Intents Trading: Why AMMs Are Going Out of Style | Aori

Transcript

if something just has the association of there's there's this like you know kind of concept of like throwing the baby out with the bath water right like like kind of just because you know just because the bathwater is bad doesn't mean you should throw the baby out the window too right I think just because trafi is bad because it's centralized does not mean that we can't learn from it and and take all aart the best parts of it and I think what trafi got right that we are way too arrogant as an industry to accept is they got trading right the polygon Community grants program was launched before 1 billion tokens all four polygon Builders season 1 is live right now and features 35 million in madic to support the next generation of polygon products builders on polygon these grants are for you join the

aggregated future today by applying at polygon.net on zerox Research is a recommendation to buy or sell Securities or tokens this podcast is for informational purposes only and any views expressed by anyone on the show are solely our opinions not Financial advice Pacho Ryan and our guests May hold positions in the company's funds or projects discussed what's up guys welcome back to another episode of ZX research today I'm joined by Ry and Joshua from aori thanks for coming on guys yeah thanks for having us thanks for having us mate um before we get started do you guys want to tell a little bit about what you guys are building yeah yeah uh happy to start um

yeah so Joshua from a um yeah I guess at real high level what we're building at aori is a uh you know we like to use the term like high frequency intense protocol to sound really cool and make the VCS really happy but in reality that just means like you know you want to trade in defi you want the best possible price on whatever chain you're trying to execute on um and you need it to execute this block or you know whatever the block is that you you demand it to be executed by um and we're the best at that so yeah really really simple definition of Yi perfect um and why did you decide to build aori what was the like some of the problems that you guys wanted to specifically solve so you

might notice I'm I'm losing my hair I've lost a lot of it um I don't know how many years left I have of it um but that's mostly the fault of uh defi protocols um you know uh it's nothing it's none of that Health stuff or testosterone or whatever it's a having to be a market maker on a bunch of dexes uh made me want rip it out um so yeah that's the main reason uh but no just to be more specific um I was a market maker previously um made markets on a across mostly options but towards the latter half uh of my career in Market making um was tra to be the dedicated Market maker for a lot of dexes for a lot of amm for a lot of onchain clubs um even some offchain auction platforms um and it they these things were just

like completely intellectual like thought experiments that went way too far um and ended up causing lots of lots of negative externalities and harm to users that I think a lot of Def protocols and defi Founders want to ignore um and that's really what uh I think I think the reason a exists is to minimize uh the amount of harm to the end user or actually make it negative in a lot of cases um that's really the end goal is make dii as great of an experience as possible and one where you're with every action you do you're not leaking out tons of Alpha tons of value to be captured by people who are looking to exploit you um so that was kind of

really what what led to the creation of a that makes sense um and I want to get into all of that specifically around amm design later in but I guess like to start us off a bit how does aori work on the back end what I I remember you guys using tort contracts is that still the case um if so why did you decide to use Cort and was it difficult integrating it yeah it's I guess maybe I'll give like a little bit of like a um a little bit of like a a detailed road map into or I guess like not a road map really more like a like why we chose to use something like for example a lot of users Pro a lot of people unless they're really technical probably have no idea what cport even is right um so so everybody's really familiar with amm

these are you know your standard onchain market makers that you know you add liquidity to on both sides and you know the definition of market making is I'm going to be a buyer and a seller at all times um I'm going to I'm gonna have bids and I'm gonna have asks at all times and that's exactly what an automated Market maker does it puts liquidity on both sides of the book and lets you uh uh you know even like fancy stuff with like V3 you can you know add custom ranges or whatever but like it's basically just like I'm announcing my you know bids and asks ahead of time on chain uh and this leaks the most uh you know uh uh exploitable uh trading uh information possible um then there are you can get a little bit better with things like CH clobs where like you're

able to update like on really fast blockchains like on salana with like you know serum like open order book Phenix um where you can update your quotes really fast so it's much closer to that of like a centralized exchange and as a market maker you're leaking less Alpha um and then uh I guess like one step in the right direction but in a completely different way is something like cow swap or Unis swap X where uh it's completely offchain it's all based on these things called signatures you probably if if you metamask or rabby or any of these wallets you've definitely used a signature before it's that gasless little button that just says sign um these are basically auctions that auction off your order there are problems with auctions I'll get into

those probably later if we were interested in talking about that um but what aori is is based on this smart contract that openc created called cport um cport is basically um a signature um platform for trading where they basically made this hyper generalized signature trading platform where like you know Bach I can I can sign a message that says I would like to sell one ethereum and I want at least 3,000 usdc and then you can sign an a mirrored order that says yep I'm willing to send you 3,000 usdc if you give me one ethereum and then we both take our signatures we plug them into the smart contract and then we when we settle that trade uh we both get our tokens and if we both have the correct number of tokens in each of our wallets the trade

clears if you've ever done an nft trade on openc or on blur um you have used cport and and that was what what what you used um and aori was like I guess like the idea behind aori is you know openc put in a lot of barriers to be able to trade ERC 20s for ERC 20s they specifically want we like okay only nfts for nfts or nfts for eth or whatever we were like okay well what if we made it what if we try built a deck on top of this um and that's really where he already came from is like like what if we built a decks on top of this thing uh and we did and uh worked really well um we don't use Cort anymore we've made some optimizations like openc built C port for nfts so it's got a lot of gas

optimizations for nfts that it doesn't have for your C20 so we just we just made a lot of those optimizations but it's the exact same architecture same infrastructure style signatures for trades um and then settlement you only pay gas at the time of settlement um so yeah that makes sense and um one more thing that I wanted to ask about that's slightly related to this is that when I I know we said you don't use cport anymore but let's say I say okay I want want like 3,000 usdc for one e that goes on as an intent um but one thing that you guys talk about is limit orders as intents um what does this mean how does it I guess what are the implications because we hear a lot of this intens just limit orders yeah I think uh crypto

and defi Founders in general especially are obsessed with creating narratives um and like don't wrong I think like a lot of the work like like the enoma guys have done has been spectacular on defining intents as things that can go Beyond limit orders but today uh intense are just swaps they're just limit orders um and so when we say intense as limit orders we say like let's boil down an intent to its most basic format and that is basically input token output token and whatever the chain that I want to sell that trade is on um even like one thing that uh um like the enoma team has like talked about is uh um that uh an intent should basically just be I have a portfolio at time one and I have what I

want my portfolio to be at time two um and a solver has to get me from State one to state two and that can just basically be a collection of swaps for example um so that's really like how we we view intents at their like most fundamental form we we actually before they were called intents we were calling them virtual orders because they're basically you know like we consider like real orders to be like I'm paying the gas I'm creating an onchain limit order so a virtu ual order is just a signature that's just it's all held offchain it's basically waiting for some you know counterparty to fill it uh effectively uh so with these like virtual limit orders um you know or intents um is like can I define that

anything else other than the the ab swap tokens um are there is are there more Express potential expressive I guess like options that I can have as a swaper such as like I want X for y over time Z yes that's a that's a great Point yeah the really like we've we've kind of written some about this some we've published some we haven't um a lot of this so when you when you look at intense as like this if you boil them down to just pure limit orders right like if you look at defi in general uh what is defi but a bunch of protocols that are effectively expressing combinations of swaps so like let's look for example at like a money market

protocol like compound or a right like what is what is opening Alan opening Al loone is I'm sending the lending protocol my collateral token and I'm getting back whatever asset I'm borrowing and then after some time after n blocks in the future I can then send back the loan token plus interest and then get back my collateral token that's just two swaps like any that's that's not that complicated it's just two swaps and so like what we've explored and we've built some like proof of concepts of but like if anybody wants to reach out to me in the future is interested in kind of stuff like this we're we're big Believers that the future of all dii protocols will be built on top of intent limit order Primitives like aore so I

can build a money market protocol that's built entirely in like a te for example or something similar um on top of aori that ex that like I just said I'm just doing one swap to get into a into a loan position and then I'm calculating the interest in this environment and then after some time I'm uh asking the user to pay back the loan and if they don't pay it back I'm going to liquidate them and it's all done entirely through basically these solvers who are finding you optimal swaps um so yeah I I'm a big believer that like all def5 protocols in the future will move to this like perss are perss are just a bunch of swaps money markets are just a bunch of swaps um you know this is kind of the the direction I think it'll all go we're all we're all swapping

information in the end of the day I did have a question going back to uh you know what you were talking about in terms of some of the motivation Beyond a which involves you know almost like providing users protection against you know various exploitation so like going back to one of the points that you made earlier about how the motivation of AI was to uh you know like solve for how to protect users that are getting exploited could you speak a little bit more to like what users you're talking about and the types of exploitation that exists honest I mean like it's funny you say what types of users like like honestly there's only like a handful of

users who use any existing onchain clob onchain amm auction platform that aren't being exploited the better question is like who's not being like who's who isn't being exploited because like when you look at like amm for example like the everybody knows about impermanent loss in lvr um but they don't realize the extent to which these trades are being exploited every single Swap you do on the Unis swap front end or through any kind of router um you're almost certainly whether you know it or not having your transaction be front run and back run by uh either a a Searcher who has some kind of an agreement with a vertically integrated Builder Searcher so a block Builder that like is ordering

all the transactions so it's not even like you know Jared from Subway anymore is like the exclusive one doing this there's a bunch of them that are just like basically front running and back running you versus CI liquidity versus liquidity like your trade is being exploited whether you know it or not um and and additionally uh there's been some like really good like threads and like papers about this about like the maximum exploitable value like like there's some crazy stuff now where like you it's so complicated now that whenever you do a swap on like Unis swap uh for example um the best Arbor that's arbing your your trade is likely doing something like a front run a jit liquidity then there's your trade then there back running you then they're burning their liquidity uh

and out after after they they jit made and then take burn their their just time liquidity at the end of the trade so they're doing like these like multi-legged transactions to like screw you um onchain clobs and like auction platforms are a little bit more complicated in terms of like how the user is being like taken advantage of on chain clobs the you're minimizing a lot of the uh problems of uh announcing your strategy ahead of time like a big like impermanent loss the problem with impermanent loss is you're telling me every single order you're going to place and then I can basically say I can maximally Arbitrage you against a more liquid venue like binance for example um on a nonchain clob you can update your orders really fast and you can be really

granular with them um the the issue though is you can only update your orders as fast as the block time and now there are a lot of people who are like like and this is not like talking [ __ ] about any of these like really fast new blockchains that are coming out like Monet or you know Mega E or anything like that they're spectacular Tech like I think they're going to be incredible platforms um but specifically with relation to believing that like an onchain clob could work on these platforms like I did a panel at ECC and I thought this was really funny one of the uh one of the uh uh panelists uh without naming who they were uh they were they ran one of the biggest onchain clobs and they were telling me like yeah whenever we started our protocol um you

know we didn't have any issues whatsoever with gas um but now um You you know and they're on a very fast L1 that has very low gas and they're like within the last year gas has 10 xed and our spreads have increased by several basis points just because of how congested the network has gotten and it's like that's the inevitability of any onchain club like you're you will always be worse than binance because I'm having I can't update my orders as fast and I have to pay gas to to to place these orders so what what that means is the end user whoever whoever's tapping into this liquidity is getting worse prices um than they than they should be um and auctions like just to Breeze through auctions real quick the issue is just with auctions if if binance is if if the

the asset we're auctioning off like on cow swap is like eth and binance is quoting me one price if I'm a bidder on this auction like if I'm a solver I'm never in my in my wildest imagination going to bid the fair value of ethereum I'm always going to bid under the fair value of ethereum because I have to make profit if I don't make profit then what am I doing here like why am I solving on this auction platform right so that's kind of a lot of the when we kind of architected aori the big premise was like how the hell do we minimize this as much as humanly possible hey guys permissionless the event for crypto natives is heading west to Salt Lake City and we're bringing the biggest names in crypto together for an incredible campas event join me in the rest of block works on October 9th to

hear for more than 200 Industry Titans including Nick white from Celestia shiram from Igan Lair and Keon from Mona the conversations at permission list will be covering the hottest fees in crypto including modularity reaking the Bitcoin and Solon ecosystems Ai and crypto rollups l2s institutional adoption and of course the November election that will be right around the cor we couldn't be more excited about this event and it's happening in one of the most beautiful places in the US ticket prices will be increasing towards the end of June so get your ticket today make sure to use the code so I get some credit and bragging rights from the rest of the team uh use ZX10 for 10% off on your ticket one more thing we're also hosting the permissionless re hacka font

in collaboration with blebox the theme this year is a multi-chain future to build a better cross chain experience so I don't have to have a heart attack every time I bridge to a new chain all hacka found participants will receive a free ticket to attend permissions so if you're a developer come join us in Salt Lake City team up with fellow devs and build the multi-chain future polygon Labs is developing the next generation of open- source ZK Tech to aggregate crypto liquidity and user bases empowering developers to grow in a unified web of interoperable chains with the a layer the a layer is a machine for Network effect don't worry about boost wrapping liquidity just build anywhere in the a layer to tap the liquidity and users everywhere to support the aggregated future and boost scrap quality project the polygon Community

grants program was launched with 1 billion tokens All For polygon Builders season one of the community grants program is live right now and features 35 million in madic to support the next generation of polygon projects for Builders on polygon or projects interested in Mig grading these grants are for you join the aggregated future and feel empowered to grow in an infinite web of unified interoperable chains apply today at polygon.net how are you guys going to onboard solvers or I guess what how I wanted to ask how do you approach that and um as far as I understand most solvers on most intent networks are run in-house currently um is it's going to

be the case for you guys for some time how how do you see this even working out eventually I guess I want like also the big picture Josh take a bit yeah yeah so we have like just to touch on aori and then the ecosystem in general so we have no internal solver right now like we candidly like we Outsource 100% of our solving and I think it's probably going to be that way for a long time um there's nothing precluding us from running a solver but like I don't know I don't want to get into the algorithmic optimization game that [ __ ] is complicated and really really hard like like it's really hard um I I I'll talk I'll talk a little bit about like solvers today as well like so and the solver ecosystem in general so onboarding solvers has been like really

really great like for a couple reasons number one like all the solver teams are really great they're really open they're super happy to integrate because like fundamentally at the end of the day the way that they make money is getting orderflow and we're another source of order flow so they want to integrate us and so it's been really really like easy to onboard solvers and work with them because like we do a lot of the tech integration for them we really are happy to talk to them and communicate about like do we think that they're going to have a chot to win orders on a so we're we're really above the board with them in terms of keeping everything like you know we don't want like 20 solvers who uh and there are some solvers that do this like which is kind of insane um

anybody in the audience who's not like technically uh you know proficient um you know like the the the lowest level like the fastest possible coding you can do is in like assembly or like something really low level like Ruster C like there are teams that have basically Rewritten the entirety of like Unis swap in uh a rough C environment so that they can be faster at calculating Unis swap routes than anyone else like in assembly and [ __ ] like it's it's insane how hard they're optimizing like there are teams of like five to 10 people whose whole day job is like let's shave 10 micros seconds off this week of our of our quote and that's not a millisecond that's not like one 1000th of a second that's one 1000th of one 1000th of a

second like this is insanely like Fast quoting we're talking about um so there are a lot of solvers they're really really great teams to work with like um right now almost all of them all of them basically are optimizing for generic or sorry generic swap solving only so like input token output token which I think that will likely be what solvers do forever I really would be shocked if I saw any kind of intent uh infrastructure take off that's not just pure swaps over the next like five years at least unless something catastrophic changes in the tech like there's no way um the the even C like I'm not even talking about like you know just pure same chain swaps even cross chain swaps will be just swaps or cross chain intents will be just swaps

yeah depend it depends what you're swapping though right like right now crypto is at a phase where uh swapping uh swapping anything is with respect to token some sort of financial value but that token there's a data structure and there's some information that is being swapped ultimately and so I don't know I could imagine a world where like swapping information that isn't necessarily Financial tokens exists like what like what information I'm curious uh I mean yeah this I don't know this is like a totally different uh podcast and no it's it's interesting a different area of exploration but like you know right now on the internet when I'm you know signed

up for Instagram I'm swapping all sorts of my information about my you know who I am and my age and who my phone is connected to and you know they're able to sort of paint this picture on like who I am and what I might like none of that information is financial I think that'll probably like anything that's I yeah I guess exclusively sure what I'm talking about it's like Financial right like for whether it's spot tokens or it's like things like tokenized derivatives for example like tokenized options or perss or um even even like binary options like bets basically um what you're describing like data swapping I think that they I think that kind of solution already exists I think it looks like like hyperlane or like

layer zero right like it looks like a generic messaging protocol um if it's if it's like I don't I'm not trying to get something out I'm just trying to send some data and maybe I expect some back I think it'll look like layer zero or hyperlane or or like a GMP or something like that but like any kind of financial swapping or any kind of like uh dii protocol will will likely be solver based um I I I don't know if I'd call it like the I call it like the market maker iation of everything uh you could maybe call it the solation of everything I don't know I I I need to work on the marketing of this of like what this term is going to be called but like yeah I really hope that that's the future if that's not the future then we've done

something catastrophically wrong um yeah cross chain bridging now has just boiled down to like in intense bridging as well it's just swaps I'm swapping X on Solana for X on E so it's the same thing um and what does the a architecture look like I remember reading that it's based um wh why why did you guys decide to go with a base architecture what was the primary reason for that and did you consider any alternatives there's there's two things to talk about there number one is like cross chain swapping and two is like what based why is a based or like what is based mean um the first is like uh for for the base part just because that's that's easier to tackle real quick like I really hate how

uh how bastardized the term based has gotten um I don't mean like based in terms of like posting like edgy memes on Twitter like for any of the listeners I'm I mean Bas is in like based sequencing is what we're referring to um uh although don't me wrong I'm based on Twitter but not like not like in a sequencing way um uh in in a sequencing sense um I believe that a a a app or a rollup is based if what they do is you do some kind of transaction building offchain and then it it doesn't it can be on a blockchain it can be not on a blockchain it doesn't matter but then the transaction gets sequenced to the layer one to interact with or to kind of in in interact with the layer one so I think a really good example of a based

rollup is op is opy or blur like I think these are by definition based rollups You're Building transactions offchain that have multi-legged like things like signatures for example or some kind of complicated data structure and then whenever whenever some some you know uh condition gets met either the maker the taker the blur Foundation the openc foundation anybody can then take that call data they can take that that transaction and then write it to the layer one and actually execute it on chain so I think that's what a based app is I think any other definition of it is like probably way too granular I think it literally should just be you build your transaction offchain and for your app and then all the transactions on the

app get sequenced on the whatever your destination chain is I don't even say when I say L1 I don't even mean just mean like ethereum you could just pick arbitrum or uh or like salana or whatever it doesn't matter just like you're settling them in whatever order that the app users want them to be settled on the Chain um right um and the way a based is like that's what we do we allow you to build a signature offchain build a transaction offchain and then based on your needs like based on how fast you need that transaction settled um it then gets settled directly to the layer one um the second part that's like kind of tied in is like cross chain bridging and like what cross chain bridging and infrastructure it looks like in general I think like going back to like the Soliz of everything like I

think um I I really would be shocked if there was a cross-chain bridging protocol that exists in like a year or two I mean maybe they'll exist but like they won't be the majority of cross chain volume I think what'll probably happen is apps will do dedicated deals with private market makers or private solvers that have liquidity and they will directly tap into them to provide them quotes for whatever token they need and whatever format they need on destination chains and I think that's actually probably better because I think what that means is you'll have this very open competitive market to capture each app's order Flow by the solers and market makers for cross chain order flow and that means that the end users are going

to get better pricing it means you don't have to do these things like these convoluted like auction mechanisms for some of them like I I really think like that's probably what it's going to look like but what but what happened to defi what happened to open transparent Finance like what you're describing kind of sounds like trafi all over again it just sounds like you're it's Market making in a traditional sense it it is but it isn't because what we're doing is we're if you if you just if something just has the association of there's there's this like you know kind of concept of like throwing the baby out with the bathwater right like like kind of just because you know just because the bath water is bad doesn't mean you

should throw the baby out the window too right I think just because trafi is bad because it's centralized does not mean that we can't learn from it and and take allart the best parts of it and I think what trafi got right that we are way too arrogant as an industry to accept is they got trading right they got trading right there was 500 years from the creation of the Dutch East India company's IPO even the Romans had like stock markets right like there's a lot of cool papers on this and like stuff they had like a very sophisticated stock market that was almost as sophisticated as like you know pit trading in like the 1990s like it's really there's really cool stuff on this like trading is a solved game like tra continuous limit order books effectively or continuous continuous limit limit order markets are

like you know the most efficient for any kind of liquid asset and auctions are the most efficient for Price discovery of illiquid assets like this is kind of like known economists have known this for a long time there's great papers on this um so we should not be trying to reinvent that what we should be trying to reinvent is the settlement and storage of all data relating to ownership of assets that is where like we should be trying to decentralize What's called the Clearing Corporation at the CME not the CME like people don't know this but like the there's the life cycle of a trade in trafi goes exchange and then at the end of every day they take all the trades that happened on The Exchange and they send them to the clearing Corp and then the clearing Corp

then Nets everyone's trades out and they say okay here's all the balance changes that need to happen and then they send those to like your what are called custodians so that's like Vanguard Black Rock State Street uh bny melon um these are like basically banks that are just holding on to stock certificates for you at the end of the day and so what we should be trying to decentralize is the second and third tier of of execution not the first tier um if we can use things like signatures for example we're even able to capture the majority of the decentralization in that like unless you actually your private key and public key signed for that trade like that trade can't even hit the clearing layer like so I would say like right now aori for example and like this is true of C I'm

not going to just say this about aori I'll be fair to competitors like this is true of cow Swap this is true of unit swap X this is true of like any signature-based platform is we basically have the same trust assumptions that like a centralized sequencer has we can censor your orders but we can't um but we can't edit them we can't check we can't settle them without your permission because like your private key is the one that signed for this and so that means unless your private Key signs for an edit or a cancel or anything like that like that trade can't clear so the the what we should be working on and innovating on now is probably things that look like censorship resistant Solutions things like tees for example I think are like a good good path down here but there are problems with tees

but there are other ideas with censorship resist and stuff with like avss and whatnot too but sorry I'm rambling but yeah no I like I really like that analogy I really like that analogy you present with the exchanges and then there's this Clearing House and this whole supply chain and I yeah I agree like there's a way that we could adaptt these traditional models in a way that presents all of these things that we don't normally have like access to or uh we don't normally have the ability to observe uh these relationships so that we can make informed decisions as Market participants uh you know making that more available for everyone for sure is I think a really good North Star to keep keep uh you know walking toward Roy I think this one's a bit more

for you um I'm realizing maybe I should have let Josh take this one on his own no it's good um special I guess I want to understand who the main well this one's actually too pronged but the first one is who are the main users of aori um and I don't mean like apart from obviously me you all of us but and like what's the current go to market obviously you want like regular retail users but is there somebody else that you're specifically focusing on yeah so to to answer your question on on who the users are you know on one end we've got takers which which I think Josh you know defined very well those are mostly

institutionalsales which which will look I think very very clean and and I'm spending a lot of time now trying to get people to use it and and test it out so that we can improve it before we officially announce that it's actually live um and on that side it's going to be very much retail or anyone who would ever interact with a DEX and you know or an aggregator Etc to to do any form of swap so to answer your question both retail and and institutional and I guess on on the go to market side of things it's it's really just very much user-driven and and I don't mean to be the meme of you know for users by users but it's

effectively we have a set of clients that are already on board and it's a question of hey where do you guys want to trade the most and that's how we'll form where we launch first from the institutional side of things right on the user side it's it's really just hey bu can you come try this what do you think how do we improve it because though we trade and you know I mean you know that I double in [ __ ] coining here and there so I I can you know extrapolate what I would want but that doesn't necessitate that that's what magic would want or what Josh would want or what what you might want so it's very much trying to get a lot of feedback on the Fly and and just reiterating and reiterating and reiterating till we're

comfortable and then even then trying to improve it further um that's on the go to market side of your questions um now now why I asked when this is launching is is we're also working on another retail product which which I won't Del too much into but that's going to be primarily retail focused um and so that's where trying to get retail users will be very very helpful and and sort of create a nice flywheel of of order flow for us that makes sense and specifically I guess on the taker side what does that look like how do you I guess have um yeah well so thankfully you know the team is is is very much from a market making background so they've

already got a lot of connects in that space which has made my life quite easy on that front but otherwise it's you know going after these new liquid books a lot of VCS who've turned into really long only hedge funds um who manage their own books you know going to them and and asking them what what would they need you know whether it's a teawap which which is very obvious or other sorts of features um but attracting them you know a lot of a lot of a lot of these books are very secretive they don't want to be known and so that's a little more tricky and and something I'm trying to maneuver very you know niftily on on how do I find these people so that I can on board them that makes sense joshh you have anything you want to add yeah I would say like like yeah a lot of

what we're really talking about it's like what we're doing now um but I think like it's important to know like our long-term vision is that like this is how all trading will will work in the next like several years in defi like I think we're we're kind of really focused on building infrastructure for trading infrastructure for any kind of settlement that needs to occur um and that's really kind of what we're what we're what our kind of vision is is that like swapping trading defi protocols in general in the future will look a lot different they'll look they'll be effectively these kind of based applications built on top of signature signature swaps and things like that um much more so than they exist now in these like really kind of crude Let's Jam everything into a smart contract possible and so I think like our takers

for example like we use the term taker now but like these will just be defi users in the future this is what defi like I think will look like is basically having you know uh these just pure intentions that you're you're working with with and all of the execution routing optimization all of that occurs in like a very open and competitive Marketplace um between between solvers I want to double tap on your definition of based because I think it's really interesting I think that your definition is is is much different than the definition that I think people use in you know on you know these where people have these conversations um can we double tap on that um and for I guess we'll maybe try

to like I'll just try to summarize what you what you've said and correct me if I'm wrong based on like your definition you think that uh a ba it's it's a based rollup a is a based rollup not because uh there's some L1 proposer who um does the sequencing it's that the aori settles on the L1 where the proposer re receives the sequence of a transactions so in a sense that like in a sense like you guys are sort of sequencing or there's some solver or Market maker who sequences this like opportunity and then that gets

posted onto L1 and the proposer isn't doing that sequencing it's just stamping like oh okay this is what we're receiving I I I think like you're kind of just to be more specific like um in my definition of base it's very very generic so I would almost say like a really good way to look at what I think of is based is is basically a private mol it's like you could really view aor as like a private M poool for example um or like even cow swap like basically you're doing all of your transaction building in this you know offchain isolated area where call data and transactions are being built and then when conditions are met they then can be sequenced to the layer one now an app can add any rules they want to to the

proposer of that transaction on the L1 so like you know the way cow swap does it is they say that like cow swap settles all the transactions they don't even let the solvers settle the transaction so the sver send their call data and then uh cow swap is the one who actually executes that transaction on the layer one directly um we do it in a much more um I guess you know open manner where we basically like once an order has been matched we publicly announce that order to everyone in the E everybody who's listening to our API anybody can settle the transaction um and so that means that depending on who whoever whichever counterparty of that trade the maker or the taker depending on how fast they they need that transaction settled they can propose that transaction to the layer one in

whatever kind of time format or block format like whenever they need that transaction settled so I I would view a based application as basically saying you do your transaction building offchain and then you allow the execution and the proposing of that transaction to the L1 validators or whatever the Val whatever the you know sequencing structure is of the chain that you're pushing um you basically allow it to then settle directly to that chain so I would look at it more like I would think of Base sequencing basically as a rewriting of the old term private M Pool basically okay interesting interesting I'll have to go back to Circle or something Josh I was going to ask you about te and why you brought them up I

just think they're really really interesting like I think like any any way that you can like basically have this uh so like if you view like a a block a blockchain is a you know a a distributed execution environment right so it's this distributed execution environment that every node is running and then you know you have your data you know stored on that node you know and we all agree you know based on these different you know um consensus algorithms to then agree on which transactions to include and which ones not to include or whatever right um a te is just a really like interesting concept of how do you basically build a verify ably uh sensor or a verifiably you know censorship resistance SL like a Computing environment that has like some of the capabilities of a

blockchain that only exists on one computer I think that's a pretty cool idea um so I don't think like I'm not like a te Maxi or anything but like I think tees are pretty cool and like they open up like for example like we can run the a reorder book for example on like a distributed set of tees and then let people connect to whichever one's closest to them for example like there's just it just it's a really cool way to do Computing um or to do like basically get some of the properties of a decentralized network on one box which is like something that has not been possible previously no I completely agree and I feel like we I'm seeing a lot more talk on te on the timeline over the past few weeks at least um I feel like the one question that I

haven't asked yet is why and I know you have very very strong opinions on this is why you guys are building I guess limit orders as intense as opposed to amm um yeah that's really it amm are uh you guys allow swearing on your platform yeah you can say whatever you want they're [ __ ] stupid they're [ __ ] stupid like I I I I mean like look they I I get why they exist like like all things that are considered dumb in hindsight they were smart at one point right um and I think like amm were really smart for basically you could really think of an amm as like an auction more than anything more than even like an order book let's say like I won't get too deep into the technical

Weeds on this but like when amm's got started the only thing that was being traded on them was like these illiquid shitcoins like nobody could get listed on binance or coinbase or whatever it was like [ __ ] I I can't get I don't have a plug with binance so I might as well just list it on Unis swap for example right like that was the whole reason they got started and that's like a really valid use case for an amm I think that's a great idea when when price discover is occurring on an amm that's but the moment on an amm that you start trading something that's more liquid somewhere else like you are basically just letting you are basically just screwing your users and like I won't I won't like call out who but like the the some of the people some of the actors

who who really shill amm um have done some of the most like genuinely despicable things in in crypto like research and in crypto analytics in general like publishing papers like that that basically deny things like impermanent loss exist things that deny that like Arbitrage or lvr exists on these amm like that's the kind of stuff I really hate and so like I'm I'm a very like I don't know a tenacious person and like you know I whenever whenever I see something that's not good I like to call it out and then when I see like people who are being really really uh malicious um for for whatever bags that they're holding um and they're denying that something malicious is happening like I want to call them out on that

immediately um and so that's what like a lot of my problem with amm like I don't have a problem with the philosophical concept of an amm right like just like I don't have a problem with the you know the concept of like I don't know [ __ ] Playdoh or something man but like don't lie to me about it don't lie like don't lie that's bad and I I like to one thing I like to really pride myself on is like when I see somebody like lying I'm gonna call I'm gonna call it out um and so I think like a lot of the discourse around trading in crypto has been done by people that don't understand anything about trading in crypto and or and or they do understand and they're trying to take they're trying to scam you um and so that's why like I I feel like I'm a

spokesman for the LPS like I want to be I want to be the voice of the unheard of the of the LPS of these amm and these these these these platforms and so like yeah lend lend me your lend me your your LP assets and I'll stop you getting absolutely Reed by like five trading firms that are like front running you I I had a tweet one time that like a lot of guys like a lot of my friends like to make fun of I was like I was like you're getting front ran back ran and butt [ __ ] and like you know I I I really want to protect uh protect the uh the LPS from from those three events occurring to them at least non-consensually at least non-consensually right like so yeah that makes sense I um I

think I want to ask two questions um first is one thing that you touched up on is that amm were originally built to be able to trade [ __ ] coins like El liquid [ __ ] coins on them you guys have permissionless asset trading um which is cool but I don't I guess what I'm wondering is how do you I guess incentivize solvers to pick up these orders and to fill them like is there no risk for them to keep or because they're just doing it like no no oh sorry didn't me interrupt you you have more for the uh so so here's okay solving is inherent so like uh solving today is Market making without taking any risk I'll give a little like like uh TDR for for the

listeners who aren't familiar with much of Market making like Market making or I guess solving today to start with solving and what why it's solving is super super easy like solvers do not take any risk whatsoever right now not in the future but right now um right now what they do is um they basically they run an algorithm they find you the best possible you know route like I'm going to break up your trade through these 15 different pools and you get the best possible number of tokens out um we're all very familiar with this that's super common that is what is called like Market making routing basically using algorithms that market makers use all the time um to basically find you the best possible output price um this is super super common in trading uh this

kind of strategy of basically optimize routing effectively through different exchanges um solvers right now do it all atomically so that means like the solver takes your assets routes them through all these amm and all these pools on chain and then return returns you the tokens in one transaction so they're never holding any inventory or any risk um Market making is like doing that but actually holding the inventory and taking the risk um and so you're not you're not able to do it atomically like I'm filling an order on binance and then I have to fill that order on coinbase and execute that hedge or I'm gonna lose money um and so uh yeah I think um that's like the way we basically handle the solver solver risk side of things is

like solvers right now are doing everything atomically but like in the future and you you're starting to see this a little bit on like cow Swap and Unis swap X and like a lot of these like more mature solving platforms is market makers are now coming in and filling filling Solutions using proprietary liquidity so they're like they're not routing your trade through a bunch of pools they're just they have a bunch of assets in a wallet and then they just basically will quote the price on binance plus a couple of basis points of of profit and then fill that order on the on the um you you know the the platform that you you whatever wherever they got that trade from for example so um I think yeah that's it's an interesting uh rabbit hole but yeah

sorry if I didn't no it's perfect eventually you see more and more um I guess solvers become market makers yes absolutely absolutely yeah that was yeah sorry just because of efficiency because they'll be able to Source it for cheaper it's it's actually just a it's a risk uh tolerance threshold really right because like what it basically I think like any solver that doesn't start to invest heavily in understanding how to do things like Inventory management and hedging is going to get left behind um I think as we move away from a world where liquidity is stored onchain and market makers are onchain via like amm and onchain clobs we're going to look uh it

the world will look more and more like centralized exchanges with decentralized settlement and clearing and custody and and so that means solvers are going the term solver is probably going to disappear it will just turn it'll it'll basically get merged into Market making um and this is not like one one tiny little like like annoying like asterisk is like crypto's kind of like kind of [ __ ] in the definition of like Market making like crypto people think Market making means doing Arbitrage that's not what Market making is that's not Market making at all that's just arbing like Arbitrage is not a market making activity like like maybe a market maker makes a little bit of profit in trafi off of Arbitrage but it's like you know pennies compared to what they make doing

actual Market making Market making is I am going to be the best I'm going to be a one I'm gonna have buy orders and I'm GNA have sell orders in the Market at all times whether or not I can hedge and I'm GNA price the risk I have of holding these assets accordingly so like I think the best way to describe what a market maker does like I I I love this analogy like basically my trading Mentor who traded at the Chicago mer for like 25 years um told me gave me this analogy and it explains Market making in like the most elegant way possible market makers are just used car salesman like that's all they are and so like what this means basically is Joe comes to me in the morning and says like hey I got to sell this car this car the fair value

of this car according to my estimation as a used car dealer is $22,000 and I say all righty I'll buy that car from you right now for 1,900 so I'm getting I'm buying his car for $100 cheaper than the fair value of it because he needs to sell it right now and I'm like okay he's like all righty because you'll buy it for me right now I'll give it to you at a discount so I'll give it to you for 19900 and so I as the used car dealer hold that car I'm taking the risk of that car declining in value so let's say the car stays $2,000 for that whole day at the very end of the day um you know Joe Schmo not Joe I'm using two Joe's uh uh Adam comes to me at the end of the day and says I want want to buy that car

and I say all righty that'll be $2,100 and he's like what that car is only worth 2,000 I'm like you need it right now though don't you and he's like yeah I need it right now I'm like all righty well you're gonna pay me 21100 for it and so when he pays me 2100 I'm netting $200 of profit and so that is what a market maker does in trafi a market maker is somebody who's saying I am always going to be here to buy and sell this no matter what um and they're GNA they're going to take haircuts off of your you know order because you're basically paying them to always be there for you to for S liquidity um and and I think solvers will have to adjust to being Arbors to being market makers uh I W to I I want to definitely

reopen the tab the amm tab that bacho uh you know presented um uh I definitely want to revisit some of the concepts there uh well first of all the way I think about aori is um you know more Market structure more defi Market structure which I think is a great thing like solvers being able to take more risk given this Market structure I think that's you know an Inc a really potentially valuable piece of infrastructure the part where I'm going to push back is on this notion that amm suck and the reason why is because I think amm are supposed to be dumb um I don't think they're supposed to be these like super sophisticated

like vehicles or or structures that super sophisticated like market makers are going to use I think amm are are supposed to be dumb um and so I think that classification is is is not like properly used when we think about amm and I think crypto has a habit of like using these fancy terms for you know things that might not necessarily look like that in the short term for example amm as one we're not necessarily Market making it's just these passive investors holding two assets that rebalance over time that looks very much like an ETF another term is Dows decentralized autonomous organizations Dows are not you know decentralized uh maybe some are

and when they are that's these coordination systems fall apart they're definitely not autonomous either there has to be some level of communication so I just want to like revisit the amm because I feel as if like uh the space doesn't necessarily classify that correctly and so I'm curious your thoughts on that classification and whether that like can potentially update your perspective onm so so I I definitely like I the the I I definitely get your point I will say this like as earlier I described when I said I don't hate amm I hate the people that lie about amm um I think like amm are just a tool at the end of the day right like like I can't it would be stupid of me to

be like I hate hammers because I like screwdrivers you know it's like makes no sense right like you need a hammer when you need a hammer you need a screwdriver when you need a screwdriver right the my the place that I think amm can be valuable is basically the same place that like auctions can be valuable when you have an asset that you don't know the fair value of it and you need some kind of natural price Discovery to occur um in this isolated market then yes I can understand having amm be be these things that like need to be used and need to be like more creatively used the problem is people go around act like they're the best thing since sliced bread and they're like we need to use this for all trading at all times and I'm just like what the [ __ ] are you

talking about like no that's stupid like so so that's my only like caveat to what you said yeah yeah I agree I agree with that 100% sweet sweet all right so we shot on amm enough what do you think is the future of lping and amm in the hold on I'm not I'm not shoot [ __ ] on amm I I think that amm if if defi if defi has done has brought like you know sort of going back to an early point I I also think that there's a lot of value from learning from traditional Finance like you know adapting these Concepts and figuring out how to apply them in defi but if there's one like purely defi application that we've created it's amm and I think there's a lot of value

in that well I I mean I would say per and money markets are much cooler in my opinion like I think I mean money markets not money markets and and Futures they deres Market those are structures those are products that I mean but but what is it okay okay I'll like dunk on amm then okay an amm is just an order book with rules that's it it's an order book where the the Only Rule is that there's one market maker and that market maker has a predefined price function that's I mean it's an amm is not like like when you when you break it down it is it is an order book where you have basically continuous limit orders they're not technically continuous because math and solidity is discrep um well in in part in the solidity that

like amm are written in but yeah Unis swap V3 is that2 unis swap V2 and you know for example unit swap V2 balancer these aren't limit order books these are just yes they technically are because because every single like it is if you think about it in the terms of like one side of the pool is the bid side and one side of the pool is the ask side you basically just what looks a lot like you know a standard Market maker where it's basically saying I'm going to provide this level of price of of liquidity at this price and at this price and at this price and at this price V3 just gives you the ability to instead of having to post your Li your your range on the

whole Spectrum you can post it on a section of the spectrum that's the only difference between V2 and 33 in my opinion I think I think we have different perspectives on on what amm are we should we should definitely unpack this like in another time are like [ __ ] everybody's got one no no I don't know what to say after that I I just I don't know I'm gonna say it Josh what we have because I'm cognizant that you have to go so I'm going to run through relatively quickly um I'm gonna ignore the lping an amm question even though I wanted your take on it we have to ignore it now but he he can write you a thread on it yeah yeah I'll write you

a thread or jump on again I'll write you a thread yeah um this one's also for you R A but what's the end goal vision for aori what does the full Suite of aori products look like and what is the ideal experience that you guys want from the user I'll give like a quick one and then Roy feel free to jump in well maybe maybe maybe I don't even think I'm qualified to answer this Josh Josh the founder yeah just to try and help him get there so if there's anything I missed feel free to jump in like I I talked a little bit about this earlier like like the vision is aori is is the whe it's a whether it's aori or it's just the fact that we build the tooling for it it's that all defi applications um that are not you know things like like what we

talked about earlier that are just like basically just data passing um all Financial defi applications move to intent infrastructure like aori where you basically have you know a a you have a predefined end goal that you want for something to occur and where're the rails that you do it through um and and I think like that's really what my dream is is for like the future of defi to look like is to look like you you have an open you know the problem with trafi uh is not like the trading structure of like order books or anything like that I think the problem with trafi is that to be a market maker in trafi you have to have hundreds of millions of dollars you have to have licenses you have to have like regulator friends you have to be part of SRO you have to be part of

there's so much fixed cost the price of being a D5 Market maker is hey are you really good at writing algorithms that are optimized or do you have like some kind of of the market that no one else has and I think that's incredibly beautiful and like my dream is not that we have 50 million different amm competitors against each other my dream is that like we have this um very basic concept of an intent or a swap or anything like that and we basically allow um a super open and competitive market of market makers to be able to compete and provide liquidity for whatever the end goal they user wants um and so that's really kind of what I what I dream a is is the infrastructure to make that possible that was pretty good Roy I kind of don't want you to add

anything in because that was really really good thanks man appreciate it thanks for having us guys thank you for coming on much appreciated man sorry I think I lost you for a second back oh he was just being dumb guys thanks for coming on thank you Namaste thank you for having us pleasure gentlemen are you currently building on polygon or interested in migrating season one of the polygon Community grants program is live right right now and features 35 million in madic to support the next generation of polygon projects builders on polygon these grants are for you join the aggregated future today by applying at polygon.net [Music]