another such that it looks like two separate transactions right it looks like one Axel transaction going from scroll to Linea and one fra fairy transaction going from arbitrum to polygon and none of it needs to touch ethereum like it like and the more we look at it the more we realize this is actually everywhere in the space it's a huge [ __ ] problem this looks a lot

like um like extensibility where Bridges had this really like is this really pressing issue where they weren't able to reach all of the different chains and so we continue to see more app chains come out more rollups that's like it seems like that's only continuing in in the direction of more rollups more app chains and so these Bridges had a real problem that was part of the issue with connect was that you weren't able to reach all of these chains and with all of these assets right and so and so now you were able to solve this bridge extensibility issue by going and and saying no we're just going to solve it one layer above all the bridges can still operate they can still go they'll end up you know through a various amount

of bridges they'll end up reaching chains but then you have this this clearing netting like layer on top that says we're just going to coin like people are already taking the bridges there we're just going to net it out so that finality happens quicker and at better execution and so the end benefici beneficiary as the user because they don't have to wait or even take the bridges themselves like that'll just happen underneath this clearing layer and so the user ends up getting their their tokens quicker and more affordably um yeah and then by by go just then by by virtue that you're able to reach all of these chains that the underlying Bridges can't reach themselves exactly that's like that's a

big part of what we we're kind of like interested in here is that you know I I mean there there are going to there are hundreds of chains coming like I think people don't real like I have talked to so many teams that are building rollups and like so many more than you would expect um like I would say just about every project in the space at this point has talked about it internally or at least thought about it internally why because in many cases a lot of projects are either already doing it or already doing a bunch of stuff that is like similar so any infrastructure project like connect included in the past like you know we iterated on our design for several years and when we looked really closely at our design last year we were like wait this is a rollup we just we

built a rollup we didn't even mean to build a rollup we just ended up building a rollup and like and then and and it kind of like the light bulb went off and we were like okay well why don't we just use an existing rollup stack instead of like building a totally custom thing and I I think I've heard that story from a lot of different people in the space so that's like all the infrastructure protocols and then all the D5 protocols have this similar problem where they're like well okay there's going to be like there's going to be hundreds of game chains there's going to be hundreds of different like different things out there that like ecosystem specific chains things like that that get get get spun up like am I really willing to go and redeploy my protocol and bootstrap liquidity on each and every one of those

no it just doesn't really make sense and so you know you have Pioneers like FR that are that have kind of realized this and are like you know what let's not do that let's consolidate all of our liquidity on one chain fine there's some fragmentation we can fix that using Bridges and intense that that user experience will get better but our liquidity is going to be one place and because it's one place we can provide the best possible pric in compar compared to competitors that have that have just spread out all over the place um our philosophy is like I think I'm actually willing to make a very solid bet that there will be you know a thousand rollups next year at this time um and not there obviously a lot of them will be quite small um but I'm willing to bet that that will be the

case um and I think in that world you have to kind of ask yourself well like there's a thousand rollups and there's 350 days how are Bridge projects going to deploy three rollups a day to get to that point and I think the answer is that they're just not right like I think we we have gotten to the point where we we got we were like okay we can do one per week and that was we were like that was just like operating super efficiently I don't know how you can speed that up Beyond making you know your Bridge totally permissionless and if you want to make it totally permissionless that's what ever clear solves right like Ever Clear can go and be the like Canary deployment into all these ecosystems we're working with all the RAS providers like you just Mission LLY deploy every clear you know there

doesn't need to be any liquidity or anything attached to it you can deploy it to a chain or to a new rollup and then that can be the base layer on top of which like dozens of intent protocols can just Just Launch very easily the extensibility problem that you guys are faced with your confident that with ever clears clearing layer plus a you know a network of solvers or intent based protocols that is like the solution for this extensibility problem for the interoperability ility space genuinely I mean I spent now years thinking about it thinking about every single option right like ZK proof Singularity shared sequencing like those things are those things are super important and they they are very critical aspects to how we will improve interoperability but they will

be latent right like ZK proofs are like if you want to have single slot finality instant transfers across all chains at all times like using only ZK proofs we need to have proofs that are generated for all like indexing all chains and the data of all chains that happen in like like I mean faster than arbitr block time so 100 milliseconds I mean I don't know how quickly we're going to get there it can take a 100 milliseconds to just communicate like across the internet today um it's like yeah so I think that like if we really really want to solve this we I think like we need we need chain abstraction we need intest we need the L2 of interoperability and if we have an L2 of

interoperability we need a shared sequencing of interoperability which is effectively what ever clear is there was something that Andy mentioned on our last podcast just before this with uh Stefan from one balance and um it was like let so let's take this to its logical conclusion you know a thousand rollups by maybe next year is fun like there's gonna be way more than that right and so Stefan was like basically no like every user is gonna be its own rollup right and and like I like now that I'm starting to wrap my head around that it's like well with the with proper extensibility that's possible and so like the question was kind of like well okay you know if I'm buying coffee at Starbucks on my account then why does

that have to tap into like the global state of like everything everywhere maybe it doesn't and maybe we localize consensus and net out you know like my transaction with Starbucks versus like you know Starbucks and that Starbucks is roll up or something like that I how do you think like something like that local consensus could look like is that you know H how do you how do you kind of like see that playing out on some of these more like cluster type of rollup structures that don't necessarily look just everything trickling down to L1 settlement I mean ultimately everything will trickle down to L1 sedon but it's just a matter of like when and how that happens right and like I like I so I like this buying coffee of starbu

analogy because I think that there is a very very direct analogy that you can make to like the way that payment systems work today um where you know when you go and you you know you buy a coffee at Starbucks like Visa doesn't actually move your funds that would be insane like you know it it costs a lot of money and it's really complex for banks to actually settle between each other Swift and like fedwire and like AC is actually I mean a is a little bit more efficient but it's like it's still like not an easy process it still costs a lot um I think to the tune of like 15 plus dollars per transaction um uh I think that the um I I think for like a day-to-day payment where you're you know buying something for dollars or perhaps less

you you need this mechanism which is kind of how intense Works where like some actor in this case Visa goes and like authorizes the transaction basically says I'm going to put my own credibility online and like I'm going to just like I know you have the money and I'm just going to like tell the merchant hey you will be paid at some point and then the merchant at the end of the day goes and says hey like here are all of my receipts I'm going to ship them off to Visa Visa gets collects all these actions nuts them off and then goes to the banks to like negotiate the um this kind of two-phase approach actually exists everywhere in systems um I think like it's interesting because it's like you you talk to people in a bunch of different like ecosystems with with or not EOS in a different sort of like

perspectives you know there's some people that have a finance perspective some people that have like a distributed systems perspective a cryptography or information Theory perspective in all of these cases you have the same overarching pattern of like when you want to make updates to the St some shared state or a system that involves some shared state it is like there is always like layers to how those updates are made um this exists for like clearing and settlement did it exists for like databases with caching um like both reading and then also writing into a cache um and then it also exists for things like uh like commit reveal schemes right where you you don't want to like uh you want to like compartmentalize who has access to which information at what times um I don't think that model is going away like I

think what we are doing in crypto as we always have been is like just rediscovering the same core principles and like running up against the same like fundamental artifacts of Nature and physics that we like okay we have to do this there's no other option for it um I think like uh so to answer your question Robbie like basically like what we need to do in order for the system for like if we want to maximally scale ethereum right which we which we do we want we want to get to the point where where there will be like quadrillions of transactions per day on these chains um if we want to get to that point where you can run the entire world on this computer how do you scale it maximally well you have to have like tiers of consensus where you have the Baseline

maximal consensus right which is the set there that is the most secure form of consensus you get with the largest amount of participants um and uh and you know then you have like sort of modularity right you have you have certain aspects like data availability where you can have a smaller security set because it just isn't as important um and you can kind of move up the stack to rollups that have even smaller security set because they're not as reliant on like they're they're inheriting security from stuff underne and then as you move up the stack I think to intense with intents what's happening is it's actually even smaller there's no there's no like public verifiability at all and intent is basically a onetoone relationship we called this um we called this locally verified when we when we first like uh talked about like the interoperability

tral on back in 2021 I think our our understanding of this has evolved quite a bit where we we sort of see now like local verification as a slightly thing than the other forms of verification but local verification just means like okay no one else needs to be involved in verifying this like I trust you because I know you have money on the line and like I can verify that myself and that's super easy for me to verify like in the browser just verify a signature and that's it very cool yeah makes makes a ton of sense Andy did you have a did you got something there yeah I just think this like tiered approach to consensus is like is something that we're just going to slowly slowly get to so kind of

excited to see that roll out I think that the the last thing that that we should touch on is just kind of how the development of of the clearing layer is going and and just kind of what the what the next couple months look like for you guys yeah that's a really great question um so I mean we're we're live on testnet as of as of the announcement um which is awesome I think like um you know I think for for us we're a team that like really believes in having something tangible before we start talking about it so it's nice to be in this place where we where we have something that people can start like interacting with and we're already talking to a bunch of a bunch of protocols um like a really awesome ecosystem of folks that have that have like been interested in this um and it's

it's Fant like a testament to how positive Summit is is that like all that ecosystem is made up of all the protocols that were previously our competitors right so like like it's it's really great that like they're all just like yeah everybody has this problem like we we don't know how to solve it we don't want to solve it we're focused on this other much more important thing which is which is totally fair um what'll do the next so we're basically from here there's a few things that we're working on um we we have a initial kind of vzero implementation for how like the pricing and netting works in the system um that we talked about um as part of like the launch thread and blog post that uses a Dutch option um uh in the longer term the idea is as I mentioned like settlement is just like

has unbounded complexity so what we want to do is just like allow for anyone to build their own strategies for settlement and have them compete where like solvers can just opt into like what the best strategy is this is really cool because I think like there are like the ideal outcome to head towards in this case is is to like is something that's like truly data oriented right so you you can use data to not only like optimally net between part Partners but also like preempt figure out where should liquidity go or where should we have deposits available to accommodate for future netting needs and that is a like data science machine learning problem which is very interesting um I have a little bit of a background machine learning as well um as as does Lane so this is this is something that peaked our interest a lot so I think I

think the the direction I head expect this stuff to head is basically like uh eventually get to the point where people can build any kind of like arbitrary settlement that could maybe do something like post like a zkl proof um like of computation over like historical data of flows into and out of arbitrum to figure out when are we going to see potentially like an increase in flows or something like that um and then in addition to that plugging into stuff like cctp for usdc or you know fra fair for fra whatever um uh what we are doing between now and launch is like working on kind of finishing development of the the Dutch auction system and like kind of getting feedback and testing of that and then starting to think through the designs of like what does this other piece look like um and so uh and then aside from

that just like the basic stuff stuff of like you know audits um you know Security reviews and other things like that getting feedback from Partners Etc um we're targeting going live in sometime in Q3 awesome awesome one uh one quick question for you and I I know we could have a whole another conversation on this but like maybe just maybe just at a high level like you know the me like again like the layered consensus thing does that open up me at each layer like generally without getting too too too specific like how are you thinking about me in this new paradigm yeah it's a really good question so um so there's there's Meb kind of a little bit higher up in the

stock and then there's there's like how this this problem at at the Ever Clear layer higher up in the stock um there is a so any any instance where there's like information asymmetry there's there's MB right like you can have you can any instance where you just some people have access to information before others um you you can like take advantage of that information um the idea with like intense and intent networks is that the more you can get to the point where there's like a single like unified public option where everybody's expressing their intents the more likely it is you can have an outcome where there's like fairness where fine there there might be like opportunities to front run but everybody becomes aware of those at the same time and so then it's not as much of a problem anymore um but then the flip side of that is also true

where it's like if you if you have for like certain kinds of use cases like it's just better to not have that option at all and just have things be in a private private option that's like I think a really interesting and ongoing area of research anoma is doing a lot here um I know suab is thinking a lot about this as well um to be honest it's still like a little bit out of scope for us so we're we're sort of like we're keeping an eye on it as something that's really interesting but we we're more interested in like what does this mean for solvers so for solvers specifically it's kind of it's It's a there's a there's sort of this like interesting situation where you know solvers want to like net off against one another and and reduce their costs but they also like

solvers are necessarily adversarial right even if they're I mean they're all like friendly to each other but like they are adversarial actors economically and they're like competing with each other on on filling these intents and so if a solver is like preferentially being settled to uh in any way then that's a form of M because basically that means that like some solvers over the long term are going to win out over others for this one of the key design elements of ever CLE so Everclear is an arbitr morbit rollup um and it it's like working fraud proofs and things like that one of the key design considerations around it was we want to make sure that everything that the system does regard both like now that you know we're building it ourselves and then also in the future when people come and like build their own settlement

strategies we want to make sure that every single thing that the system does is publicly verifiable and in this case publicly verifiable means like it's posting a proof to chain of like okay a bunch of you know intents came in and a bunch of solver settlements came in um in in our system we we call these invoices um so a bunch of invoices came in those invoices have to be processed in order because if you break that order then you're kind of unfairly settling some people before others they might be getting better pricing um and so you know you can ver ify that ordering you can verify like the strategy that is used because there is a proof posted by the RO chain and if if the ordering is wrong then that is a that's a fault of the rollup right like and that's something that was flashable just per

like normal arbitrum orbit rules um for us I think that's like a really important piece now there is there is like one other piece which is how how does the system even ingest new intents to begin with and are those order correctly um that's one which is actually a much harder problem that to be honest I think 's one team that we've been talking to about it that that maybe potentially has a solution um but I think that that's one which actually requires like encryption where we have to sort of actually actively like hide information associated with the inbound invoices going through like like hyperlane for example into the system like when hyperlane like we use hyperlane to kind of read read new intents or new settlements for solvers

coming from Every Chain um we need we would perhaps even need to like privatize that such that uh there is no way for someone to know okay well this these transa like these settlements are going to be ordered in this way and so I could potentially try to PR on them anyway tldr it's a very very hard problem um I don't know if there is a single SE bullet solution I think a lot of it is just going to come down to like management and like continuing to find ways to like compartmentalize information in the right way awesome man appreciate you uh staying staying a bit longer uh over time and uh answering our questions um congrats on the launch minut it's been really impressive to uh to see connects into ever clear and Beyond thank you so much I really appreciate it yeah thanks thanks for

coming on always a pleasure sir and uh look forward to the next couple of months got a lot going on and see you in in Brussels yeah yeah see you guys in Brussels I think it's the beer uh yeah for anyone that's listening we're we're gonna be heavily involved in ECC and so if you if you're if you're going to be there definitely come and say hi hell yeah we'll see that