DeFi by Design EP#95: Blockchain Infrastructure at Scale w Connext Founder Arjun Bhuptani

Transcript

foreign [Music] by Design where we talk all things blockchain and cryptocurrency while striving to educate Empower and enrich what's going on guys welcome back to episode 95 the D5 by Design podcast here today with our boy Rob as well as Arjun from connects and we're going to be talking about the multi-chain world seen a lot of activity flourishing on Roll-Ups as of recently and seeing a bunch of new chains come out as well as some of the kind of older alternative layer ones still stay alive so it's been a really interesting time in in the kind

of world of which chain do I call my home um you know I've had a lot of fun on Roll-Ups recently had a lot of profitable opportunities and the same can be said about a lot of different other homes so today we're going to be talking about using connects to you know travel from one home to another and how that looks some of the vision and yeah I'm here with Rob Rob it's been a little while since a pod but we're back in Action how you doing yes sir good morning doing well uh pleasure to be uh pleasure to be on today we uh you know we talk a lot about this analogy or at least we have talked about it in the past where like the chains are like Islands or cities and then bridges are like the roads between those cities that people could travel and transfer

um from one home to another um and connect provides that infrastructure uh to travel between cities so you know I think conceptually everyone kind of understands like how to travel um bridge ux is is one of the better you you user experiences that is on the ecosystem right now um technically like we've seen some kind of uh we've seen some exploits right so I'm interested to hear how it connects uh protects against those um and then kind of what what the thesis is in the future for multi-chain um it's something that you know kind of started slowly and then and then

suddenly um now there's several chains there there was just you know one or two early on and I think that Trend will continue as we have sharding more chains pop up we'll continue to have uh more and more uh multi-chain universe and you know bridging and infrastructure between those chains becomes even more important so Arjuna it's uh it's a pleasure to have you on today um you know maybe you could start about um how you got into blockchain and then we can kind of you know let the story unfold on how you got started with connect yeah for sure um first of all thank you guys so much for having me here um it's great to be on this podcast and great to chat with everybody about it um uh how did I get into crypto

um so uh I originally had a background in physics like theoretical physics I was planning to kind of pursue Academia in that direction and then realized that I wanted to have like be able to work on something that had like more meaningful impact in my own lifetime um to a larger number of people and at the time I was really interested in like this is back in like 2016. um I was really interested in public goods and I'd had a lot of like exposure to like peer-to-peer protocols um in like the early torrenting days uh but I'd never really been that interested in Bitcoin because I was never very interested in like money in finance um and uh and that was when I came across ethereum just uh at like a hackathon um and I think that was kind of when things clicked for me where I was like

okay we have this we have all these like very deep societal issues around uh you know corporate monopolization or um you know Sovereign effective monopolization right large large organizations um both corporate or government uh related that are um where they're the desires of their constituents so the individual like people in a population are not really don't really match well to like what ends up happening at the government level um and uh and to me it always felt like okay what we're missing is like a third option um you know something where you're not having to pick between some Mega Corporation and or some like you know authoritarian government you get to pick some option that is actually legitimately democratized um some option

where individual people have like the agency and authority to make decisions in some system um and the best example of you know a digital globally accessible public good that we that I'd kind of come across up until that point was the internet um and so when when I first encountered ethereum uh again 2016-ish it blew my mind because in my mind I was like this is the platform that allows us to basically expand on the key thing that the internet pioneered which was create a globally accessible uh Network for communication to now create other kinds of public goods that let us do more interesting and more powerful things things like coordinating with each other on a local or global scale and doing that in a in a cheap way that is also

provably Fair um things like uh being able to give access to financial infrastructure things like being able to uh put like social graphs and things like that on chain it like basically like have those be kind of available to everybody so that you don't have monopolization um happening around like things like Facebook um very abstract idea uh I you know smoked a lot of weed at the time so I still do but you know that was it was very much a like early phase where I think at the time in 2016 most people who were in the space were still kind of thinking about these very very abstract Concepts from a very deep philosophical level um and so it's kind of cool to see how

things have progress since then um in uh in 2017 I started wanting to become really more deeply involved in the space uh I uh started a couple of projects um I like helped kind of uh kick off a couple of like early infrastructure things in the space and then uh eventually realized that there was this need to build uh tooling that allowed users to interact with ethereum more easily um uh this was kind of second half of 2017 when the Ico boom was starting to pick up and uh started connects with the initial idea of let's find a way to make it very very easy for people to interact with dapps and pretty quickly it became very obvious that at the time it was completely impossible for anybody to interact with any dapps because ethereum

just didn't scale at all um you know the the prospect of like I need to pay 70 80 to buy a 10 ticket like nft ticket was obviously ridiculous and a lot of the early projects that we were working with at the time had exactly that problem you know we couldn't even build any sort of tooling or products for them until they solved the scalability issue and so we became one of the first projects doing r d on l2s um we actually built one of the first l2s in the space on ethereum uh for a project called spank chain which is an adult payment uh payment processor slash campsite platform uh they were one of the only projects with any traction actually back in 20 2017 2018 and we built a state Channel network a conditional payment platform that they used at the time to do tips for Camp

Stars and then also payments um uh over the course of the next like two to three years um we continued doing a lot of r d with um the broader Elder ecosystem uh most of the people that were in that ecosystem uh area at least the researchers that are working that ecosystem have now gone on to create most of the Roll-Ups and chains that we know of today so that included you know the founders of Matic uh that include which became polygon obviously that included uh the people in plasma Group which eventually became optimism uh that included the founders of arbitrim um or off-chain Labs which obviously developed Arboretum um and uh and through the course of that we realized that there was this actually very very interesting problem uh that people weren't thinking about a lot

which was in a world where there end up being multiple different layer two options how do you actually once again abstract away the complexity for users to have to interact with many of them um we we were kind of very early to the thought process and we were very lucky in that we turned out to be quite correct um so in 2020 we we started kind of uh playing around with this idea of doing you know transfers between Roll-Ups without introducing trust and in 2021 the whole multi-chain ecosystem exploded um and all of a sudden this became like a huge and very pressing need um that's kind of a zero to 100 of like how we got here um where we're going is uh even from 2021 our key belief has been in the same

way so today when you use the internet you don't really need to think about you know you you use an application you don't really need to think about like what's going on behind that application uh as a user you you know if you go and interact with YouTube you don't have to think about how YouTube is pulling data from apis all over the world or how it's interacting with like Google's like distributed database infrastructure in the background you're just using an app um and we think that dapps kind of need to get to the same place like the whole concept of I'm on I'm on optimism and my funds are here and I need to go and use this application running on polygons I need to bridge over there to use it that that notion is just like bizarre to anybody that's not in the space right

it's it's like very web one like we're in like the early web one days you were like I'm connected to my local uh College you know Network and I need to use the internet work to connect to this other network somewhere else and I have to like manually plug in that other networks like routing address so that way I can get there and then and then I can interact with their stuff like that's that's kind of where we are right now um and and I think I think the way that we get there is by allowing for application developers to have tooling that in the same way that the internet works today interact with resources that live on multiple change resources meaning data funds state users that live on multiple chains all at once without having to really think about it so in the same way that you know you use

YouTube today allowing you to interact with I don't know Ave in a way where Ave is just in the background routing your uh LP deposit into to whichever pool is providing the best rate rather than you having to specifically say I want this to go to optimism I want this to go to polygon and I think just kind of as a final note there this really breaks into you know Rob what you were saying which is just we are heading towards a world with many more chains um that's not that's not going away like we connect our fundamental thesis and the thing that we're leaning into is that there will be a thousand Roll-Ups in the next like year and a half um and when that happens like our as a user are you going to know the difference between arbitrum one and

arbitrum two uh arbitrum Nova et cetera Etc no you're not going to care you're just gonna say I want to use an app that's it foreign yeah you're going to want deep liquidity and fast throughput and low slippage low fees and just be able to interact and it goes even further uh into the the idea of onboarding people who aren't as technically Savvy who want to use Bridges um and who really don't may not care as much per se or may not be as as deep in In The Weeds on the technical side it it again just expands the vision that you have of multiple chains hundreds of chains and basically uh obfuscating a lot of the processes that you know we

let's call this D5 power users do on a daily or weekly basis and so a more uh you know just fluid ecosystem kind of makes me sad man it's like one day there's going to be no like hard work involved which is going to be also easy and seamless yeah that's when all the alpha dries out exactly yeah precisely and another part that's a little bit sad is like it kind of loses its home feeling because you're not really aware of like what what chain you're actually on um maybe the the community transcends more into like just kind of like the group chats you're in and not necessarily like the actual chains but more of the the brand recognition will would be on the

actual uh protocol uh like the decentralized application rather than the chain yeah perhaps I mean I think um for what it's worth I think like what we're probably gonna start seeing so like right now the the main end user facing application is investment right like that's like there really are very few things that exist in the space that are getting end user traction that are not uh like Finance oriented or some sort of like Financial upside oriented um and it's just it's just kind of the nature of where the space is right now is people people like believe in crypto and so they want to get involved and like if you're not a developer that is capable of building things the way that you're getting involved is by investing in things um or if you're not in a position where

you can go and join a company to work on things but I think lens is probably the only one yeah um but even even with lens right like I would I think the vast majority of people that are using lens are probably existing crypto users that are that are you you know it's like it's providing a use case that is not um like there's there's a ton of others too like you know that are providing use cases that are not you know perhaps not like uh Financial appreciation or appreciation oriented but they're still targeting the same like core User Group um and I think the reason for that is like if you really think about it right like what is the what is the end user use case for for like Cloud there isn't one right cloud computing as an industry the the end user facing use case like

the the use case was let's go and integrate into a bunch of infrastructure that allows people to build more powerful web applications which can then themselves like interact with end users um and I think that that's something that like I I think that's kind of the the weird place that we exist right now is like we have a lot of these like incorrect signals from the market which say oh well users actually care about the brands of the chain but really they they only care in so far as their investors like and then developers care about the brands of the chain and the communities and stuff like that um but ultimately like some end user that's out there that doesn't really care about crypto at all and just wants to use an app they're only ever going to care about the app that's not that's

never going to change yeah that's interesting that that that takes the idea of like the fat fat protocol features sort of like the layer one or the base layer thesis and kind of reverses it on its head as far as just powerful apps that will scale because if you think about it it's like uh in in the reference that you make with the cloud there's there's tons of different apps on the internet from Facebook all the way to various softwares things like this and each user doesn't think about Internet Explorer Firefox or Brave it's more of like you know I'm going on Instagram or I'm going on um you know telegram right so um yeah I think I think what you what I see connects providing for the ecosystem is a means of uh allowing end users to

use whatever protocol or app that they choose without necessarily minding where it is only for whatever use case that they have so if it's a financial one right they can be they can use the app in the place where they can have the the biggest gain uh potential or best risk reward Ratio or or kind of the highest yield right if it's a community-based one where the largest communities are nfts kind of all a similar process of thought where Bridges like connects are just basically a pathway but I still find that in the current stage of the market there are I think users that are currently here which is is obviously nothing compared to 2021 and also

nothing compared to like say five years out um I I think they still have a home like you know they're still three to four chains that they use on a daily or weekly basis right they they'll trade on GMX on one chain they'll use Ave on one they'll buy nfts on one they'll use a specific yield app on another little trade options potentially on one um so you know I think that there's still something to be said about users finding their own homes and again connects just allows them to kind of test out different basically areas to to live if you think about ethereum as a digital nation state and then kind of plant their Roots um you know how they like yeah for sure and I think that that's that's kind of the big sort of asterisk

here is like we are still at least a half a decade away if not longer to to getting to the point where this stuff is like fully abstracted um and it's it's going to be I mean I think everybody always just like grossly underestimates how long infrastructure takes to build and how long some of this lower level stuff takes um kind of getting back to your point like this is this is part of the reason why like this is part of what this this whole Space unlocks right it makes it very easy to go and build applications like you can go and fork and deploy an application in minutes now rather than having to like take a you know like months like you used to um both in this space and out of the space but then the flip side of that is like the

infrastructure itself needs to needs to be so stable it needs to move at a slow enough Pace it needs to be conservative enough that it can then support that level of innovation um and so I really do kind of firmly believe you know we're still a ways out it's a it's a good like directional indicator for what we should be trying to lean into and trying to find ways to continue to like abstract away chains but fundamentally it's going it's not it's not something that's gonna gonna happen overnight uh there's gonna be like some awkward in between steps um yeah and uh you know we have we have for example some some things that we'll be rolling out soon um that allow like tooling that allows for applications to enable their infrastructure to be interacted with from any chain so that they you know as

an application you don't have to be like oh I can only target users from my chain I can you know you can massively broadening your Market but again like there's still going to be some gotchas where users are going to be like until users have like a wallet experience that actually supports all of the chains they're going to be like oh well my you know my funds transferred across chains to interact with this app and now I'm entered into like a position on this chain but like how do I actually track that position on that chain well now you have to go and perhaps use something like the bank or or zapper instead to do that rather than your own wallet so it's going to be like kind of incremental progress yeah it's slow and steady I think account abstraction is something that's big that that's happening which

for those listeners who are still a little on the edge because it's kind of confusing I'm gonna do a video about it this week but effectively think of your your metamask wallet as like almost like a wallet in your pocket and then think of think of an account abstraction like a smart contract wallet as like a smartphone like a venmo um so I mean it's a little bit more uh advanced than that but it's that's the kind of level of Step um that I think that will take for user experience as well as basically onboarding more people to the experience um and then yeah like you said just kind of more of that front-facing uh technology takes longer to build than just forking synthetics as code and you know dropping a new app or a you know a

new um you know kind of pool or something like that but I'm curious uh kind of Shifting Gears here unless Rob's got something super pressing well I've got I've got one more just kind of like similar as far as why or or June why why you think the complexity of the user experience is causing the industry to only build Financial applications yeah that's a good question um because right now blockchains and this space don't scale enough to actually build anything else um because if you think about it right like we're basically operating like a TI-83 calculator right now if it's not we're not like like the the

kind of code that you're writing on chain that like is just it's like 600 lines of solidity code that you're deploying and it's costing you like ten dollars per interaction to use that's that's that's never going to work for any real real application right for any like real use case that goes beyond um Beyond something that's like just fundamental very basic very simple public good um I think the the broader idea here is like you want to allow people to like this industry allows people to build these like very very sophisticated public goods eventually um and in order for those things to be fully incredibly neutral public goods they need to not eventually they need to work to work towards not actually having any centralized dependencies at all so that means you know being deployed on chain having the compute run on chain having

storage run on chain even having the uis be served from chains you don't have this situation where like a UI gets shut down and so now nobody can interact with the app anymore um until we get there like we're gonna get there in fact the cross chain stuff is probably going to be what gets us there like the combination of being able to interact between chains and being able to deploy many application specific Roll-Ups very quickly that can have their own kind of parameters around hey this one's better for compute this one's better for storage Etc so the combination of those two things is going to make it so that you can have these like full stack decentralized applications but until we get there like we're sort of just kind of we're basically just kind of like hacking it with certain things that we know can like we know can provide

value today and we know can like be incredibly neutral that are quite basic and like at this stage it's still very very simple stuff yeah makes sense like lack of scalability prevents these full-scale applications that serve use cases Beyond Financial ones exactly like I think eventually and pretty soon I think what we're going to start seeing as this space now I mean we we have the scalability now with with Roll-Ups and with caution communication we can actually do this today um what we're going to start seeing is stuff like decentralized Wikipedia with like legitimate incentives for for like high quality editing right we're going to start seeing you know basically any any situation where you're like okay there's this thing that exists that's a public good and that everybody interacts with that is you know uh that should be

globally accessible and shouldn't be owned by any singular company you can probably put on chain and it can be quite a useful thing um so things like Google search for example right you can you can eventually I mean this one's a little bit further away but you could eventually have like an on-chain map of the internet and allow people to search like do graph traversal through that map and that becomes a replacement for Google um just like the search aspect of Google um and again incredibly neutral on chain public good yeah and bring AI into the mix and then and then things get really weird because like AI is replacing search and you can bring AI like onto a blockchain and make it credibly neutral and I guess people could maybe like the incentives

you know you could you could benefit from training it or using it somehow yeah exactly any situation where you're like I mean I've actually I've tweeted about this once but I think the Playbook here is really simple it's like if once this once the technology in the space gets to the point where it's mature basically it's going to be like a some uh one of my one of my friends uh amine Founders Bank chain and has done a bunch of other stuff in the space as well and we've collaborated quite a bit um but he like his I love his terminology for this for what this space will eventually get to uh which is a decentralization laser beam um you're basically just gonna have you're gonna be able to play this repeatable Playbook which is going to be look for web 2 monopolies um any web 2 Monopoly you find figure

out what the means of production is for that Monopoly so what is the thing that they're actually building mode around and decentralize that so like Facebook uh it's their social graph these are decentralized that into lens uh you know uh Google it's the search graph decentralized that into like an on-chain on-chain map of the internet shoot the decentralization laser beam right at the web to Monopoly yeah it's gonna be it's gonna be really cool if it works yeah yeah it's super solid yeah yeah that's crazy yeah ai ai will be another I mean we're you know starting to see the first Monopoly there um but yeah I mean the peer-to-peer nature of it then just paves the way for incentives

um to be built around the moat and then like I'm almost glad that we we have ex like as an industry we have experience working in Dows that are crafting uh incentives um at a scale for these financially based apps which require like pretty significant incentives sophisticated incentives before we start shooting this laser beam at like web 2 monopolies because we're gonna have to have like pretty um sophisticated incentive structures to overtake you know those web 2 monopolies or or compete with them yeah for sure yeah yeah I agree with the super super uh super bear Market podcast from

the industry not being you know prepared and mature enough to We're All in This Together yeah I have the max amount of hopium in the bear and like in the in the bowl I get so like if we do this podcast again in the next bookmark it you're gonna see me be like the most jaded crankiest old man in the space foreign I don't blame you I mean all the all this all the bad stuff happens towards the tail end of the ball the hacks the exploits and the the drama and the and the crazy gains and just absolute garbage but yeah it always comes and goes and I mean yeah on the topic of kind of the uh the end of the bowl and kind of the the jadeness I think a lot of people in the industry watching this as well approve

experienced it in some sort are somewhat jaded from from Bridges because of the amount of exploits I know vitalik wrote an article about it way back when and different cross chain multi-chain it's not a crush chain world it's a multi-channer vice versa and I I'm just curious from a more Technical and security perspective what you guys have learned from uh the the let's just say the failed or the the bridges that have uh had you know some issues and you know what what your vision and Outlook um and model kind of is for security for connects yeah um yeah I mean it's good that you're first of all great transition into this because uh uh it is it is a really important topic and you're totally right like I I think everybody is concerned about this and scared about this because

it is a big problem for the space um I think bridges are and like cross-chain communication is like the first it's like it's a primitive that is at the same level as blockchain the blockchains themselves right fundamentally um it's like you have blockchains and then you have the substrate that connects them um it's something that people have been thinking about for literally decades now and there there really hasn't like we're start you know it's it's been thought about a lot it's been theorized a lot but there hasn't been a lot of like practical application until recently because the demand just wasn't there yet but of course everybody was like yes it's going to get there eventually um and arguably this is also why Roll-Ups exist like Roll-Ups are explicitly a solution to this problem where like in you know back in 2017 2018 everybody was like okay how can we allow

for Two Chains to communicate with each other ethereum and a side chain but do it in a way where there's no possible way for that side chain to to actually like rug its users and the only way to do that was to have the side chain explicitly require rely on ethereum for you know ordering of transactions and for for data availability which then became a roll-up that was then it was then called a rollout um and so you know this we've been kind of circling around this problem for a very very long time um I think it's not surprising that given where the market was given how suddenly this became a problem like a like a user problem and not just a theoretical problem how many hacks we've had um like I think it's this is like

you know when we were doing this with Roll-Ups there was like several years before people even began using Roll-Ups where we had time to like work on iterate on designs and think about like what is the best kind of way for this to work and by the point at which the first roll up started to be developed Roll-Ups is a construction were finalized um that was not true of bridges um in fact when Bridges start started to become popular in like early 2021 um and started to be used there it was nobody really understood how they worked at all and there was just a myriad of different ways to do it and the vast majority of the ways that it was being done were just fully custodial it was basically like someone running either a single server or running like a multi-sig and relaying transactions

between chains using that and like not understanding why that's not good enough um and why that's actually problematic not just for users but also just for this ecosystem more broadly I think now the level of sophistication is improving um which is good uh being in a bear Market where the hype is lower where people actually have time to build and aren't just trying to like capture as much value as possible is really really important um and I think a big part of that is also that we uh as connects and I think more generally the ecosystem is now pushing towards a more sophisticated understanding of what cross-chain communication actually means um and and how you can minimize the risk of it so one thing that we've put forward recently is this idea that um you know if you take this primitive

which is a messaging Bridge so of course the word Bridge means a lot of things to different people um you know transfers of value transfers of messages Etc we're gonna just refer to all like the the base construction which is a messaging Bridge which is let me get some information from one chain to another you know uh that information could be related to a token transfer but it doesn't actually matter um if you take a like a messaging Bridge as a Core Construction something that we've observed over the course of doing research for the past few years is um really that core model can that core primitive gets broken up into three different layers um there's transport which is the core the like simple process of getting data from one chain to another and not actually making assumptions about what

happens to that data everybody ends up building this themselves but it's it's fairly simple you know you're just like it's just kind of relaying um effectively uh and we you know you can use out of the box tooling for that which is nice you can use stuff like by economy gelato Etc um so that's something where you know it's like a it's a well-established layer of the stack and there's really no risk there yet on top of that comes how do you actually secure the message that you've now relayed across chains um and that's the place where I think most people are getting tripped up so historically the way that this has happened is you know some sort of multi-sig there's now a push towards more kind of like trust minimized mechanisms things like like clients you

know ZK like client implementations that allow you to do like more trustless communication between L1 chains um and then of course there's like the roll-up Bridges themselves which provide L2 to L1 communication that is trustless um but again like these are all these are all things that are being built right they can still get hacked and like all of the bridge hacks that have happened have not happened because of some trust they so far at least they've largely happened because like someone's implementation had a bug in it um so that really that verification piece is where like the vast majority of the security risk lies and then lastly on top of all of that you have what's called execution um which is what connects does execution is uh when you after you have taken data across chains and verified that it's

that it's correct how do you then do something with it on the destination so like from a developer perspective you know how do you call a contract on another chain from a user perspective how do you interact with the DAP on the on the destination chain um and usually any kind of application is built on top of the execution there because you don't as a as a app developer or user you don't have to run infrastructure to do that yourself um so again execution not as risky right again it's it's a it's more of a fairly simple thing that needs to be decentralized for censorship resistance purposes and stuff like that but it's not inherently providing the like it's not contributing to what we call the the surface area of risk for for bridging so let's focus all of our efforts on verification um and this is this is kind of the way

that connects differs from everything else most and actually I would say all of the other bridges that are live today basically just build this whole stack themselves and then they Implement their own verification layer um that is used in the same way across all chains connects does something slightly different what we do is we try to delegate that to whatever the best mechanism is for verification in a given ecosystem that already exists so for example um if you're interacting with an application on optimism that application is relying on the optimism role of bridge there is no way that you are going to get out of trusting the optimism role a bridge is going to work right because fundamentally if that construction fails then the entire optimism role implodes there's nothing that anyone can do about it so for uh

communication that happens to and from optimism connects uses the optimism role of bridge we actually passed the message through the roll-up bridge and if something goes wrong we fall back to the optimism roll up seven day exit window um similarly on arbitrary we use the Arbiter mullet bridge on polygon we use the polygon POS Bridge because that's the one that is underlies all of the assets on that chain and then for other systems where there isn't a canonical Bridge available or isn't something that's very very established we're working towards using a basically like an aggregated verification approach um there's been some chatter recently about this this model called Hashi which is a a multi-bridge multi kind of verification mechanism that Aggregates you know a bunch of different options

Aggregates like Wormhole syncs Etc um to create one system where you would have to have the majority of all of those other Bridges fail in order for some hack to occur um and so from our perspective we would then delegate verification to them we would sit on top of Hashi be an execution layer for them and like fully have them control Security in the system we believe that by doing this we like connects and in general this space can get like a level of security that has not been seen at all so far for Bridges and really is like the end state of how you can do secure communication between chains so the in a nutshell that is if I'm sending a message from arbitrary optimism um that is going through the arbiterm

roller bridge to ethereum and then from ethereum to the through the optimism roll-up bridge to optimism there's of course some bells and whistles on that we batch those messages uh significantly to cut costs um we do certain kinds of things to like help reduce the latency of those transfers and then we have a system that exists on top of that to execute the message on the destination chain and in many cases that can happen immediately but in a nutshell you're still kind of relying on the existing infrastructure that is already securing the space yeah very interesting to hear you kind of break down the components of a bridge and how most people build the entire stack themselves you guys focus on the execution layer and delegate the security part the

verification part to those who are building that natively like optimism on optimism on arbitrum and it allows you guys to focus on the fun part which is the scalability like the execution is where like you can actually go and do something on that chain and not just not really like have to worry about the verification of the security part because you guys are plugging in the most secure bridge that you could possibly plug in because it's the canonical bridge on that on that chain how do you like I guess if my question is more like on the latency because does that mean like when you got when someone Bridges through connects they're

basically you know bridging through the optimism uh verification client and you guys have to wait the seven days like you mentioned there were some bells and whistles I guess I guess like this is probably about as technical as as we could go but like what what are some of the bells and whistles you guys put in to reduce latency when going through one of those native bridges yeah that's a really good question and that's that's actually the main thing that I think um one of the main pieces of value that we provide as an execution layer um that you know where we we do it in a way where again we're trying to minimize any additional security overhead we're minimizing any additional trust overhead compared to the underlying Bridges but of course like you know this is this is part of the the like uh uh the part of

the secret sauce that connects is providing to a developer like an application developer that wants to build a Crossing app um so there's there's two things that we do um the first is what you mentioned about how do you deal with like the seven day you know optimistic roll-up exit um so the one of the there's there's kind of a couple pieces there so first is the the seven day window is is a somewhat arbitrary um it's and not just arbitrary it actually represents something that's a little bit different than people think it does um a lot of people believe that the roll-up when you when you make a transaction on the rollup that transaction isn't final until seven days in the future that's not actually true um your transaction on a roll-up finalizes as soon as the rollup operator posts the next kind of batch of

transactions so post your transaction to ethereum because at that point you and anyone else including people that want to dispute the roll-up can verify that the sequencer has not committed fraud so why do you have seven days you have seven days because you need to give enough time for the entirety of the roll-up state to actually be posted to ethereum um because the Roll-Ups are all pushing for evm equivalency right now that actually requires interactivity it requires you to play this game where like uh you know say I'm disputing the optimism roll up like I would need to I'm I'm a person that's like watching the system has caught fraud I would need to put forward the fraud and then optimism the optimism sequencer would need to reply with some State and then I would need to reply with more fraud and they you know like you would have to go

back and forth on that several times to get to the point where the entirety of the optimism state is replayed on ethereum to prove that something has gone wrong for bridging you don't actually need to do that all you need to know is you need to know is there fraud in the first place because if there is you can fall back to the seven day window to adjudicate where did the fraud come from but you don't need to wait the full amount of time to know if they're like the kind of binary thing of is there even fraud to begin with so what connects does is we actually shorten that window significantly um we wait for we we pass the message through the roll of sequencer so the only way that fraud can occur is if the rollup sequencer actually commits fraud in the first place which again you

would have to assume that optimism is actively committing fraud then on top of that we wait 30 minutes um and in that 30 minutes we have our own system of Watchers that is basically checking to see if any dispute has been initiated against optimism now it stands to reason again because of economic incentives and because of the way that this like the way that optimism works and the way that people watch the optimism roll up that within 30 minutes at least someone will have initiated the dispute obviously the dispute will take seven days to finish but within 30 minutes you know has fraud occurred on chain um and so we only need to wait 30 minutes when exiting the roll up um and that's and you can be very very confident that that can happen with a very high degree of security because if something went wrong

in order for something to go wrong you would need basically the optimism sequencer whoever is sequencing the optimism chain at the time to commit fraud and for all of the Connex Watchers to not notice that none of the optimism sequencer Watchers or the none of the optimism sequencer Watchers also didn't notice and on top of that all of that needs to be worth it to the sequencer who is already putting up significant amount of value to stake uh in order to sequence the roll up in the first place um so generally speaking we think that that's like provides effectively the same level of security for the kinds of things that you're trying to do um and this is kind of the this is also like the the core basis for how like cross roll up interactions the core kind of thought process for how cross-rollope

interactions using a shared sequencer will work in the future we can touch on that in a little bit but um that's a little bit more of a technical topic but it kind of touches on the same same core point um the second thing that I mentioned that we do um is we uh we do what is called Fast execution and this is a really interesting concept that we've created um and that actually makes sense script economically so um there's this model that people have been doing for a very long time called Atomic swaps uh Atomic swaps is like how the OG way that people are doing interoperability uh on you know you had all these Atomic spot protocols yeah exactly you had all these Atomic swap protocols between like Bitcoin and ethereum and what you would do is you'd put up some BTC on bitcoin

into a multi-sig and then I put some BTC on ethereum uh in a multi-sig and then we would like basically I would I would like you would uh you would give like one of the signatures to unlock the multi-stick and I would give the other one then we could basically atomically unlock both at the same time so that you get your funds and I get my funds um and there's no way for only half of that to execute um what's really interesting is that that core idea is basically I am paying I'm basically letting you as a service writer I'm letting you as a counterparty um front to me capital on the destination chain right front the BTC to me on on bitcoin uh in order to then receive my eth and I I can ensure that that happens atomically because uh because of the mechanism that has been

built underneath that core mechanism can be generalized it doesn't just need to be for transfers of value you can actually do that for any instance where the counterparty so the person who you're interacting with is uh wants is is or sorry the any instance where the counterparty is able to call a contract on the destination chain so let's let's take a little bit more material example say for example I have my funds in usdc on arbitrum and I want to trade usdc for eth I want to do it on optimism because optimism has the best usdc for each price and it even including the gas cost of getting there it's worthwhile to do um now obviously I could bridge and you know bridge out of arbitrim to ethereum bridge optimism Etc or I could do

something like bridge to optimism and then swap on uniswap there um but all of those things are are just a bit of a pain in the ass right like as a as a idealized user I don't have to like think about any of that I just want to go to uniswap and use the app so one way to do that is for me to uh give you know say I'm swapping 100 100 usdc one way for me to do that is for me to put that 100 usdc into escrow and for me to go to Rob and say hey Rob can you uh swap 99.5 usdc on optimism two eth and then send me the eth on optimism and if you do that programmatically I can basically ensure that this this the 100 usdc that I

blocked in escrow on arbitrum will be paid out to you now in effect what I've done is I've paid you as a service writer to just front capital for me and to do the transaction immediately um and you're willing to do that as long as you get paid back obviously with some fees um that core concept is is what we're calling fast execution um it's something that you know it's like similar in concept to the fast liquidity model that a lot of people have been talking about for exiting Roll-Ups but more generalized to any public facing function what this means from our protocol perspective is this is really cool for any user facing interaction our Network can just execute it immediately um the reason it can do that is because we have a network of off-chain liquidity

providers and executors called routers those routers just front the capital to execute the destination chain transfer so a unit swap call for example and then they will get repaid by the protocol because there is a message that goes through the the slow path going through ethereum uh from you know arbitrum to optimism and when the message arrives on optimism routers can basically claim against the funds that you paid them on optimism or on arbitrary to begin with and so that is that's like the mechanism of what effectively lping on uh a connect looks like those are those are the routers and then the fees on what our users and what our community sees on Bridges or unconnects the fees that they have to pay which is often times higher leaving a roll-up going to mainnet

versus going from mainnet to a roll-up obviously outing gas fees um as part of it that that fee is what you just explained as well is that kind of fronting process um you know and there's I know there's risk in there but the but the uh yeah is is that right or or is that UI front facing fees separate from that kind of fee that the uh that the person who's fronting uh earns yeah um yeah that's a good question so there's there's one little Nuance here which is there's two types of liquidity providers in connects there's like the active liquidity provision and they're not just providing liquidity they're also doing a lot of like execution and stuff so there's like basically the nodes in the network which are routers

um and what they're doing is they're charging a five VPS fee every on on any like fronting of capital that they do um which really it seems like nothing but routers are still earning a pretty sizable amount and we're only just getting started um so it's pretty awesome um and then it's a really efficient business right you're basically like locking up like liquidity for an hour and you're getting five BPS on it you can you then your liquidity unlocks and you can lock it up again so like you can do that you know uh you know somewhere anywhere between like up to like 12 times a day or something like that so if you have 100K in liquidity you're basically getting 12 times 5 BPS per day uh in fees uh and so over the course of a year that's like you know uh several hundred

percent annualized um uh but there's a there is another kind of liquidity provider on connects which is the passive liquidity um this is a this is again a bit nuanced uh the reason that passive liquidity exists and connects is because we want to make sure that we're you know as a as a user you're getting the right right asset we don't want to give users like a wrapped connect asset or something like that so what we do is on both at the at the source like when you're getting started and then at the destination we'll swap from like the canonical assets so for example polygon POS usdc into the connect specific asset which is next usdc uh that can next usdc is basically a pointer to usdc lockdown ethereum so what we're basically doing is pricing your chains liquidity against ethereum

liquidity and then that next to usdc of course is controlled by connect so we can burn and mint it across chains so uh just I guess Andy to answer your question um the fees as a user that you will see will be the five BPS fee from a router and then you will see some like slippage associated with those amms um that slippage could be positive or negative usually I mean it's a stable swap um uh with a you know which are obviously quite efficient so like the the slippage that you typically see is very low I mean we've done like very sizable transfers during like the usdc d-peg for example and like the max slippage that we saw was point four percent yeah yeah it's always really really reasonable even with size yeah

um you know even yeah pretty much any chain as well um the Roll-Ups and ethereum always have really good liquidity for sure I mean from my experience so yeah um Arjun kind of wrapping up here uh one last thing that we asked a lot of our guests as as we close up is if you had to explain connects to one person or one entity who would it be and what would your quick elevator pitch be yeah the next is a platform for cross-chain application development right we think uh every single application in this space is going to at least be interacting with users from across chains if not be natively deployed across multiple chains in the same way that today as an application developer you build a web app that

interacts with resources over the internet you will build a cross-chained application like cross-chained app that will interact with resources all over the interchange um uh the way that connects works is by delegating security to the canonical bridges of each ecosystem so you're basically retaining the same like trust and security as the underlying roll-up or underlying chain um and the goal is that by doing that we can provide the best most secure experience for people to build this new wave of applications that is coming nice nice you missed part of the question as um who would you explain it to but I'm just going to go ahead and develop Elon Musk oh Elon Musk um yeah I mean our ideal basically our Target user

is obviously we have the end user Bridge UI and stuff like that but like in my opinion I think bridges are going to Disappear Completely like it doesn't make sense for them to exist it's sort of like you know it's sort of like uh I don't know it's sort of like email like it'll it'll probably exist but it's just like eventually you're gonna have chat baked into everything and you're like why should I use email anymore um except for spam uh and and uh and so I think our Target our Target user the people that we really really want to talk to and the people that we want to interact with connects uh and build on top of us are developers people that want to build new and interesting applications sweet well thanks a lot for coming on Arjun I really appreciate it it was it

was a great chat looking forward to see what you guys build um and also see you guys out there in Lisbon I'm sure and um yeah man take it easy keep on building yeah awesome yeah thank you guys so much for having me yeah it was a pleasure man thanks for coming appreciate it