ETHKL24 | Soufiane Hajazi, Bloxroute | Intent Networks and The MEV supply chain

he is from block rout and he going to break down intent networks and M Supply Chain be small by the way I really picture because it really summarize what's eum and Def in general here you have okay Co taking off okay yeah FTX is on fire you have here tall building ethereum all the D and you have different l2s that trying to connect to this uh Utopian city called etherum you have Luna as well here at the details well great shout out to anyone who made this and of course you see that there is like no Bridge no connection somehow between ethereum and and Bitcoin and I think that today talk about scalability and makes sense so yeah today talking about intent and how does that impact the Dynamics of the IM supply chain so just a in about me so software engineer at blar and what we do is we are running uh in E INF different chains salana I and BST working closely with block Builders and Traders help them get fast and assistent trade on these Network and yes we are hiring if you're interested talk to me so yeah my uh so I hope that you as excited as these guys to Embark to this journey with me so we can go through the why the what and the how M and how do that impact the M exraction know that I'll be saying M and M so than uh so yeah my very first idea I would like to car that intent are everywhere people are constantly moving from state zero to state if I'm in Babylon for example I wanted to create a chicken for a fish I want to go from a state one which is owning a chicken to state zero of not owning the chicken but owning the fish so the way I do that is I would first go knock on the doors of all my neighbors and see if anyone is interested in my chicken if not I would go to the market and just start shouting who wants a chicken and exchange of so intents are everywhere but executing them was really really bad okay so what happens over time is that we were getting constantly better and better in providing platforms and ways for people to execute their intent safely and efficiently whether it's a barter to a simple intent of going from point A to point B what you would do just take your grab or Uber and just put your uh the Parker arrival and that's it and the whole platform will take care of all the technicalities of fighting the closest driver the fastest way etc etc so my very Point here is that with two companies understood that early they understood that people users generate intent okay and if you are able to execute those intents in a very efficient and fast way then you are creating the value because right now as the other talk said we three is losing its TR and even viic says that there is like a creativity free I think that the very best thing that would unlock this tree would be building an intense Centric world so that people can have a smooth user experience on on V5 and that's kind of the big thing and if you are building a d building an application on any platform and you are not considering intense then you are in a and of course if you are so unlikey you would go from state zero of having zero Pizza to BCC to State one of having one pizza and zero DCC so what are Inns inters basically the idea of describing the output you want without caring about the process I want to have an nfx given one E I don't care about how do you execute that how best you a it all I care about is getting my n so the very core idea behind intent is to distinguish between two things users have explicit preferences and implicit preferences if I go to uniswap interface and I put one e this is my explicit preference I want to swap one e for whatever USB would that be but my implicit preference is is that I'm allowing un to set the flipid and also I'm accepting as little as x amount because of the speeds Etc that go LP and the network and of course these are customizable but for somehow an average Jo wouldn't optimize they wouldn't even know what is and we have also non optimizable preferences which is in this case you can't tell Unis swap to use another liquidity pool besides its own okay so today my point is all about working on the implicit and making them easily customizable for people without letting do any grud work so that's exactly what happened when you tell people about implicit preferences and how they can get them right uh so in sense have so many use cases I'm not going to go through them but you can basically say I want to exchange usdc for USBC I don't have e please pay for the for the gas using my usdc you can also if I'm a bank for example I want to execute a swap I can say please execute this swap with only regulated B for example licensed B in these states so I'm clearly explaining my intent saying that I only want to trade with company who are registered or block Builder who are comp or whatever and of course I can also ask my transaction to be bed or my intent to be bed for gas so why intent why do we care about intent why do we want to transition from a transaction Centric defi to an intense Centric defy three points first point is that defi is becoming more and more fragmented liquidity is available cost we have seen man network and we have tens of l2s we have Sol we have Cosmos it's a lot of network and liquidity by Nature exists everywhere and for me when I want to trade some liqued trade I'm not aware of that liquidity I'm not aware of whether trading on ethereum would be better for me or trading on polygon would be an even better option because I'm on the G and maybe there is like a an Arbitrage opportunity there so liquidity is fragmented execution as well like right now we have dkm we have this we have aristic rollup we have offchain evm and the whole interium road map is all about building this L2 e where you have a of uh parall the ends being running auction also the logic do you take for example St it has its own native language called Cairo for you to write my contract and for me going from E to to I not only need to learn Sol but I also need to learn Kyo so and the logic there could be different there are certain constraints with the compilers so The Logical things logical logical can also changes depending on which network so this is like a transaction Centric way uh world right now this is what's going on we have Eve she wants to buy an NFC but she only has D okay I only have 100 I want to buy this nft but this nft contract only accepts you what I'm going to do is unfortunately I couldn't find an d e liquidity pool I would the first get some usdc and then from usdc get some and only then get my nft here I'm I'm expanding the surface of attack because I'm exposing myself to so much here like Circ can on Me on the first R can sandwich me on the second etc etc if I screw up on setting this Li so the idea is if uh with the current transaction cic uh users are always exposed to am and because they certainly don't know what's the best route for them to execute that trade so an intensement word would be like this uh Eve would go can anyone fulfill this for me you can keep the change I'm only willing to give up 100 D for this nfx that's it I don't care about what you're going to do to execute it I only care about you die and giving me the NXT and then we have uh some actors we call solvers steers who are these guys I'm going to get to them in a second who are like the big ogs who understand the liquidity where it is and the best rout for you they say okay I'm going to do it so because they know exactly all the liquidity pools and what's the best L they would split the trade they would take half of that die and swap it to HDC of it USB and then the rest to East and from East they will go to NF and at the end they get to they got to save one Di and they get to to to keep it so this is kind of how efficient trades are really really important for and that's how and that's why also uh Outsourcing your intent is really uh important because of how much you can be uh how much you stve and so so yeah we have current router you go from say I have this amount of compound open I can either use this regular router or an Al router but since with un liquidity will be behind so it will be so much difficult to maintain those Auto routers and that's why we definitely need need in so who are the solvers so solvers are basically entities running some highly highly complicated algorithms that are constantly pulling liquidity from different sources think of them as Searchers but also not only Searchers but solers because they are able to get you the best route for your transaction to get exec we think that they are honest solvers they are honest actors since they wouldn't do anything bad with your intent they listen to liquidity as I said and they it's also going to be a permiss right now not really permiss list you need to get per to the but the idea is for anyone who is capable of uh providing best solution to be able to participate into this open so over would get intense because of the maxing ring and the heavy lifting computation they are willing to do they would either go to an aggregator on layer one layer two or simply go to use a tier toer network if there is like The Coincidence of once and this is the whole idea behind the call protocol if I'm willing to buy usdt and someone is willing to to sell usdc buy you get the idea we can just save up on the lp fees and network fees and just exchange the exch or I can just simply go to a centralized exchange but no lot centralized exchanges because you need to get Co so here is the what the landscape looks like users are creating intent solvers they could be a dow they could be dependent solver could be a community solver go from the discovery the creation of transaction and then L into the settlement layer which would be either any any chain it could be I I have one B&B I want get some E I don't care so the solver do the brid ETC for you because you don't know exactly uh how which is the best bridge for you and how safe is that uh so intense are something working it's not just a CH we have protocols doing intense we have C protocol go check it out it's mentioned it's uh it's an intense protocol where we have solvers tens of solvers and anyone by the way can soling there and get the native token of the protocol and this is like the transaction world when you have meare transac Bo lotive and blocker so little bit running out time so at block we have an intent Network so if you are a solver of dab and you are generating or you want a intense you can just connect to the network and get the be the the first one to get those intents to solve them or if you are adap you can just make sure you are connected to the Des overs and making sure that intents of your users are getting as fast as possible to uh to the so yeah definitely the future would be to have a a domain specific language for intent uh let's say we want to have something similar to gping hey please I take whatever BMB from my wallet and I want to have same e that's it I don't want to interact with any interface I don't want to connect my wallet I just want to have human language uh kind of interaction so of course there is risk of centralization over getting bigger having more and more uh complex algorithms the requirements for the hardware are going to be really really interesting so there is a high risk of solving being only at the hands of very few Elite uh actors and then there's also the solving in incentive what if I submit an intent but no solver is excited or in solving my intent so the idea is that what C protocol did is to incentivize solvers by their own native token so you actually not only get to keep whatever M that is created there but you also get some of their token just to incentivize people to uh Sol in so this is the current M Landscape on ethereum you have transaction uh you go to you have a transaction you have two options whether you go to the public men pool this is where you're going to be eaten alive by B who are constantly listening to uh the men pool transaction they basically GNA hand with you and uh yeah that's bad for you and then you have the second option which is private main pool and how you do that is by basically submitting your transaction to Builder like block rout or you can submit it to an orderflow option like meare or me blocker the idea behind this uh uh orderflow options is to create like a conditional order flow where a user would say okay I'm gonna submit you my transaction but I would want to only share with you the call data or only the pr and two addresses and whatever MV you get to create I want to get a share of it and that's exactly what's happening with Medcare so Medcare we take a transaction we share it broadcast it to sers seers going to back you and they're going to send you like 90% out of that ad that they created for the transactions and and yeah that's if you want to participate in this option you can do it or just submit that as a private transaction and then we have Builders running low infra low latency inra and high complex algorithms to build most profitable block and then from the builders we send the block to uh to the block this is after to the PDS an intense Centric M Landscape would be intense in added to this intent pool uh solver solving the intent so the solver would take care of the execution so they will construct the transaction you don't have to uh worry about any front TR in or swi these guys know exactly what's the best way to submit the transaction uh then action from solver solver would go to the Builder and from Builder go soes there is the say that intense can cure cancer but they would never cure uh any so M within an intense TC would be like a service C for me it going to be more it's G to be fair because you have solvers competing between each other let's say I want to go from point A to point B and I'm sharing my intent with both Uber and grab so would expect those two to compete so they can finally the best and cheaper uh ride instead of just eating me alive or getting me wrecked so this is kind of how gets mitigated within an intent landscape as long as we have healthy competition as long as we have uh an open protocol for anyone then we get to be more fair and uh yeah so time is up and thank you so much if you have any more questions all right thank you very much for that