Chain Abstraction Fireside: Insights From Builders | Vivek Gupta | Brendan O'Toole | Kushagra Singh

hello everyone uh it's pleasure to be on the Denver stage and talking about chain obstruction before we begin I will uh request a quick round of introduction from VI and Brandon here uh anyone can go first Brandon you want to go first uh sure yeah so my name is Brandon otou I'm a senior product manager at agoric um bit of background on agoric agoric is building orchestration capabilities that enables developers to create userfriendly apps that allow the seamless uh movement of assets and services across chain hi uh first of all it's great to be here uh in among some of the brightest Minds uh in the web3 space and ecosystem here my name is VI I am the CTO of octo octo is a uh end to endend chain abstraction solution where we are uh basically building a SDK using which users can express developers can express their intent that will be orchestrated across the chain uh using multiple layers and components uh the main intent here being to simplify the usability and the experience for the app developers and user experiences for end users as well that's amazing so uh that brings me to my first question like right now now web 3 seems very fragmented we have multiple chains multiple protocols there are different consensus mechanisms everything seems so clanky so what's your opinion on that you Brandon you are going to first yeah um yeah so yeah it's it's it's already obvious at this stage that fragmentation is a big problem in the industry um I think the it's what's interesting about fragmentation is it's allowed innovative solutions to run parallel because people have had to figure out how to coexist but I think the the real challenge is not the fragmentation itself it's being the lack of connective tissue that has bridged these ecosystems together to uh act as conduits for that liquidity so um you know there there's such a huge addressable Market out there if we can all bring it all together and that's what we're trying to do with orchestration we're trying to um bring these players together we're trying to abstract away the complexities to to enable this amazing you know desirable endgame of of this this ux and um you know these are the real problems that we're focusing on that's amazing uh V your opinion right I think it's a very good question uh because the fragmentation in the web3 ecosystem today is real I mean uh there is a whole selection of networks protocols liquidity solutions that are spread across uh with their own set of uh complications uh that make it very difficult for a developer to build on them right because uh a developer has to worry about how am I going to orchestrate my transaction across multiple ecosystems how am I going to uh orchestrate my transaction different chains what if the user has liquidity spread across multiple networks how am I going to aggregate that liquidity and then execute the uh transaction and uh the reality is that's not what the developers or the end users eventually want right if you ask an end user what do they want want they want their intent to be solved right for example if I have an intent to invest in a per protocol then I should be able to invest in the per protocol without having to worry about what the network is what the protocol is where my liquidity lies and the same applies to an app developer an app developer should be able to develop adapt without having to worry about which network am I actually executing this transaction on how am I going to execute this transaction on Etc so the real question here is how are we going to achieve that and for example how Brendan mentioned about creating this connective tissue uh in this fragmented ecosystem I believe that uh it is a matter of creating this solution at four different levels I mean first of all you need to solve for coordination then you need to solve for execution then you need to solve for data and liquidity and then finally settlement of finality on chain right so uh take the example right example take an example of the purp uh example that I was saying so example if I'm an end user and I wanted to make an investment in a per protocol where my assets are actually lying on three different chains in three different ecosystems some of it is lying on Solana some of it is lying on Aptos and some of it is actually lying in say ethereum right how am I going to then be able to execute this uh transaction say on a hyper VM right or hyper evm right so the what what you need to do you need to first of all find out what is the best way for me to aggregate this liquidity then you need to be able to execute these steps and aggregate that liquidity and then finally be able to execute that transaction on the target Network and then do the final settlement and so if you are able to do all of these steps and execute this transaction across all of these step in a single click of a button or a single expression of intent from a developer then that is the unified ecosystem or that is the unified blockchain world that every developer or every user is actually looking for right so I think fragmentation is real but the end State here would be where everything is Unified and automated the that's amazing explanation that uh that brings me to a question specifically for Vi uh what is exactly the chain abstraction solution octo is building we are talking about the complete stack can you put some light on it sure so see octo uh like I said we are building an end to endend chain obstruction solution uh the start of which is basically an interoperable wallet that we are building an interoperable wallet that gives you delegated access to execute transactions across any of the chains and any of the networks that we support at this moment we support more than 30 uh networks and we are expanding as we go along now if I were to go back to the four layers that we were talking about right so how are you going to do coordination how are you going to do the execution how are you going to do and in in in uh octo we have actually built out a solution for each one of these three uh each one of these four layers and made it so simple that a developer is able to just Express their intent and is able to execute right for example uh we have built out a solution called as unified liquidity layer what this unified liquidity layer does is that it actually Aggregates liquidity transparently without the user having to worry about where their assets are lying we find the best ways to do it we find the most cost efficient ways to do it the fastest ways to do it and then aggregate that liquidity for the user similarly uh we are also building out a solution called as dwnn dwnn is what we call as decentralized wallet Network decentralized wallet network is I believe one of the world's first fully decentralized wallet Network that is actually driven on top of APs operators where the user is actually able to uh Express an intent get the signing done and the signing transaction can then be executed across multiple chains then we back it off with uh what we call as internally decentralized transaction Network which actually finds these paths and executes the transactions across multiple networks so in summary if you were to go back to the original framework that we were talking about coordination execution uh uh data and liquidity and then settlement we are actually operating our solution across all of those things and that's why we say it's total abstraction uh or we say it's an in to end abstraction solution where a user is able to express their intent with just one single expression of intent and everything else on the blockchain is taken care of by a platform that's amazing explan can I chime in there yeah yeah so um yeah it's worth distinguishing that yeah what you guys are doing with Oto you have this it's a lot bigger ecosystem that you're trying to abstract you're trying to abstract kind of the whole everything in between um the way agoric is approaching it is is different again this is speaks to how there's a lot of people trying to solve this problem in the space um no there's it's not a monopoly it's a market people are doing it different ways they're going to do it in ways that compliment each other you know there may be ways that agoric and OCTA work together to to overlap and increase the surface area of this orchestration fabric um but the way agoric looks at it is um you know we're we're a smart Contracting platform um in Harden script but we look at it at you know our our orchestration smart contracts um are these programmable smarts that can automate this cross-chain uh multi-chain function and so the nature of our smart contracts are you can do things that you can't do anywhere else you can um you can code timers you can have actions take place in in future when a certain date is met you can have um transactions happen over multiple blocks um you can have things pre preset to to trigger under certain events and so um you program the smart contract it has all these features that allows the user the Builder to to to code in this uh this amazing experience from start to finish um and so it it puts more of the power in in the Builder the Builder hands because the builders are ultimately the decision makers they need to choose orchestration tools to provide that exceptional ux um and so when they do that um they have to they what they pick is important because they might not it might not be a one- shot solution they have to do slightly a few different ways um yeah I just want to kind of add the context of how a goric is a little bit different do right so uh and and I I think uh I agree in that agoric is doing I think beautiful work uh especially on the execution layer of the overall framework that we were talking about in the way that we are actually trying to build uh the dtn network is pretty much uh in the conceptual thought of the way we were actually talking about in that we eventually end uh plan to have a whole uh you know network of solvers and then uh for the for the users to be able to express their preferences of who the actually the solver would be and how the solving actually would be done so while in the first layer that we have actually built out we will probably be one of the only solvers in the network but eventually the solution is actually built out and that's the reason why we actually call it decentralized in that we going to have uh you know whole uh network of solvers and then be able to find the most optimized way to solve it across multiple solvers that's amazing explanation uh but but uh whatever the point you said uh Brandon like that brings me to a conclusion that more than smart contracts we need smart uh coordination inside the blockchains how is agoric orchestrating it and how is the solution working do you want to put more light on that yeah so uh I see I don't see how smart coordination is possible without Smart contracts right the the smart contracts are the composable uh you know reusable programmable like insert your thing here that uh you can program these smarts into right and so um the coordination aspect is is like without coordination of anything in any Market you you don't have anything like think about Amazon take away Logistics take away payment gateways um what are you left with right so it's how you string it all together and you know the hardest part with coordination is it's not uh sending a transaction from one chain to the other it's sequencing all the things that need to happen at the right time um and so with you know orchestrate um smart contracts on theoric you can you can do that in the way you need to um the Builder picks his own Journey right you can you can you can uh build in the the parameters that address that user's intent ideally the you know if we're talking about like top down the user is never going to care about what needs to happen for their intent to be delivered no user wants to know how the sausage is made they just want to eat their sausage and go home they they don't care like about validators they don't care about bridges they don't care about any of that and and so at a very simple level you know that's what's needs to be abstracted away and it's on it's on the Builder picking his tools to decide that and pick that Journey that's amazing so uh this brings me to a question like uh we have spoken about different methodology that we are solving the user experience for the end user so what is the endgame for user experience on blockchain like uh VI you want to go first here sure so I think uh the end game is a fully automated blockchain where the user is able to express high level preferences and everything else is transparently executed for the user right uh so you know user experience is something that uh we actually at Otto uh kind of obsess about I mean that's kind of the mission statement with which we actually started off the whole project with the whole premise that the experience for both the end users as well as for the developers is actually not sorted in the web3 ecosystem and that is probably the biggest blocker to mass adoption of the entire product the ecosystem while we love the decentralized principles and the whole concept about how people have power in in their hand the trustless ecosystem we not never be able to uh bring the actual value the true power of it in the hands of the masses unless the experience is sorted for both end users as well as for the developers so what's the end State I think end state is fully automated for example if you look at Sexes right sex has actually solved this in certain ways right for example if you have to interact with a sex today it's fairly simple for example if I want to swap or if I want to buy if I want to sell it's all single click of button everything is sorted but what is not solved there what's not solved there is decentralization right so it's it has centralization risk it has centralization problems and we have seen uh often we have seen fairly commonly that the centralization disk actually plays itself out in the ecosystem and so uh an end state would be what you would see in the web two world you know for example how I hit the like button in an Instagram and I don't care about whether a socket protocol is sockets are being used whether it's an HTTP connection whether Json are used used I don't really care about any of that as an end user right in similar fashion uh in the end state that we have people will not care about what my liquidity solution is where my liquidity is lying what my networks are what my Protocols are all I'm going to do is I'm going to express high level intent give my preferences and everything else will be automatically sorted while maintaining the principles of decentralization security and trustless execution that defines the very principle of the web3 ecosystem yeah I mean to your point like if we weren't building semi or fully decentralized Solutions everything would be easy right like we don't build these things because they're easy we build them because they're they're needed because those are very hard to solve problems how do you build a fully decentralized trustless ecosystem and yet give that the experience that you want to build yeah no my point my point is like we're in the blockchain space for a reason we're not Building web2 Solutions all the time we're we're we're you know we're it's not always as easy as decentralize 100% of the time so I agree with like kind of what exactly and I think we at one of those curve points uh pretty similar to how we didn't know how to send uh an email over a network right I mean that was a hard to solve problem but today that's not a very difficult problem to solve in the web 2 world in the web 3 ecosystem I think we are at that curve and I do think that in next 3 to 5 years we would have actually solved a lot of problems which will make some of the further building for developers much easier right example when we say that we want to solve for user experience while also solving for developer experience the real problems will be solved people will be able to focus on building their business Logics actually being able to solve and crack for user problems when they don't have to uh worry about solving the problems of the infrastructure where they don't have to worry about solving the problems of the platform and I think I personally believe that we are at that curve and as soon as we have crossed that curve is when I think the mass adoption will come in through Mass solving or use cases that go beyond example Financial intents to uh generic intents yeah my one word answered the question though is it should be invisible the end state of ux in multi-chain world is it's invisible the user has an intent the user presses a button and signs once and the user gets their outcome and whether that's a developer who built out experience or an AI agent who has Vibe coded the app for them because it didn't exist and provided that outcome whatever gets there first right so the end game is chainless so the end game of user experiences chainless like you won pure agnosticism there you don't know where it's you're you go to an app you don't know if it's in Cosmos if it's in etherium if it's bridging if it's um if it's got to do some complicated validation behind you don't know the network security exactly that's amazing so uh that brings me to my last question like uh that's a pretty interesting one that I want to know like uh how do you think like narratives like AI fit into web 3 I have I have a few ideas um I um so like I talked about it like agentic Frameworks that utilize AI I mean I I think it's we're not far away from and this is what's interesting about orchestration and Building orchestration Solutions There's an opportunity for orchestration Builders to create like digestible composable reusable components that an an AI agent could uh integrate and deliver that users's intent so you know if we are creating these things in the right way if they're maybe maybe they're lightweight um what's stopping an AI agent delivering that user's intent by using that orchestration and perhaps building the app that's that that sits on top of it um and what's also interesting is with wallets like maybe wallets just become an interface for um for an AI agent and you you you direct it you give it its power you trust it you say I want to complete this intent and users stop interacting with wallets all together because they trust their AI agent and it's just an interface for for authentication and signing so sure pretty much I think uh it'll be about uh improved efficiency and usability so what an AI agent could actually do is uh learn on top of all the user behaviors so for example you are actually trying to solve for an intent in your various path an AI agent could actually determine which is the most efficient path and keep optimizing it dynamically as an example right similarly uh on a usability front for example if you had a wallet and if you could give a very high level intent and then have the AI agent actually uh uh understand what the intent of the intent is and then do the execution for the intent at the solers then you could actually have an AI agent solve for usability too so I think from a usability and efficiency standpoint AI agents could actually play a very big role once we have actually solved for the abstraction and coordination because in absence of abstraction and coordination even in AI agent would find it very difficult to coordinate across multiple chains that's amazing thanks thanks a lot guys for the session uh we learned a lot about user experience today uh so happy to host you guys today yeah pleasure to be here thank you thank you