The Pomp Podcast - #776 Building Decentralized Exchanges For Retail Traders w/ Mark Lurie

Starting point is 00:00:00 What's up, everyone? This is Anthony Pompliano. Most of you know me as Pomp. You're listening to the Pomp Podcast, simply the best podcast out there. Now let's kick this thing off. Mark Lurie is the CEO of Shipyard Software. They specialize in building decentralized exchanges, especially things for specific types of users in the DeFi space. In this conversation, we talk about decentralized exchanges, all the technical trade-offs that go into building them, how they think about liquidity management, and what the change in incentives are in this new system. I really enjoyed this conversation with Mark, and I hope you do as well. Before we get into this episode, though, I want to quickly talk about our sponsors.

Starting point is 00:00:41 First up is LMAX Digital, the number one institutional crypto exchange. They offer clients the deepest pool of crypto liquidity on the planet, underscored by 100% uptime track record through volatility spikes. They leverage LMAX Group's liquidity relationships and ultra low latency technology. LMAX Digital is the market leading solution for institutional crypto trading and custodial services. If you've never heard of LMAX Digital, it's probably because you're not an institution. They have no retail, only institutions. They feature a central limit order book streaming spot Bitcoin, Ether, Litecoin, and Bitcoin Cash, all paired with US dollars, Euro, and Yen. LMAX Digital, they're secure, they're liquid, and they're trusted. Learn more

Starting point is 00:01:20 at lmaxdigital.com. Again, lmaxdigital.com. Next up is Compass Mining. Compass Mining is the world's largest marketplace for mining hardware and hosting. With Compass, everyone can mine Bitcoin. Their team makes it easy to start mining wherever you want, at home or in one of their 23 hosting facilities around the world. Through the Compass marketplace, retail miners can access mining hardware with similar prices and purchase plans as the world's largest mining companies. Compass miners own their machines, they choose whatever mining pool they want, and they mine directly to their own wallets. Miners who don't want to host their machines can order ASICs directly to their doorstep. Simple and low-cost hosting agreements coupled

Starting point is 00:02:05 with best-in-class customer service are the reasons why Compass is the simplest and most popular way to mine Bitcoin. Start mining your own Bitcoin today by visiting compassmining.com. dot io again compass mining dot io go check them out and let me know what you think last but not least are my friends over at okx crypto is all about democratization and freedom of choice but many companies limit their offerings to centralized trading products the crypto companies leading the pack in terms of innovation are those that extend access to the industry's cutting-edge products and services bridging c5 and d5 if you're searching for a platform that reflects crypto's promise of a more open and less restrictive financial future, look no further than OKEx.

Starting point is 00:02:47 On OKEx now, you can easily switch over to the new DeFi mode, connect OKEx's bespoke Web3 wallet via browser extension, and start exploring opportunities at the bleeding edge of crypto. From the DeFi dashboard, you can monitor your portfolio of self-custodied assets across a range of blockchain networks and generate passive income from yield farming with top DeFi protocols. In the NFT marketplace, you can participate in exclusive drops and trade non-fungible tokens without secondary market fees. Meanwhile, the GameFi section is your portal to the latest and greatest in play-to-earn and blockchain gaming. Venture to the forefront of crypto innovation and connect with OKEx DeFi today. Again, go check them out at OKEx.

Starting point is 00:03:28 That's where you can find OKEx DeFi. All right, let's get into this episode. I hope you guys enjoy this one. Anthony Pompliano runs Pomp Investments. All views of him and the guests on his podcast are solely their opinions and do not reflect the opinions of Pomp Investments. You should not treat any opinion expressed by Pomp or his guests as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of his personal opinion. This podcast is for informational purposes only. First, let's kick this off with decentralized exchanges in general.

Starting point is 00:04:00 For those that don't know, help us understand what the difference between a centralized exchanges and a decentralized exchange. Sure. So in traditional markets, the way you exchange assets is through a central limit order book. This is what you're probably familiar with if you've gone to the New York Stock Exchange and you've tried to trade assets. And basically what happens is a bunch of buyers say, I'll buy X units at Y price, and a bunch of sellers say, I'll buy Z units at W price. and there's a stack of orders and they're happening thousands of times a second and they're matched at the margin and that's the market price. And that's how most markets work in the traditional financial system. The problem is that that doesn't really work on the blockchain

Starting point is 00:04:50 for two reasons. First, it's really slow and you just can't have thousands of orders popping on and off, given how slow the blockchain is. And the second, it's pretty expensive, right? The gas to post an order is pretty expensive. And the storage cost for keeping stuff on chain is pretty expensive. And so a central limit order book just isn't really practical. And so that's why this new model called automating market making was developed. And so I can tell you about that. But first, any questions? No, let's do the AMMs, the automated market making. Explain what that is. Okay, great. So first some context here. So Shipyard designs decentralized exchanges that uses automated market-making technology. And we're building a bunch of these different exchanges that are kind of best execution for specific types of users. And the reasons we're good at this is for a couple of reasons. One is, you know, we've been in the space for a while. And the second is, Abe, my partner and our CTO, is one of the foundational researchers in making automated market makers work for this type of context.

Starting point is 00:06:05 So how does an automated market maker work? Well, instead of this central limit order book, there's a standing pool of capital on the blockchain anyone can deposit assets into. So let's say you put a bunch of ETH and a bunch of USDC in, and that just sits there. That's liquidity pool. And there's a math formula in the automated market making that says, here's the price. Anyone can trade with this pool for a price determined by this math formula. And that's like a very, it's pretty simple code. And it means you don't need thousands of orders flowing around. You can just have a trader that comes and says, well, I want USDC and I'll put an ETH or vice versa. And the automated market maker will always quote a price.

Starting point is 00:06:52 Now, why do the liquidity, why do the people who contribute capital to the liquidity pool do this? Well, a fee is added for every trade. And that fee makes the liquidity pool larger. And that creates yield and return for those liquidity providers, which are known as LPs. So that's why the whole system works. It's kind of like a marketplace. Everyone's happy. Absolutely. So obviously, the most popular decentralized exchange so far has been Uniswap, or at least historically, the most popular. When you think about that, you guys say, hey, look, that's a great idea. We love decentralized exchanges, from what I understand. but we think that there are certain technical trade-offs that are made to get the most amount of liquidity onto a decentralized exchange. And so that may not be the ideal venue for everyone to trade there.

Starting point is 00:07:42 It's really good for, it sounds like, very large transactions. So if I do hundreds of thousands or millions of dollars in a transaction, then going to somewhere where there's tons and tons of liquidity is great. But if I want to do, let's say, a $1,000 transaction, going to those large liquidity kind of venues

Starting point is 00:07:59 may not actually be the best thing because of some of these fees. So maybe help me understand in terms of how you look at the difference between let's say Uniswap and then the product you guys built. The first one is called Clipper, which is more focused on like the retail trade. Like what are the big differences here? And depending on who you are, why should you go to each one of them as a trading venue? Makes sense. Yes. So that's exactly right. So Uniswap was really the first decentralized as exchange or DEX. And it has a very simple formula, which is essentially it has a price based on the ratio of assets in the liquidity pool. And this is kind of a one-size-fits-all

Starting point is 00:08:41 approach, right? Everyone goes and uses this automated market maker and the pricing formula is X times Y equals K, asset one times asset two equals K. And it turns out like a one size fits all, a pretty simplistic approach isn't the best for certain people. Some people have very small trades and there's a different math formula which would result in different prices. Some people want really big trades and they want a different architecture. And that's why you don't have one huge market in the traditional financial industry. You have many different exchanges for different types of trade, traders or instruments. And so in the case of Uniswap, because their formula is, and pricing formulas is kind of based on the ratio of assets in the pool,

Starting point is 00:09:27 they want as many, as much liquidity in the pool as possible. So Uniswap has billions of dollars. And what that means is that when you do a really large trade, let's say a $500,000 trade or a million-dollar trade, you're going to get less slippage, slightly better prices. And that's how Uniswap competes. But it turns out that a small trade, let's say like $1,000 trade, doesn't really matter from a price perspective if you're trading in like a $10 million pool or a billion-dollar pool. The slippage, the price change is pretty low. But you're still paying high fees on Uniswap because Uniswap needs all these fees to generate yield, to attract billions of dollars of capital that then helps the large traders. And so retail traders are really

Starting point is 00:10:18 subsidizing whales. And so we looked at that and we said, hey, that doesn't really make sense. Like if we're retail traders, we don't want to be subsidizing whales and paying high fees just so they can have more liquidity. And so what we did with Clipper is we say, let's carve out a portion of the market focused just on retail traders. This is Clipper.Exchange. And so for retail traders, what we did is we capped the liquidity pool. We said, let's only let $20 million in liquidity. And what that lets us do is drop the fees dramatically. And that means slippage increases a bit, but for small trades, it doesn't increase that much. And the drop in fees outweighs the increase in slippage. And so what we have is better prices

Starting point is 00:11:01 overall. Now, it means we have terrible prices for large trades. There's really high slippage, but that's okay. That's not what Clipper's focused on. And so what we've done is just made a series of design trade-offs because we have a specific user in mind, which is the retail trader. And so if you are a retail trader making trades under $10,000, you should go to clipper.exchange instead of Uniswap, and you will get a better price. If you're a large trader, you should go to Uniswap. Or maybe you should go to another exchange built for large traders. Because execution really matters. And if you're not getting the best execution, you're losing money that could be in your pocket. Makes complete sense in terms of there's specific design or

Starting point is 00:11:42 kind of technical trade-offs that you make. And if you optimize for certain types of people, then you'd make different decisions. Help me understand, what are some of the other types of decentralized exchanges that you all see as opportunity? Obviously, don't reveal anything that would be detrimental to the business. But are there other things that you're like, yeah, we definitely see the opportunity to build specialization and other kind of design trade-offs?

Starting point is 00:12:05 There's so many, so many. We're still at like such an early stage of design for decentralized exchanges and DeFi in general. So we're focused on small trades, retail trades on major pairs like BTC and ETH against stables. There's another design for really new tokens where volatility is super high and price is a lot harder to determine and market depth is lower. And so there's a whole other design just for that. There's another design just for large traders because they really don't want to be front run.

Starting point is 00:12:46 They have different concerns. there's uh you know there's there's another design for um you know different types of assets whether it's derivatives or synthetics etc etc and so there's there's just like a lot of different there's as many designs as there are types of traders and users because you're always making design trade-offs and if there's a different type of user there's going to be a different type of design trade-off for them got it and then when you think about how uh these decentralized exchanges Obviously, there is decentralization around the technology itself that you guys are building, but you guys are a centralized company. How do you kind of balance the centralization of the company and hiring employees and a marketing team and software developers and all that and kind of a more traditional corporate structure with the building of decentralized products?

Starting point is 00:13:37 It's a really tough problem. Um, and you know, the, the, I think a lot of DeFi projects struggle with this decentralization dilemma. Um, so in our case, Shipyard Software is a U.S. corporation. We develop the software, uh, the code for decentralized exchanges. Clipper is our first exchange. Um, Clipper is public software that runs in a decentralized way on the blockchain, even though we developed it. So we are not Clipper, but we are the developers of Clipper. Over time, we want decentralized governance, but that's a really hard and tricky thing to do. You know, you have very centralized companies that essentially dictate every product decision.

Starting point is 00:14:20 And then you have really decentralized companies like SushiSwap, very decentralized organization, but they have all sorts of dysfunction as a result because there's not clear leadership. And so, you know, it's a tough problem. And I think it's to solve this decentralization dilemma, you have to kind of progressively decentralize. There's no clear way to do that, but it's something we're committed to figuring out. Yeah, it's really interesting when you start thinking about the various challenges here. And also, I guess part of this is, you know, decentralization. I think for the most part, people think about the importance of it is either security, some sort of resistance to state attack. They immediately jump to, I think, what is the classic uses for decentralization.

Starting point is 00:15:05 In some way, what you're talking about here is less about the ability to resist state attack or something around security and that nature, and it's much more about a different way of having the order books operate and get liquidity onto the actual trading venue, and it changes some of the economic incentives. It allows for you to do different technical trade-offs, and so it's unique to kind of think through why people are creating these decentralized exchanges, not just how they're doing it as well. Yeah, that's right. And, and, you know, if you think about decentralization, I think you can be clever about how you're decentralizing, right? So you said a few one,

Starting point is 00:15:44 you know, one is resistant from resistance from state censorship. You know, another is governance by its users. There's a whole bunch of others. And so, to say that everything needs to be decentralized is maybe a little bit of a red herring. Instead, you can focus on specific elements of the product that need to be decentralized in the way to ensure censorship resistance, in the way to ensure there's clear leadership, but users have a say in the roadmap. And so I think there's a way to balance all these things that accomplishes the goal without throwing this label like, you know, you know, decentralization or bust over everything. Yeah. Because that can create dysfunction.

Starting point is 00:16:33 Got it. And I guess, Joe and John, what questions do you guys have? And then I'll continue. My question would just be around going after specifically retail traders versus more institutional traders, right? Like, why was that the decision? What kind of tradeoffs are there when you cater to one group rather than the other? Yeah, it's a great question. Our mission is to, as a company, is to help create free and fair markets and for as many people as possible, right? Because markets are like an engine of human progress. And often we see they're not really free. Often they're manipulated or constrained, especially by authoritarian governments. And often they're not fair as in the rules are changed after the fact. And so we thought, well, like, how can we manifest that mission?

Starting point is 00:17:17 Well, we really need to touch the most number of users. And that meant focusing on retail because that affects the most number of people as opposed to a handful of institutions and market makers. So that's why we did it. And then we kind of thought deeply about, OK, well, like, what are the design tradeoffs we'd make just for that user? But there's a tradeoff there from a business perspective. And that's that small trades, you know, we get lots of transactions, but not high volume, right? Because they're small trades. If we were to focus on institutionals, we'd have small number of trades, but huge dollar value in transaction volume. And, you know, that's a vanity metric that a lot of people in the market care about, even though I would say they shouldn't. But, you know, that's a trade-off from a business perspective, but we're being true to our mission. And so we feel good about it. John, what do you got? Mark, can you talk about the different types of customer acquisition you guys have thought

Starting point is 00:18:13 about through incentives and other forms? That's a great question. And I think it's one that people in crypto need to focus more on. Classic, you know, entrepreneur's mistake is having a great product, but no way to get users, no distribution. Distribution really matters. So how do we get distribution? We plug into DEX aggregators.

Starting point is 00:18:37 So, you know, DEX aggregators are like kayak.com for flights. You go to kayak.com, you put in, you know, where you want to go from Boston to Salt Lake City, and they show you the airlines with the best prices. And most people just go for the flight with the cheapest price. Similarly, in DeFi, there's DEX aggregators like OneInch and Matcha, which is built by ZeroX, and you put in the trade you want to make, and they'll show you the price you could get from 100 different DEXs, and they rank them. And because we have the best prices on small trades, we come up first on small trades.

Starting point is 00:19:12 And that's where we get most of our flow. What we also find is that people get excited about our community and they see us on the DEX aggregator listings. And then they start coming to us directly and realize, hey, I should just do all my trades here because I'm getting the best price and just go direct. So we've had a lot of organic usage come up. But our main customer acquisition channel is DEX aggregators. And I would recommend that other DEXs or bridges or what have you really focus on that as a distribution mechanism because they really need to have one.

Starting point is 00:19:47 If you are okay with sharing, can you just talk about the incentive for those DEX aggregators? Is it on a percentage basis of the trade or is it just on an overall one-time fee? How does it work there? So different DEX aggregators have different models how they make money. Um, but their mission is to provide the best prices out there for their users. And so the reason Dex aggregators want to integrate with us is because we have best prices so they can then deliver that to their users, right? They're constantly competing with each other.

Starting point is 00:20:24 If they don't have the decks with the best prices, then people are going to stop using one inch and they're going to start going to matcha, uh, or vice versa. And so that's the strategic imperative for them to integrate well-performing products. The way they make money is, you know, some have a token, some charge a fee, you know, some charge a spread on gas. It depends, but they're taking some, they're usually taking some fee of some sort somewhere. Mark, when you think about building this stuff, talk to me about like the team. Is this something that takes like two or three developers a couple of days? Is this like a, you know, six month, 12 month roadmap type stuff?

Starting point is 00:21:07 Like how difficult it is actually to create these DEXs from a technical standpoint? And how do you think about allocating human resources to do it as you kind of work on multiple ones across time? Yeah, it's a really good question because on the one hand, you know, one developer can fork a DEX and throw it up on the blockchain in like, you know, a day. Um, but a lot more goes into success than that. Um, one is, you know, you have to think deeply in order to be differentiated. And so we, we have a research team that really spends a lot of time thinking about what are

Starting point is 00:21:44 the right technical trade-offs that takes a while. Then we also have, you know, uh, uh, uh, marketing, uh, team and community is so important in crypto, uh, to get people involved, to get users. And so we also have a marketing team. We work with a great firm called Serotonin. And then there's also a legal regulatory aspect. So there's like a lot that goes into it. You know, I'd say we have about 15 FTEs, like full-time equivalents at this point.

Starting point is 00:22:14 And I think we're running pretty lean. So this takes real resources to do well. You know, you can just throw something up and people do. But those are hit or miss. And as many examples of successes you see there, there's probably 10 or 100 X of of, you know, failures. And I guess I would say I don't think they're innovating or doing something thoughtful. And a lot of times you see that in the yields. Yeah. And so I guess my last question for you is just around capitalism is all about how do I build something, service a need and make money? When you build the decentralized exchanges, there's like an incentive alignment with the

Starting point is 00:22:57 liquidity providers, obviously. How do you guys make money? How do you think about like where you make money versus where maybe you shouldn't make money or you should pass that through to other folks within the ecosystem? Like how do you think just about the design and almost like system decisions that you have to make around how you sustain yourself as a business? That is a great question. I'll talk a little bit about the DEX, but I won't answer that in detail, partially because in some ways it's TBD. Okay, that's fine. So how do the LPs make money? The more volume there is on us, if you can hold the liquidity

Starting point is 00:23:43 constant, the more fees will come into the pool, right? It's really not fees. It's that the prices are favorable. And so the pool appreciates. So like in our Polygon, in Clipper's Polygon pool, you know, right now we've been going for a month. We're running at 50 to 70% annualized yields with no liquidity mining. That's just like natural annualized yields if you're in the pool. and so that's like there's money being made there now i won't comment on our business model but i can comment on how other dexes are doing things right so uniswap they take uh and sushi swap they take a fee a percentage of that fee and that either goes to their token holders or it might go to a treasury uh and then that funds operations so that's a pretty common business model you see

Starting point is 00:24:35 in the space. We have not done that. It's certainly something we could do. There's also legal considerations around it. But I'm generally of the attitude that we're so early in the space that if you just make really good things that people use, the rest will figure itself out. And we've raised a bit of money to cover ourselves in the meantime. So we raised for Polychain and and a few other businesses, firms who, you know, who help us cover our expenses while we make a big splash in the market. Sounds like you need money for the pomp boys. That's what I heard there.

Starting point is 00:25:13 All right. So I guess one other thing that that made me think about is as you're building this, why should somebody come work with you guys rather than go work at Coinbase, Gemini and some of the centralized exchanges? Are there like fundamental differences in the experience they'll have in the ethos of the business and the cult? Like what is the big difference? Or is it just like, no, literally it's just technical architecture, centralized or decentralized. You're still working on exchanges and there's not as much difference as you might think.

Starting point is 00:25:44 I think if you're thinking about where to get a job, I think that you need to think about a few things. First is the mission of the business, right? I've talked a little bit about ours, so that has to be something you believe in. Um, second is the kind of like size and stage of the company, right? If you want to, uh, work at a really large company, then like, you know, you probably won't like a startup. And I think Coinbase at this point is a pretty large company at a startup. You can have enormous impact.

Starting point is 00:26:15 Uh, you can move really fast, uh, and you can, you know, make progress on really innovative things, uh, really quickly. And then I would say within a startup, like, you know, you want one that's doing, it's like thinking deeply, because if you have a team that has experience and is thinking deeply about users, they're going to be successful one way or the other. And so you're going to be getting on a rocket ship. And the final thing I'd say is management. I think you want to join a team where leadership cares not just about the mission of their

Starting point is 00:26:49 business, but wants to make sure that also furthers your aspirations, your career aspirations as an individual and is willing to mentor and encourage those. So, if that is all stuff that you want, then I really encourage you to reach out for us. We're looking to build best-in-class product, best-in-class research, best-in-class marketing community, you know, and so we're hiring, um, we'd love for you to, to go to shipyardsoftware.org contact us. You can also follow me on Twitter at Mark Lurie. Awesome, man. This is super cool. I love the idea of understanding the technical trade-offs and basically building specialized, uh, products for special users, you know, or certain types of users feels like that is a great wedge into the

Starting point is 00:27:34 market. And it's obvious because what Clipper is top five, top 10 decks in the world by users. yeah by users okay so when you yeah obviously you guys are at a disadvantage because you have smaller uh uh you know kind of transaction sizes just given the nature of of who goes there but i mean that alone speaks for itself right in terms of uh uh people want to want this stuff so it's pretty cool you built that uh where can we find people to uh to find you on the internet do you want us to send them you want to go twitter you want to go linkedin smokescreen what do you want I would love, you can find me on Twitter at Mark Lurie, M-A-R-K-L-U-R-I-E. You can also visit clipper.exchange.

Starting point is 00:28:15 That's clipper.exchange, C-L-I-P-P-E-R.exchange, where you can get the best prices on trades under 10K. For those listening, you should really use it. And if you want to read a little bit more about our team, you can go to shipyardsoftware.org and learn about who we are. awesome you guys got anything else nope thanks for coming yeah thank you legend keep it up i can't wait to see the next one you guys launch it's gonna i feel like you're just on this mission they're just banging out dexes so i'm uh i'm excited that's what we're doing we'll let you

Starting point is 00:28:51 know when the next one launches thanks so much sounds good mark thanks

The Pomp Podcast - #776 Building Decentralized Exchanges For Retail Traders w/ Mark Lurie

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.