The Evolution of the DEX Space | Will Warren - 0x and Matcha \\ Antonio Juliano - dYdX

you guys are Ki hey guys come on up will Antonio yep yep come on up let's do it here you go thank you for coming thank yeah all right I can start hey guys how's it going I am Antonio the founder of dydx and joining me is will the founder of zerox we're going to talk today about the history of dexes so will we've both been around in the space for a while now I remember you started 0x when I was at coinbase like eight or nine years ago and all of us knew what you were doing you came and hanged out with us us a lot at coinbase and I really did not understand what you were doing when you tried to come in and explain what the heck a DEX was to me and the rest of us at coinbase so how did you come up with the idea how did you get the conviction to start a decentralized exchange before that was even a thing yeah first of all thank you Antonio for for joining uh so pretty soon after the erc20 token standard was created this is around mid 2015 uh started experimenting with ethereum smart contracts in my spare time I was still in grad school so I'd spend like most of my nights and weekends like hacking on ethereum Smart contracts uh but pretty soon got kind of this core thesis that the erc20 token standard is such a powerful primitive uh and that allowing tokens to be globally accessible uh on a public blockchain like ethereum is is such a disruptive and Powerful new technology uh kind of developed this core thesis that all forms of value would end up tokenized on public blockchains uh like ethereum and uh so kind of like billions of different tokenized assets so things like traditional assets stocks bonds Fiat currencies uh things like video game items and airline miles and even things that we couldn't really create or couldn't exist before uh smart contracts platform like ethereum and one of the things that so initially we uh my co-founder Amir and I were trying to build a decentralized exchange venue so a a company that owns a decentralized exchange uh and we were in the San Francisco kind of ethereum developer community at that time and as we met more and more teams uh in that space so like augur and nosis uh and Mel inport and maker Dow the one thing that they all had in common is that they all required exchange functionality in order for their use case to be possible uh but they weren't necessarily focused on decentralized Exchange it was just a necessary building block uh so after enough of these conversations it kind of became clear that there is a much bigger opportunity to create a platform where anyone could create a decentralized Exchange Marketplace uh rather than just creating a single Dex application on its own uh and so that that you know kind of interacting with ethereum developers in those early days in SF that is really what kind of made the idea of a decentralized exchange protocol kind of Click for us um and and so that that's when we started developing zerx protocol uh in yeah around the end of 2016 interesting yeah I remember when dydx was founded in 2017 I actually was inspired by you I think I went to a talk that you gave in San Francisco at the time and I remember having this experience at coinbase and coinbase back in 2015 was an awesome place there were a ton of people Brian Armstrong of course Fred Olaf falic came and gave a lot of talks there and I remember it took me a while to wrap my head around what the heck is ethereum what is a smart contract but once I did it became clear that this is a new paradigm of computing and I was like there must be something that you could build on top of this that's going to be really valuable and it took me a while to figure out what that was but I remember going to the talk you gave and I was like damn this actually seems like the first thing I've seen that actually makes sense an exchange of value in a totally decentralized way and that's really how the Genesis of the idea for dydx happen where I took a look at things like ZX and there were kyber and some other decentralized exchanges at the time and asked myself what comes next after this and to me it was pretty clear it was margin trading it was derivatives it was more sophisticated forms of trading but I think this is a really good way to look at the evolution of markets look at something that's really early in working and ask yourself okay does this actually make sense is this going to stick around and then second of all what's going to come next after this so that was really the Genesis of the idea for dydx and obviously dydx and ZX have changed a lot over time I'm curious in a minute or two how would you describe the biggest impact of the change from what ZX was at the beginning to what it is today yeah good question so a lot has changed over over seven years now uh you know I think we we originally went out with this vision of uh let's build this open-source decentralized exchange protocol and our team we will not build any sort of user-facing application we're not going to build a business we're just going to provide this infrastructure and we're going to allow any anyone to go and create their own kind of marketplace or decentralized exchange on top of it and I think over time it's it's become clear that uh in order to build something that is going to be sustainable uh that is going to be used and stand the test of time uh building directly for a customer and building something that a customer is willing to pay for is critical to building the right thing uh and so you know if you really want to create you know create something that is going to drive a lot of value and that is going to stand the test of time you have to be building directly for your customer and that's actually one of the things that I really looking back like I really admire about your approach in in founding dydx uh at the time we were you know we were fully focused on let's just build this this public neutral infrastructure we're not going to interact with any end end user applications we'll just build for developers and I remember early on you were you were very much focused on how do we build the very best product for our customer and laser focus on that and I thought that was like a very unique Approach at the time uh and yeah I'm kind of curious uh yeah how how that kind of has evolved over the years I think it's stayed the same I mean this is just how you build new things you have to have a Clear Vision mines for who are you building for what's their problem and how are you going to solve it for them and this is the continuous question I ask myself when we build dydx and how dydx has evolved has changed a lot over time so we started with margin trading actually and fast forward to today we're focused on derivatives and Perpetual contracts and I think that really mirrors the evolution of the crypto Market as a whole back in 2017 margin trading was the dominant way that crypto was traded it was the highest volume market today that's perpetuals and derivative contracts dydx went through a pretty significant change about I what like four years ago now or so where we built our first perpetuals contract at the time and nobody had done this before we didn't know whether it was possible in retrospect a lot of these things seem obvious it's like oh yeah of course you could build a smart contract for what a Perpetual contract is but it's not obvious at the time and really going out into the unknown and solving these things that users actually want to use and this was right around the time when bitmex was blowing up perpetuals were becoming a really big thing in centralized exchanges so we ask ourselves okay is this a clear Market yes definitely it was really blowing up users wanted this do we believe this is going to come into defi yes of course you know this is our basic thesis that more of trading is going to become decentralized over time and then third of all how do we actually build this and then responding to what you said well I think this is I don't I don't know I have to make this point so often honestly but it's kind of a controversial opinion in crypto that product should come first like what are you actually building what users are you targeting who are you building for and I think people just don't ask themselves this enough I think that in crypto there is so much hype there's so much noise that you can get really caught up in it you can get caught up in the buzzword you can build your next L1 and collect your1 billion like that's not actually interesting to me what it's interesting to me is building an actual product that end users use and I think that is that's remain the same like dydx has changed a lot but that's the one thing that's remained the same so I feel proud of that and yeah I see that Evolution a lot in zerox as well yeah and so building for customers first building a product finding product Market fit and then gradually decentralizing over time it it seems like that's kind of the trajectory that you you followed uh and yeah curious you know one of one of your big customers uh as a derivatives platform are market makers and I'm curious you know as you spent more time building for market makers and learning about what's important to them how do you build you know a a margin trade or you know now it's a Perpetual platform that's doing a billion dollars a day how do you get to that point by interacting and learning from market makers yeah and it's interesting so talk about knowing who you're building for but a lot of these marketplaces are two sided right so you usually have two different types of users so our biggest classes of users are as will said more professional Market making firms and these Traders really require a higher amount of throughput really Advanced trading features an API experience that's similar to a centralized exchange and it's hard to do that in defi they require quite a lot of throughput so one of the things that both ZX and dydx have championed for a long time is the use of order books in decentralized exchanges so that remains the case for dydx today and really the big challenge that we had to solve in the last few years was how do we build a performant order book in a fully decentralized way that had never been done before um and we didn't again know if it was possible but the big solution to that that we've come up with in the past year or two is we've built the first ever decentralized or but offchain order book um and this is running in production today on the dyd exchain um and again we just really asked ourselves who are our users what do they care about they care about performance they care about reliability um they care about Advanced trading using an order book rather than just lping on an amm and that was a big Genesis of why we focused on that so much yeah and and so one of the things that's interesting is like in the early days of ethereum there was kind of like an era of order books that weren't Central limit order books meaning that there was no matching engine there was no guarantees around price time priority uh orders could get front run there could be trade collisions and uh the central limit order book model uh it seems like that's one of the things that really is important for attracting professional gr you know institutional customers uh and I'm curious if yeah if that kind of evolution was uh something that was directly requested by market makers or if it was something you kind of uh came you know came to that conclusion on your own yeah and I mean honestly give you guys most of the credit for popularizing order books within decentralized exchanges but absolutely I think the key piece is how do you do matching in a decentralized way on an order book so when there's orders that cross how do you actually fill them this is a fairly hard problem um so it certainly was something that seemed required again we really start with the product first and asked ourselves who are users okay they're market makers where are they trading they're trading on centralized exchanges how do we build that how do we make something that's really similar to a centralized exchange but in defi these are hard problems but that's what you're supposed to solve as technology ologist so you know we've gone through a number of iterations on this dydx has taken a progressive approach to decentralization and we started with running the order book and the matching engine in a Central Way sort of similar to what a zerx Rel layer um is or used to be and now we fully decentralized that so I think it's been quite a journey Switching gears a little bit and I feel like we should challenge ourselves to ask the hard questions I'm curious how you and and I'll answer this too how did we both react to the rise of UNIS swap about 4 years ago cuz I feel like that really took the def Market by storm and probably both of us as well yeah I mean there is there's no denying that Unis swap version one completely changed the landscape and the entire defi space wouldn't be the what it is today without Unis swap coming and introducing the concept of a constant function Market maker and I think you know enormous praise to the Unis swap team uh you know we are a DEX aggregator uh so we you know aggregate Unis swap liquidity in a lot of ways so you know there's always some tension in in uh in the space uh competitors versus uh Cooperators so coopertition uh interestingly so when Unis swap first came out I was uh actually didn't really think it had that much potential yeah me too and and the reason is because uh Unis swap version one is is very Capital inefficient so uniswap version 3 introduced this concept of concentrated liquidity you can kind of push all that liquidity closer to the to the spot price of the market but in unit swap versions one and two you had to essentially provide full range liquidity uh and this is super duper Capital inefficient one of the things one of my greatest assumptions which was turned out very wrong was that it wouldn't really be possible for hundreds of millions of doll to end up deposited into a a constant function Market maker like Unis swap it just seemed completely like a crazy concept that people are going to put hundreds of millions of dollars into this thing and that is what is necessary to create a sufficiently liquid Market turned out super duper wrong uh but it also uh yeah I think it was uh you know an important learning as well uh Unis swap also made it possible for longtail tokens to be liquid so one of the key yeah kind of the key uh disruption or the kind of the key uh the key of Unis swap was that it lowered the barrier to entry for standing up brand new markets so instead of having to go to a binance uh or a a coinbase and to attempt to get your token listed uh now anyone could create a token in 10 minutes they could go spin up a Unis swap Market in 10 minutes they could deposit some tokens and passively provide liquidity and this was critical to this tokenization thesis uh and that was one of the things that I think Unis swap just completely unlocked uh and you know definitely wouldn't be here today if they hadn't yeah I think it's really interesting the way you started this talk with the conviction you had at the beginning was there's going to be so many tokens tokens for everything it needs a medium of exchange and I think all your points I thought as well on Unis swap you know I felt like it was okay this is fine this is kind of like a simple exchange that maybe people will use a little bit but you have to understand where the space was back in 2020 2019 or so when Unis swap was founded there was just not very much volume there was not very much capital in crypto and especially in Defi and it was really hard to foresee the rate of explosion that Unis swap had at the time and I had a really couple really important learnings from that one of the things that not that many people know about dydx is that that dydx used to be the biggest decentralized Exchange in the world by a lot four or five years ago so in 2019 2020 dydx had over 50% market share for decentralized exchanges we were sitting pretty yeah the market was really small but we felt like we understood this Market it was going to grow and we were the leader and the thing that happened to us in 2020 is we got absolutely by Unis swap like we went from 50% market share all the way down to 1% market share in the course of 6 months and we grew a little bit over that period of time this was like the bull market in 2020 um but not nearly as much as Unis swap they just absolutely started going crazy going exponential and it's really the reason for that was was hard to understand immediately to me but over time it became very clear that it was because they had the ability to trade so many markets on Dy and sorry on Unis Swap and dydx did not have that we were focusing just on three markets Market at the time and unisoft swap started having hundreds thousands of markets and we really had to change our strategy around that and you know fast forward to today dydx is around the same volume a little bit more than Unis swap normally derivatives are higher volume than spot um and we've made a comeback but the biggest learning that I had from that is there's this really core driver I feel of what becomes popular in crypto and I think the core driver of what's become everything that become popular in crypto is it's really about the number of things that can be traded about the number of assets that exist on the chain and this basically defines and in my opinion explains pretty much every Trend that's become really popular in crypto it's like okay crypto is invented there's like five assets you can trade back in the day Bitcoin eth coinbase and a few other exchanges support those fast forward to 2017 icos become a thing now there's hundreds of assets that you can trade fast forward to 2020 Unis swap enables thousands of markets to be traded there's another bull market it blows up fast forward to 2021 2022 nfts become a thing now there's like tens of thousands hundreds of thousands of things that you can trade and I think this is the really core thing that we're focused on now at DDX and the core learning that we've had is really about the number of markets that you can trade about the quality of those markets and liquidity that you can give but I really think one of the cool things about crypto is that it's so transparent you can see exactly what your competit are doing you can look at their smart contracts you can see exactly what their metrics are and I think for anybody that's building in the space the thing I would encourage you to do is do that look at your competitors and be really honest about that and it's hard to do especially if you're losing or especially if you haven't won yet you know to be really honest what's the core reason why your competitors are beating you or what's the core reason why Unis swap dydx whatever else has become popular at this point how can you do better what's the next thing you can build and that was the really big learning I feel like I had from Unis swap yeah permissionless Finance the ability to permissionless innovate and create anything you want on the platform without asking for permission yeah I agree so on that note I'm curious uh I I know that uh you know dydx chain uh I know that you're making a push to add you know have the ability to support more tokens more more markets and I'm curious how you're thinking about that over the next years so this is the core thesis that we're trying to play out on dydx now can we have hundreds of markets thousands of markets on dydx and there's a lot that goes into this but the big thing we're developing right now is the ability to have permissionless Market listings on dydx so in the same way you anybody could come to Unis swap just send a transaction and add a market to Unis swap could you do that for a derivatives market and that's a hard problem but again that's kind of our specialty is solving these things and we really feel like we can build that in the next couple months and I think that's incredibly exciting when combined with derivatives because with derivatives you can trade anything all you need is an oracle price and you can start trading the thing and I think it's just again a continuation of the thesis we've seen play out in crypto for a long time now it is permissionless finance and really the thing I try to ask myself is what can you uniquely build with defi if it's not unique um if it's not something that's uniquely enabled by defi or by whatever technology that you're building on you shouldn't build it if it is uniquely enabled by whatever technology that you're building on and solves the needs of actual customers there that is the business that you should build so that's the question I always ask myself what's something fundamentally new that we can build with the ydx and I think you see a lot of that going on in crypto and and that basically explains everything that's become popular so far strong plus one well we are at time that was quick uh but I think we'll leave it on that note thank you so much Antonio really appreciate you uh walking us through some of your thought process and uh thank you all for attending appreciate it awesome thanks a l