"People Are Seeing The Writing on The Wall; DEXs Will Rule Crypto:" dYdX's Antonio Juliano

In this week’s episode we speak with Antonio Juliano, founder and CEO of dYdX, one of the biggest decentralized exchanges on Ethereum, focusing on derivatives. dYdX recently made news by being one of the first major DeFi projects to move to a Layer 2 scaling solution.

He talks about how dYdX will change for the user on Layer 2 —from transactions that cost cents instead of hundreds, being able to use one asset as collateral for many different tokens, to unlocking higher leverage. Juliano also explained why dYdX decided to go with StarkWare’s zk-rollup solution, instead of another Layer 2 solution or even on another blockchain, and it comes down to decentralization and security.

One of the main goals for his team this year will be to further decentralize dYdX, Juliano said, from the order book to the control of their smart contracts. So yes, for those reading between the lines, that means decentralized governance, and Juliano isn’t ruling out a dYdX token.

Juliano believes derivatives are going to be the biggest products in crypto and so his long-term goal is for dYdX to become one of the biggest crypto exchanges, period. He knows DEXes are just 5% of total crypto volume, but that’s up from 0% two years ago. He says, a lot of people are seeing the writing on the wall.

The podcast was led by Camila Russo, and edited by Alp Gasimov. Transcript was edited by Owen Fernau.

🎙Listen to the interview in this week’s podcast episode here:

You’re a paid subscriber, which means you get the full transcript below. Subscribers also get exclusive access to The Defiant’s Discord chat for the community, here’s a new link to join.

Share The Defiant

🙌 Together with:

• Zerion, a simple interface to access and use decentralized finance
• Balancer, one of the leading DeFi automated market makers (AMM) for multiple tokens. Dive into their pools at https://balancer.finance/!
• Kraken, consistently rated the best and most secure cryptocurrency exchange, which can get you from fiat to DeFi
• Casper, an enterprise-focused blockchain which aims to introduce unprecedented security, speed and scale for businesses

Antonio Juliano: I got into crypto back in 2015 when I had my first job out of college, which was at Coinbase as a software engineer. And I kind of got to Coinbase pretty differently than the way most people got there. The way most people got to Coinbase, especially back in 2015, is that they were really into Bitcoin, and especially back then in 2015, Bitcoin was really all there was in terms of the interesting stuff going on in crypto. But I didn't really know that much about it.

And through the interview process met a lot of really awesome people, was really fortunate while I was there to be able to work with just a lot of the top people in blockchain at the time were at Coinbase, or if they weren't they were probably coming to Coinbase and giving talks. Like Vitalik came really early on in the life cycle of Ethereum to talk to us, and Olaf Carlson-Wee, was there giving presentations on smart contracts and gas usage on Ethereum, and all this kind of stuff. So that just got me really excited about what was possible to build based on that new technology.

And I wasn't exactly sure what I wanted to build at that point on Ethereum, but I was pretty convinced that there would be something to build on top of smart contracts that would be really big. So I really see smart contracts as this new computing paradigm where for the first time you can write these programs, they execute totally autonomously, totally deterministically, obviously not controlled by anybody. And that's just really fundamentally different than what you can do in any other area of computing. And every time you see some new step in terms of something new you can do in computing, they're basically always is interesting stuff to be built on top of it.

“…for the first time you can write these programs, they execute totally autonomously, totally deterministically, obviously not controlled by anybody. And that's just really fundamentally different than what you can do in any other area of computing.”

So anyways, that was back in kind of 2016, eventually left Coinbase in 2016, and after a few months, I started my own thing. And the first thing I started actually wasn't dYdX, it was something else, it was a search engine for decentralized apps and this was back in 2017. So I built this whole thing out. I was working on it for a couple of months and literally nobody used it. I think I had like 10 users literally ever. And the main learning that I took away from that was in addition to building something that's useful, the timing of when you build it is really important. Like probably someday, there will be a search engine for decentralized apps. But it certainly wasn't back in 2017 when there were less than 100 decentralized apps total, and what is the point of a search engine if there's nothing to search for?

“…the main learning that I took away from that was in addition to building something that's useful, the timing of when you build it is really important.”

So after I spent a few months on that, I took a step back and was thinking about well, okay, if this isn't working out, what's something else on top of Ethereum that I could build that’s interesting. And that was right around the time when the very first decentralized exchanges were starting to come out. So things like Kyber and 0x were just starting to come out, took a look at this, spent a lot of time wrapping my mind around it, and got pretty excited about it.

And I thought that’s kind of the next logical thing to build after just base level spot, which just means like a fancy word for like buying and selling decentralized exchanges, would be more advanced financial products, like margin trading and derivatives. Because I really think of finance as a stack, especially decentralized finance as a stack where first you have decentralized money like Bitcoin and Ethereum and stuff, then you have decentralized spot exchanges like at the time, it was Kyber and 0x, and then the next logical thing to build on top of that is derivatives. But you can't build anything higher in the stack without things lower in the stack, and that's kind of what I was getting at back with the timing point. So been working on dYdX since then, we’re founded in 2017, so I've been around for a while now. And yeah, I guess kind of the rest is history.

CR: You didn't really have a financial background when you started this project so it's interesting that you came at this very just technically like purely financial product from a technical perspective. Was this because finance is kind of the low-hanging fruit on Ethereum, like what makes sense to build on the network?

AJ: Yeah. I think this thought certainly extends beyond dYdX to kind of all DeFi. But I think the area, that trust minimized computing can be the most effective in is finance because trust is so important in finance. And in the traditional financial industry, it's just really run through mostly centralized intermediaries, so it’s very ripe for disruption if for the first time you can start building these financial building blocks that aren't controlled by anybody. And really see the internet, back in the day is the decentralization of information, but the blockchain is really the decentralization of value. And all value is it's kind of like a fancy word for money.

“…trust minimized computing can be the most effective in is finance because trust is so important in finance. And in the traditional financial industry, it's just really run through mostly centralized intermediaries, so it’s very ripe for disruption…”

So I think that's kind of the fundamental reason why it makes sense to focus on financial products as certainly the first use case for decentralized applications.

CR: Totally. So can you explain a bit more about how dYdX actually works? We've seen this model of liquidity pools with automated market makers take off with Uniswap and then there's this more traditional model for exchanges which is order-book based but it's harder to make that model decentralized. So yeah, can you get into the backstage of dYdX? First, is it still non-noncustodial and what kind of framework do you use to match trades?

Hybrid Exchange

AJ: Yeah, good question and that's a good backdrop as well. So we at dYdX are what's known as a hybrid exchange, so that means that we have some centralized components which are our order book, and our matching engine services, and then some decentralized components, which are, of course, the smart contracts which run on the blockchain and make everything noncustodial.

So given that we run on an order book, we don't run on automated market makers, and kind of the reason for that is just order books are a lot more efficient in terms of being able to make the exchange have much more liquidity for the same amount of maker capital. Automated market makers are great, but they're very capital intensive. So you have to have, if you're Uniswap, or Sushiswap, or whatever, billions of dollars locked on the exchange to approach a similar level of liquidity as other exchanges that operate on order books have. So that's the way it works right now.

“We run on an order book, we don't run on automated market makers, and the reason for that is order books are a lot more efficient in terms of being able to make the exchange have more liquidity for the same amount of maker capital.”

And then I think the other piece that's really critical to dYdX and the main kind of differentiator for us over most other decentralized exchanges is that we're focused on more advanced financial products. So the current product that we're really focused on is a synthetic, which is known as a perpetual contract. Probably a lot of people in crypto are already familiar with this. It's basically financially the same product that was really popularized by Bitmex and it is now super popular across the space on exchanges like Binance and FTX.

But it's currently quite literally the most popular product in all of crypto by volume. Just in the past year, the volume of perpetuals has surpassed literally all the volumes for all the products including spot trading on crypto combined and it's just looking like it'll continue to grow from here. So that's kind of what got us really excited about offering these types of products, this kind of the potential addressable market. The reason they're so popular is because they can be traded with leverage, which basically just means you can multiply your gains and your losses when you're trading and trade as if you had more capital. So it's a much more capital-efficient way to trade.

“Just in the past year, the volume of perpetuals has surpassed literally all the volumes for all the products including spot trading on crypto combined and it's just looking like it'll continue to grow from here.”

So that’s the main reason that they're popular. On dYdX, of course, they're totally noncustodial, completely transparent. And those are the main reasons I think, some users are choosing to trade on dYdX over other platforms that offer perpetuals.

CR: Right. I think perpetuals was a very killer product in in crypto, because it very much simplified going long on different assets, I mean, this futures contract that doesn't really ever need to get paid, so you don't need to deal with that trading logistics and you can just hold it and go leverage long on something.

So this is hugely popular on centralized exchanges and BitMex is the biggest exchange by volume because it's the place that people go to trade these perpetual. But while you’re still among one of the biggest DEXs by volume, you're still lagging spot exchanges, like Uniswap or Sushiswap, why do you think that is in DeFi that spot trading is more popular than derivatives trading?

A Matter of Time for Derivatives to Rule

AJ: I think just has to do with kind of the maturation process of markets. And the DeFi markets are fairly separate from the CeFi markets, I think the same thing is starting to happen in DeFi that we saw in CeFi. Like in CeFi, back in the day when I was at Coinbase in 2015, Coinbase was more or less kind of the top exchange in the world. But one interesting thing that happened to Coinbase in 2016 is that BitFinex pretty rapidly became the most popular exchange in the world by volume. And the reason that was is because BitFinex offered margin trading, and people wanted to trade on leverage, and were using BitFinex to get margin to do that.

And then we saw just this past year, actually, so it's actually a really recent phenomenon in crypto, that like I said, the derivatives volume with like BitMex, Binance, FTX and friends, all just launching perpetuals and those becoming super popular within the past year. So I think it's something that takes time, just fundamentally, perpetuals are more complicated, both to build, to use, all of this, than spot trading is. So both for us, this is a bit of a generalization, but it's technically harder to build dYdX than it is to build a Uniswap or something like that. Just because these contracts that we're offering are fundamentally more complicated and you’ve got to deal with liquidations, you've got to deal with oracle prices, all that stuff. And that's what we've been working on for the past three and a half years. But it does take time.

“it's something that takes time, just fundamentally, perpetuals are more complicated, both to build, to use, all of this, than spot trading is.”

We're only just now about to launch our Layer 2 product, which we think is a really big step up for trading perpetuals. So I think it's just a time thing. In every market ever, basically, it starts with the asset and you start with spot trading, and then you start trading more advanced financial products on it. So it just doesn't really seem super likely to me that that's not going to be the case in DeFi, I think it just takes time.

CR: Makes sense. And so this maturation, for that to happen, what do you think is missing? Is it a matter of right now DeFi is mostly being led by retail traders, and then institutionals will come in and they will want these more complex products? So it is something that has to do with the user or also the product itself, like maybe there are features on dYdX that are missing to make dYdX beat spot trading?

AJ: Yeah, I definitely think it does start with the user. Derivatives are more complex than spot trading is, but also the types of traders that trade derivatives are much more professional. They're not all hedge funds, or crypto hedge funds and things like that, certainly, there are individuals that do it as well. But the product experience and the level of professionalism of the product, and a number of features like reliability for a derivative product is much more important to the users than just spot trading is.

So I think for that reason, if you want to make a product for perpetuals, or for advanced financial products, it has to be really professional, it has to be really performant, it has to have lots of features. And again, that just takes like more time to build than a very simple interface like Uniswap, that's just buying and selling a token. And that's not a knock on Uniswap at all, they have found amazing product-market fit for their target user, which are for the most part, and this is a generalization, but for the most part, it's just users who want to trade DeFi tokens and are relatively simple. But again, in terms of allowing the DeFi space to level up in terms of the sophistication of traders, just takes a little bit more time to make like these more complex products that they need.

CR: And speaking of users, can you dive a little bit deeper on what users you're seeing on dYdX? Like, where are they from? Are they hedge funds? Are they retail traders?

dYdX’s Users

AJ: Yeah, so I think most of them are still retail or individual traders. Certainly, there are some crypto hedge funds that trade and they're actually very important for the most part. Those are our market makers. We've, I think, done a pretty good job behind the scenes for our Layer 2 product launch in terms of just getting a lot more market makers and programmatic traders trading on the exchange, which is really important for liquidity and also volumes. But in crypto, it’s really different than traditional finance in crypto. In crypto, the funds really follow the individual traders not the other way around. Whereas in traditional finance, the retail traders just follow the funds or just trade directly with the funds on like Robinhood or things like that.

“In crypto, it’s really different than traditional finance in crypto. In crypto, the funds really follow the individual traders not the other way around. Whereas in traditional finance, the retail traders just follow the funds”

So the name of the game, if you're an exchange in crypto is making the best possible product for individual traders. And again, for us, that doesn't mean we want to market to somebody who literally just bought their first Bitcoin. It's we want to market towards the more sophisticated individual traders like I was talking about. And that's what we're focused on right now.

CR: And in terms of geographies, do you know where they are based?

AJ: Yeah, for the most part, it's international. Derivatives trading is really big in Asia. One of the big things we've been focusing on the growth side is expanding in Asia in a really big way. We just recently hired our first employee in China and they're bringing other people on board in other markets. So that's the main market we're focused on. But it's really global actually. It's pretty distributed, a lot of different countries in Europe, some countries South America, India, Russia, to a lesser extent, but it's pretty global.

CR: Cool. Because there is no U.S. trading, right? U.S. traders are locked from the site?

AJ: Yeah, that's correct.

CR: Okay. And that's for all products or is there some part of dYdX that's open for the U.S.?

AJ: Yeah, it's for our perpetual product. So all of Layer 2, basically, and the new product it is not available to U.S. customers, but margin trading, borrowing and lending, which are currently still operating on Layer 1 are still available to U.S. customers.

CR: Okay, got it. And I mean, I guess that's a regulatory-based decision, right?

AJ: Yeah.

CR: Okay, and then going to the Layer 2 announcement, what is this move going to mean for dYdX? How is the experience going to change for users?

dYdX’s Move to Layer 2

CR: Yeah. In a couple of really important ways, I think. It's a little hard to overstate the importance of this product launch for us. I mean, the first is pretty obvious, just the fees will go way down because users will no longer have to pay Ethereum gas fees. If you've ever used an Ethereum wallet, or DeFi product especially recently, you're probably used to paying pretty exorbitant fees and dYdX Layer 1 was no exception to that. I was oftentimes costing us like $100, literally up to like $1,000 sometimes, to execute a single trade.

And it's just really hard from a business perspective. Who's going to use an exchange where you have to pay hundreds of dollars in fees per trade? Probably not that many people. So that's the biggest advantage is now our fees, because the scalability is roughly 1,000 times better or so, so our fees will be much more in line with what users are used to on more centralized exchanges. But it doesn't stop there.

“The biggest advantage [to Layer 2] is now our fees, because the scalability is roughly 1,000 times better or so, so our fees will be much more in line with what users are used to on more centralized exchanges.”

There's a couple of other really important product benefits. And the other thing I want to stress is the product that we built for Layer 2, it's not just like we took the Layer 1 product and put it on Layer 2. We built an entirely new product from the ground up. It's still operating on perpetuals which we also had on Layer 1, but it's pretty significantly different product.

“It's not just like we took the Layer 1 product and put it on Layer 2. We built an entirely new product from the ground up.”

So the second big advantage is cross margining, which sounds like a fancy financial concept, but it's not that complicated actually. Basically, what it means is that now you can use on dYdX just one account or one pool of collateral to collateralize any different position you might want to trade on together. So say you wanted to trade like Bitcoin and ETH, you just deposit your however much into dYdX, and then immediately start trading all the assets you want to. Whereas on Layer 1, due to the scalability and some other concerns, we were only able to offer isolated margin.

So for example, same thing you want to trade Bitcoin and ETH on dYdX Layer 1 perpetuals, you would have had to deposit separately to every different market that you wanted to trade on. And that was especially capital inefficient for market makers. Like imagine if you're a market maker on dYdX, and there's only isolated margin and dYdX has like 50 markets, you have to put down collateral like 50 times to be able to market make on every market. And that was the main thing that was holding us back from launching a lot of new markets on dYdX. So we're really quickly going to be launching roughly 15 to 20 new markets after our public launch of the Layer 2 product. And it's really just driven by this cross margining.

“So we're really quickly going to be launching roughly 15 to 20 new markets after our public launch of the Layer 2 product. And it's really just driven by this cross margining.”

Some other advantages are the speed and performance of the user interface. So now when you make a trade on dYdX, it executes immediately, as fast as centralized exchange does. No longer do you have to make a trade and then wait for your transaction to get mined and see that dreaded spinner on the website for a few minutes while your transaction is sitting in the mempool.

And then the final improvements is, is also pretty important actually, and that's just a really big step up in terms of the performance of our price oracles. And basically, what we did, I think is pretty cool actually, we basically took the exact same price oracles which are reporting on Layer 1 Ethereum, like the top one. So we're partnering with Chainlink, and also MakerDAO to be able to do this. And we took the same prices, like the same security, same everything, but we're putting them on Layer 2 instead.

Higher Leverage

And the cool thing about that is now instead of just reporting a price every 15 minutes or so, we can report a price roughly every few seconds. And this is really important actually for being able to unlock higher leverage for us in a safe way. Because the thing that was holding us back on Layer 1 to offering higher leverage is the lag for price oracles.

So a lot of times it would take 5, 10, sometimes 15 minutes for a price change to be reflected on Layer 1 Ethereum, due to transaction costs and things like that. And that's kind of scary if you're a leveraged exchange, because then thinking you have to think about then is what's the maximum price drop of assets that you're supporting before the price oracle actually updates, and you can liquidate people? And that used to be 10 minutes, and now we've gotten it down to seconds, single digit seconds on our new product. And that's what really unlocks us being able to offer higher leverage safely.

“you have to think about then is what's kind of the maximum price drop of assets that you're supporting before the price oracle actually updates, and you can liquidate people? And that used to be 10 minutes, and now we've gotten it down to seconds, single digit seconds on our new product.”

So, starting with 25x leverage on the BTC and ETH markets for the new product, that's really important as well for our users. Binance put out this pretty interesting data about their users’ usage of leverage on Binance futures and kind of the average is around like 20x or so. So we're well able to support that with the 25x leverage that we have. There's certainly always going to be some, I think it's roughly 5% of users or so, will use the 100x leverages of the world, and we're not quite there yet. But I think we're able to support 90% of the leverages users will actually want to use.

CR: Will users have to create a separate account now to use this new dYdX product? Or will it work with their old, with the same Ethereum account? Is it a separate wallet on StarkWare, practically, how does it work?

AJ: Yeah, that's a great question. And this is actually one of the really cool things and one of the reasons why we chose StarkWare. It's basically a rollup that settles on the Layer 1 Ethereum chain, and that just means that users can continue to use their current Ethereum wallets, their current Ethereum addresses and everything like that. So there is basically a signup step. There's one transaction, you have to send basically to a smart contract, that's basically creating your account on Layer 2. But that's it basically. After you've sent that one-time, that one transaction, then you can just start depositing like you would to any normal exchange and start trading immediately. So it's a pretty seamless user experience, we think. It's something that, I think, is really unlocked by just building on a rollup that settles on Ethereum.

CR: And this transaction, how much does it cost? Is it like a regular Ethereum transaction?

AJ: Yeah, it's like a regular Ethereum transaction basically. I don't know if this means anything but it's the cost roughly 100,000 gas or so, 150,000 gas which is kind of in the ballpark of a Uniswap trade.

CR: Right. Okay, so this would be like kind of the bigger upfront costs and then trading on dYdX itself that would just be cents?

AJ: Exactly. Yeah.

CR: Okay. And why StarkWare on a ZK Rollup itself? There are at least a few different Layer 2 solutions. So why this specific one?

Why StarkWare and ZK Rollups

AJ: Yeah, absolutely. So the three main things we considered were obviously Zero Knowledge rollups with StarkWare. We considered Optimistic rollups and kind of Optimism is the leader there. And then also considered other Layer 1 chains in terms of why not other Layer 1 chains, it has to do a lot with we wanted something that was compatible with Ethereum and compatible with our users’ Ethereum wallets. And that's just critically important. In a lot of these other chains, they're not quite there yet in terms of being able to offer really great wallet experiences or cross-chain and things like that. So for that reason, we decided not to use a different Layer 1.

“…in terms of why not other Layer 1 chains, it has to do a lot with what I was just talking about we wanted something that was compatible with Ethereum and compatible with our users’ Ethereum wallets. And that's just critically important.”

In terms of why ZK Rollups over Optimistic Rollups, so it's for a couple of different reasons. First of all, we think the security and decentralization of Zero-Knowledge Rollups is better than Optimistic Rollups. Zero-Knowledge Rollups, basically, they use cryptography, this new concept called zero-knowledge proofs. And high level, the way that they work is you can take an arbitrary amount of data, so in our case, say that's like all the trades for the past hour, or whatever on dYdX, and then you can run a Zero-Knowledge proof for a stark-based proof on them and transform them down into this constant size data object.

And the cool thing about it is no matter how many trades you have in the batch, the data object is always the same size. And that's what's known as the proof. So you just put this proof on the chain, and then immediately the contract knows how to validate the cryptography around the proof, basically. And it's quite literally, cryptographically, impossible for an invalid state to be submitted to the chain, because of the Zero-Knowledge stuff. So that's really cool.

“And the cool thing about it is no matter how many trades you have in the batch, the data object is always the same size. And that's what's known as the proof. So you just like put this proof on the chain, and then immediately the contract knows how to validate the cryptography around the proof.”

Because one of the things about Optimistic Rollups is they rely on game theory to basically be able to secure the chain. And that means they have to have like a really long challenge period. So very high level way, Optimistic Rollups work is the operator can submit any state to the blockchain, even if it's invalid and then they rely on like other people watching the chain and then there's a way after that, if they submit an invalid state to be like, no, aha, here, I'm going to present to the smart contracts, the operator made this invalid state, so we're going to roll the state back.

And that's the way Optimistic Rollups work. So it relies on like incentives rather than cryptography, which we think we'd to continue to rely on cryptographic primitives as much as we can. And the other thing…

CR: Oh, sorry. Does that mean that the withdrawal periods for Optimistic Rollups would take longer if you need to wait for a challenge? I mean, does it make it just slower to trade?

AJ: Exactly. Exactly. That's actually the thing I was literally about to mention, but good thought on that. Yeah. And that's actually a really big piece of it. Because you have this challenge period, like you just said, on Optimistic Rollups, you have their really long withdrawal times where nobody can withdraw and usually, that’s what with these withdrawal periods are roughly a week or so. So that was a problem for us, because who's going to use an exchange where you have to wait like a week to withdraw? It's obviously a pretty big product limitation.

There are ways you can get around it. And the main way people are thinking about getting around it, both for Zero Knowledge Rollups actually, and Optimistic Rollups is using what's called like a liquidity provider that can basically front the withdraws, and can immediately pay you out. But the issue with that, with Optimistic Rollups, is the withdrawal period is still pretty important. Because you need, if suppose you're this liquidity provider, you need capital proportional to all the withdraws that everybody wanted to do on dYdX for the entire week, you'd have to have that much capital.

And for an exchange, that's so much money. Because people are moving their funds, in and out, all the time, it'd be like millions and millions and millions of dollars. We or any of our partners don't actually have that much money to take on this liquidity provider role. We're actually doing this, basically, what I just said, for Zero Knowledge Rollups, so you actually don't even have to wait on the new dYdX for a batch to be submitted, which happens roughly once an hour. A batch just means like that process of doing Zero-Knowledge proof and putting it on-chain, that will happen like once every hour, but you don't even have to wait for that for withdraws, because we're doing that thing where we operate the liquidity provider.

“And for an exchange, that's so much money. Because people are moving their funds, in and out, all the time, it'd be like millions and millions and millions of dollars. We or any of our partners don't actually have that much money to take on this liquidity provider role.”

But the cool thing about that is now the liquidity provider only needs capital proportional to an hour of withdrawals on dYdX, and that's much more tenable, right, that's probably like roughly like a million or so. And with reasonable fees, you can allocate that effectively. So, yeah, those are all the main considerations.

CR: And so in that case, are you, is dYdX the liquidity provider is it a third party?

AJ: Yeah, we're starting out doing it ourselves with our own capital, but we definitely hope to open it up to other partners really soon as well.

CR: Okay, got it. The other piece with Layer 2s is how well they can process smart contracts or if they even can right, I think that's the other big consideration. So obviously, if you're doing this on the StarkWare, with the ZK Rollups that's kind of a solution that does support smart contracts. I mean, like what solutions out there have this limitation?

AJ: Yeah, that's a great question. And I think one of the big selling points of Optimistic Rollups is that they're Solidity compatible. So theoretically, you can just take a smart contract that was running on Layer 1, and maybe with a small amount of modifications, get that up and running on Optimism really simply. It’s not that easy on Zero-Knowledge land right now. So basically, right now for our new smart contracts, StarkWare and this is the reason we're partnering with them, but they basically built all of the Zero-Knowledge stuff about our new protocol for us.

So that's what we've been working with them on for the past six months or so. So it is possible and you can build pretty complex stuff like perpetuals, and dYdX, even on Layer 1. Or some of the more complex smart contracts that are on the chain, and already, we're able to move them to Layer 2.

One of the things StarkWare is working on right now, which is, I think, just recently launched is they basically launched their own programming language called Cairo, which allows just any developer to write their own program that runs on Zero Knowledge proofs or Stark proofs, for the most part. We didn't use that for our stuff, because they just recently came out with that. But we’d probably use that if we were to build other things on top of StarkWare.

CR: Oh, that's interesting. Okay, so that's an interesting set of tradeoffs between the different Layer 2s. So with Optimism it's easier to build smart contracts and maybe develop complex applications there, but the withdrawal times are longer, and you don't have cryptographic security. But then with something like ZK Rollups, it's harder to build smart contract based applications, but you have the other benefits that you mentioned.

So I think with all the scaling issues that Ethereum is having now and it costs hundreds of dollars to do sometimes one single transaction, and it's becoming unusable for many individuals. The question is always why isn't everyone just going en masse to Layer 2? You gave a glimpse of that. It's just hard to build on these networks. But I'd love to hear more about it. Because I think everyone's kind of asking that question, when is everyone going to go to Layer 2 and all the gas fees will come down for us?

Why Isn’t Everyone Going to Layer 2

AJ: Yeah, it's a really great question. And hopefully, we can, dYdX, hopefully, be a trailblazer in terms of motivating people to move to Layer 2. Like I said, I think the product is just so much better. It's even beyond not having to pay $100 gas fees anymore to the point where people won't want to use products that are on Layer 1, if they're similar products that are available on Layer 2. And that's obviously a really big motivator to get developers to build things on Layer 2.

“It's even beyond not having to pay $100 gas fees anymore to the point where people won't want to use products that are on Layer 1, if they're similar products that are available on Layer 2.”

But in terms of why it hasn't happened en masse yet, I think it's a combination of things. I think, first of all, a lot of these scaling solutions are pretty nascent or they haven't really been tested in production yet.

That was one of the things that drew us to StarkWare actually, is that they have a history already of running in production. So they're running as of a year ago almost, I think, with another decentralized exchange called DeversiFi which has been running on StarkWare without basically any issues. So that gave us a lot of confidence that it would be production-ready.

Things like Optimism are super exciting, and I'm also excited about it, are not quite production-ready yet. I think, as far as I know, Optimism is running on mainnet right now, but has some limitations around that mainnet. Let's definitely keep an eye on what happens when they open that up.

Loss of Composability

In terms of other Layer 1 chains, yeah, a lot of them are on mainnet. That's just happened recently. Like I said, a lot of the developer tools, a lot of the wallet infrastructure, stuff like that, isn't quite there yet. So there's a ways to go on that. So I think that's mostly it and then I think the other thing people worry about sometimes which is a super valid worry with Layer 2 is the loss of composability on Layer 1 with Ethereum. I think it would be interesting to see how this plays out because it seems like a lot of different projects are moving to different Layer 2s. We're moving to StarkWare, Synthetix was moving to Optimism, I think like Curve or something is moving to Mirror. The list goes on. And I think that's definitely a real concern.

“a lot of different projects are moving to different Layer 2s. We’re moving to StarkWare, Synthetix was moving to Optimism, I think Curve or something is moving to Mirror. The list goes on.”

One of the reasons I think you see some of the synthetic protocols like us and Synthetix being real trailblazers in terms of moving to Layer 2, is that, at least in my opinion, interoperability is a little bit less important for synthetic protocols, because you don't need as many things to exist on Layer 1. For example, suppose Uniswap were to move to a Layer 2 as an example. You need, as Uniswap, all of the tokens that your users want to trade to also be on that Layer 2. Right? Who cares if Uniswap is on Layer 2 if none of the tokens you want to trade are on Layer 2? Whereas for us, because it's synthetic, we don't actually need tokens to be on these Layer 2s, we can basically just have only one collateral token, which in our case is USDC beyond that Layer 2, and then we can just create a synthetic contract for any asset based off of that.

“…interoperability is a little bit less important for synthetic protocols, because you don't need as many things to exist on Layer 1.”

And then the other thing I'd say on that point is we at dYdX take a really full-stack approach to building products. So we build everything from the smart contracts to the backend that runs our matching engine to the frontend, to eventually the mobile apps. So we really think by building everything vertically integrated, we can build a better product than what would otherwise be possible if it's a lot of different teams kind of working on the same thing with different levels of the stack.

CR: That's so interesting. I hadn't really thought about why the first two projects were synthetic or derivatives projects and it really does make sense. So with your move to ZK Rollups-based Layer 2, are you betting that most other DeFi projects will move there? Because like you said, it's about composability, right, that's the important piece for Layer 2 scaling to work for DeFi to maintain these money legos snapping together?

AJ: Yeah, I wouldn't say we're counting on it, it would certainly be good, like the more things that are on the same Layer 2 that we're using. But I think it just goes back to the point I was just making in terms of, we try to take a full stack approach, try to build as much things in-house as we can, basically, and not have to rely on like external people. That said, we do rely on some external integrations that actively make our product a lot better.

For example, one of the things we're doing on our current Layer 2 product right now is an integration with the 0x API, which is basically 0x’s like DEX aggregator that they have, to allow users to deposit any asset and have that automatically converted to the USDC that's held as collateral in our contracts, but make a really seamless deposit experience. So you go to dYdX, you could deposit ETH, automatically gets converted into USDC under the hood through 0x API, and that's all possible because the composability there. So I think it's definitely important, but all is not lost without it. I think a lot of the advantages of so much more scalability and the other stuff I was talking about before far outweigh some loss of composability for us, especially in the short term.

CR: Okay. So wait, I missed the part about the 0x API, will you be able to add that to Layer 2 or no?

AJ: Yes. So it's on Layer 1 actually, is how that works. So basically, when you're depositing funds to Layer 2, they're still on Layer 1 on Ethereum, right? So we basically built what's called a proxy contract, which is a contract that sits in front of the StarkWare smart contract, that will basically take your funds, call 0x API, convert them on Layer 1 Ethereum into USDC, and then deposit that to Layer 2. So it's still the same thing. All of that interaction doesn't actually happen on Layer 2, and you have to pay gas for it and stuff like that. But we definitely think it's still a pretty good usability improvement for our users. Because a lot of people will come with USDT or something, or ETH and we want to make it easy for them to use the product as well.

CR: Oh, that's interesting. Okay, so that means that for projects building on Layer 2, they can still use parts of their features and functionality and have them run on Layer 1, if they require other projects to do that. So you can move to Layer 2 without needing the entire Ethereum ecosystem to be there to maintain that same functionality.

AJ: Yeah. I think for the most part, that's true. You can basically use the stuff on Layer 1 Ethereum, especially when you're doing deposits to the system or withdrawals from the system. Because one of those kind of touches both Ethereum and StarkWare in our case. So you can definitely use them in those cases.

CR: Oh, that's interesting. And I wanted to talk to you about this cross-chain DeFi world. Because you mentioned briefly that you had considered going to other chains, but you decided that to you it was important to just stay on Ethereum because of the infrastructure. Still do you plan on moving to other chains when there is more infrastructure built and when you see there's kind of more of a community there or do you first plan to just focus on Ethereum for the time being?

AJ: Yeah, it's a good question. I guess the answer is we're not sure yet. I'm probably not in the immediate future we would migrate to anything else. One of the cool things about Layer 2 and especially the StarkWare system is that it actually doesn't really matter what the base chain is fundamentally, with a few changes the stuff we're building on StarkWare could say run on Mirror or Solana or something like that as the settlement chain and then you got to get the power of Solana times the power of StarkWare, which is even potentially more as an example. So you could definitely do that at some point, if we needed to, but I think it's probably not necessary yet.

“One of the cool things about Layer 2 and especially the StarkWare system is that it actually doesn't really matter what the base chain is fundamentally, with a few changes the stuff we're building on StarkWare could say run on Mirror or Solana.”

CR: And then I wanted to ask you about these other synthetic asset platforms like Mirror that kind of like grew really quickly recently because they added US stocks, token derivatives on the back of the whole WallStreetBets thing and Robinhood drama. Is that something that that you plan on doing, adding these traditional asset derivatives on dYdX?

AJ: Yeah. It’s we're definitely interested in it are definitely considering it for sometime in the future, I've actually been surprised by this, honestly, but I think just like in crypto, a lot of people have tried this over the years basically, just putting real stocks on crypto exchanges. I think Mirror is a good example of that. Maybe more mainstream example of that just the cool work FTX is doing with like listing like tokenized versions of stocks and things like that.

But if you look at the volume on those, they're never really that high, compared with the volume on Bitcoin or Ethereum, or things like that. And I think the reason for that, and especially if you're a decentralized exchange, and you know, as a decentralized exchange, basically all of our users are super crypto fluent, and super crypto native, right? Still, we want to improve on this, but still you got to know how to download MetaMask or imToken and backup your private keys and all this stuff. If you know how to do that, you probably understand crypto and you probably want to trade like Bitcoin and ETH more than you want to trade Apple or something like that.

“You got to know how to download MetaMask or imToken and backup your private keys and all this stuff. If you know how to do that, you probably understand crypto and you probably want to trade like Bitcoin and ETH more than you want to trade Apple.”

So I think for the most part, that's my hypothesis on why they haven't taken off yet. Because even for centralized crypto exchanges, even if you're an FTX or a Binance or something like that, it's a more challenging user experience to get into the crypto ecosystem than it is to start using Robinhood or something.

So yeah, definitely keep an eye on it, is the TLDR. But we kind of want to see some more volumes hitting some more other exchanges that offer like these types of products before we get there.

CR: Yeah, that's an interesting distinction. It's like people who are already trading on DEXs, they probably just want to trade crypto and not traditional stocks, anyways. Okay. And then I want to just talk about DeFi more broadly your thoughts on this space, the incredible growth it's seen so far. How has it surprised you how TVL went from a billion early last year to over 40 billion today, ETH going to 2,000? This crazy explosion in crypto and DeFi, do you see this speed of growth continuing? And what is the long-term vision for dYdX in the context of DeFi? Do you see DeFi just overtaking traditional finance and dYdX leading the charge there? Or do you see it as something that will continue running in parallel to Wall Street?

The Future of DeFi

AJ: Yeah, definitely. So first of all, in terms of the level of growth we've seen in DeFi and just more broadly, like crypto space, it certainly surprised me. I would imagine it surprised pretty much everybody. And really the catalyst, I think especially for DeFi was Compound launching COMP, which seemed like a little bit of an innocuous thing. If you remember where we were back in last summer, and that was still kind of in the crypto winter and Compound was launching their token. And the thing that they invented, which was really awesome is liquidity mining. And they weren't the first to do it, they're actually some other centralized exchanges. There's one called FCoin and a few others, which did something similar to liquidity mining. But it really caught on to DeFi, which I think was surprising and exciting to a lot of people.

And then fast forward to a lot of the adoption we started to see with Uniswap, I think just Uniswap, finding product-market fit was a really watershed moment for the DeFi space. Just in terms of great products period, whether it's centralized or decentralized, just go to this really simple page, just trade any token for any token, has way more tokens than all the centralized exchanges do. They're listed way faster. The liquidity is pretty good as well. And then a lot of the stuff they added on top with the governance, I think was super cool as well.

So lots of really interesting stuff, I mean, I think it sometimes just happens like this, where it feels almost out of the blue that these things happen. Because for us and Compound, and to a certain extent Uniswap as well, we've kind of been around for a couple of years now. And then sometimes it just turns the corner and you turn on the jets for this space, and everybody starts using your products, which is awesome.

“…for us and Compound, and to a certain extent Uniswap as well, we've kind of been around for a couple of years now. And then sometimes it just turns the corner and you turn on the jets for this space…”

Yeah, so then to answer your second question in terms of how do I see dYdX playing in terms of DeFi, what are our goals more broadly? Our goal is pretty big, our goal is to become one of the biggest crypto exchanges, period, but on a three to five-year time horizon. I think it's still going to take some time, DeFi isn’t going to be built in the day or even a year. And I think there's still a lot of work for us to do in terms of continuing to optimize the user experience, continuing to decentralize more parts of our stack that we're really going to focus on over the next year or two. And it just takes time for two-sided marketplaces to be built up in terms of liquidity and stuff like that.

“Our goal is pretty big, our goal is to become one of the biggest crypto exchanges, period, but on a three to five-year time horizon. I think it's still going to take some time, DeFi isn’t going to be built in the day or even a year.”

But I think, obviously, I'm biased on this, but I think we're in a pretty good spot. Like, in the thick of derivatives, as I kind of touched on before, we think derivatives are pretty, obviously already are, and are going to continue to be, the biggest products in crypto. So, if your goal is to become one of the biggest crypto exchanges, basically, by definition, have to support the biggest product in crypto, and we feel like we're doing that. And then obviously the other bet that we're making is that DeFi is going to be really big on that time horizon and hopefully, we can continue to help push forward a lot of the cool things that are going on in this space more broadly there too.

And I mean, DeFi is super exciting. But still remember, it's like, 5%, or whatever of the volume that's going through all of the other exchanges, all the other centralized exchanges in crypto. But again, that's up from 0%, two years ago when this stuff like literally wasn't invented. So I think a lot of people can see the writing on the wall, and that's why people are excited about it. And we're excited as well.

“So I think a lot of people can see the writing on the wall, and that's why people are excited about it. And we're excited as well.”

So lots of optimism for the future. The one caveat I will give is, I feel like especially in crypto, in all new markets, but I think especially crypto because there's a price tag on everything that you can literally always see, a lot of the innovation or a lot of the hype seems to come in booms and then consequently busts. So I would anticipate that that's certainly going to happen at some point in the future. I mean, it's not really rocket science, look what's going on with some of the valuations of random tokens and things like that. And there probably be corrections and things like that.

But the important thing is that that the teams in the space that have been here and have been building are going to continue to keep building. It's not like the progress of dYdX moving towards full decentralization, or our next product or whatever is going to significantly stall if there were a correction or something like that. And then every time there's a really big hype cycle people wake up and they're like, oh, wow, this stuff is really cool. Yeah, it's almost like you put two years of effort into this for Uniswap or whatever. It's like, yeah, we did, and nobody was paying attention.

“…every time there's a really big hype cycle people wake up and they're like, oh, wow, this stuff is really cool. Yeah, it's almost like you put two years of effort into this for Uniswap or whatever. It's like, yeah, we did, and nobody was paying attention.”

So it's nice when people are pretty hyped on it as well in a similar way. But also, the periods where you can just like heads down, build are pretty nice too and you just have to have faith that long term, the market’s going to be there and you're going to be able to capitalize on that.

CR: Yeah. Well, I think like most of DeFi was built on a bear market. So I think, if there's another one, no reason to think why development will stop then. So, last question, you mentioned that you want to progressively decentralize dYdX, how do you plan to do that?

Further Decentralizing dYdX

AJ: Yeah, good question. So honestly, we're still in the architecture phases of this, we’re still trying to figure it out. And that's one of the main things we're going to be focusing on, especially over the summer. For us, as I touched on before, the really big critical piece that we need to decentralize is the order books and the order matching. And we run those in a centralized way right now for performance. And we'll continue to do that on our Layer 2 product for now.

So figuring out different ways to potentially optimize or decentralize the order books, that'll be a really big project for us. Also, decentralizing the control of our smart contracts. So we still operate under the like multisig module or paradigm, where basically our contracts are upgradable by a multisig contract, which is delayed by a contract enforced time lock. But it's not as decentralized certainly as a different protocol that may have decentralized governance. So potentially focusing on decentralized governance like we've seen a lot of other protocols do, involving community more in decision making things like that are all things we have slotted for the rest of the year.

CR: So I mean, that would imply a dYdX token?

AJ: Yeah, I mean, we're definitely thinking about it. Again we haven't announced anything around that or anything like that. But we think it would be value add if we were to do something like that, in terms of allowing the community to get more involved in governance and things like that. I mean, it doesn't have to, like we saw Compound’s, even before their COMP token, they were doing a lot of interesting, innovative stuff with governance.

But I mean, we're not blind basically. We see what's going on with all the liquidity mining and with all of the governance and exciting things with the community that are going on with other projects. For us, our hypothesis around it is, we're not fundamentally anti-token or anything like that. We just really think that the way you get really big is, first of all, you build an amazing product and you find strong product-market fit, and you start growing without a token. And then once you do a token or are able to throw the fuel on that fire, once you have a product you believe in and that can scale and that you think is high quality and can grow, and then you like I said, just throw the fuel on the fire, that's how we think we’ll become really big.

“We're not fundamentally anti-token or anything like that.We just really think that the way you get really big is, first of all, you build an amazing product.”

CR: Pretty much what Compound and Uniswap did. They started without a token and they've been successful with one. Okay, so that'd be a really exciting and interesting development. We’ll obviously be keeping track of that on The Defiant. So anyways, we're at the end of the hour and it's been such an interesting conversation. Antonio, thank you so much for joining me.

AJ: Yeah. Thanks so much for having me and the great questions, really appreciate it.

Share The Defiant

The Defiant is a daily newsletter focusing on decentralized finance, a new financial system that’s being built on top of open blockchains. The space is evolving at breakneck speed and revolutionizing tech and money. Spread the word and share!