This transcript has been edited and summarized for clarity.
Aditya Palepu, co-founder of DerivaDEX leads a discussion with investors Haseeb Qureshi, Dan Matuszewski, and Niraj Pant. Topics covered include participant’s DeFi thesis right now, AMMs vs open order books, and what is (and isn’t) working with governance tokens. We also discuss the design of DerivaDEX’s insurance mining program, and the unique role an insurance fund plays in a decentralized derivatives exchange.
Prefer a podcast?
Introducing the participants
Aditya Palepu: I’ve been an institutional algorithmic trader at one of the leading prop shops in the world, DRW. The past couple of years, I was working at Enigma as a blockchain engineer. Most recently this year I’ve been focused on building out DerivaDEX to be one of the preeminent exchanges on the market.
Haseeb Qureshi: I’m a managing partner at Dragonfly Capital. We’re a crypto venture fund. We’ve been early backers of DerivaDEX and I’ve known Aditya for a few years now.
Dan Matuszewski: I’m Dan, part of CMS holdings. CMS holdings was formed with two guys Bobby Cho, formerly of Cumberland, and, DRW, which Aditya had mentioned he was at, and Julian, who was the CTO of circle trade. Previously I ran Circle’s OTC and liquidity provisioning operations. We know the derivatives side as it stands currently pretty well, obviously pretty excited about the next generation of stuff that we’re seeing come out, which is where DerivaDEX comes in.
Niraj Pant: I’m a partner at Polychain where I focus on early ventures, everything from exchanges and more traditional business models to more sci-fi stuff like smart contracts, which we’re going to be talking a lot about today. Before Polychain I did a lot of work in privacy, particularly on the research side and did R&D backend engineering at a bunch of different startups in the Bay area.
Introduction to DerivaDEX
Aditya- We’re very excited to have a complimentary yet orthogonal group of people that are helping contribute to the vision here. Before I kick things off with a broader discussion on DeFi. think it’s first helpful to those of you who might not be familiar with what we’re doing at DerivaDEX to just outline our initial motivation.
Traders are really looking to speculate and hedge with a wide array of underlying assets in a secure censorship resistant and efficient manner. It sounds very obvious that you should build a platform that is secure and usable, but it turns out in practice, we haven’t really seen someone effectively accomplish both to our estimation. So, traders by and large have two options, right?
You can either opt for a centralized exchange where you have good liquidity, good participation, good UX, but is very weak when it comes to security of funds, and very burdensome to even participate. Or on the other hand, you have decentralized exchanges that paint the opposite story, right? So you have, theoretically at least strong security of funds, low barrier to entry, but very weak when it comes to liquidity, unforgivable UX, very low participation.
As a trader myself, this doesn’t sit well with me. This is not how it should be. And so first and foremost, this is what we’re set out to solve. It’s our thesis that there is a very significant and exciting opportunity to build the right kind of platform in this space.
Part I. “What is your DeFi thesis?”
Dan- I was definitely more skeptical of the DeFi stuff 18 months ago. I think it really started to change for me when the liquidity got to a point beyond where I thought it was going to be able to get. There’s certain activity inside DeFi, that is becoming better than activity you’d be able to do in a centralized fashion.
Really, there’s price and liquidity. That’s not to say the security you get with DeFi doesn’t matter. People very much care. But really the liquidity, it was the big issue.
You’re seeing this really definitely showing up in the lending protocols that are happening on chain. Uniswap, was like, all right, we’re just going to list everything and make a way for everybody to provide liquidity. And it’s great because you now have liquidity day one, and sometimes there’s a bad side of that too. There’s no safeguard whatsoever. So you have a ton of outright scams but you can’t have one without the other. Unfortunately, everything is going to get listed. You’re going to get the bad with the good, but that’s a really good way to like bootstrap stuff day one, have liquidity, have a market, get it going.
But the problem is, centralized execution for Bitcoin and Ether is almost free now. And beyond that, there’s margining that is sometimes hidden to the average user that people are taking advantage of on the exchanges that DeFi is not going to be able to compete with yet.
Haseeb- Globalization of crypto is a big story for what we see as the progression of DeFi and the role that it’s gonna increasingly play. Obviously the big story, as of this month was the indictment of BitMEX, sort of the death knell for the centralized exchange free-for-all that traditionally the perpetual swaps have been.
If you zoom out, what’s happened with DeFi is really a case of convergent evolution. If you think about what was most attractive in centralized exchanges, three, four years ago, it was almost all spot markets that had moved to margin trading. And it’s now moved into derivatives. If you look at the progression of DeFi, it is mirroring the same evolution. It was all spots trading, very, very small volumes in the early days. Then eventually it moved into being margin trading, where DyDx had kind of come to the forefront. Now today, I think you could argue we’re still in that margin trading phase where arguably of course, Uniswap is dominating volumes, but a lot of that is actually coming up from people margining, by levering up through doing liquidity mining shenanigans or levering up on Compound.
We really don’t have a robust DeFi-based on-chain perpetuals market yet, or derivatives market. Primarily the barriers are technological. The one thing that you can always bet on is that technology only ever gets better, and barriers to implementing the thing you want to implement, just keep getting lower and lower and lower until at some point you just breeze right past them. And that I think is part of the reason why we’re so excited about what DerivaDEX is working on.
Niraj- We’re really excited about new use cases. What sort of new behaviors are enabled by DeFi and by crypto broadly? We’re seeing a lot of really exciting stuff happening in the market. A lot of cross chain interactions, a bunch of BTC is on Ethereum for the first time. LP shares are this primitive for people to use in their DeFi applications. DeFi is kind of an industry now. There are funds based around it. People trade it actively, both retail and institutional and that’ll continue to grow. I think DerivaDEX is enabling people to trade out of this utopian financial system where you just have one wallet address and you can go trade everywhere. And I think while the tech is still early, there’s a number of really strong benefits to what’s happening in DeFi. I think that there’s a lot of exciting stuff happening and I’m really excited about what DerivaDEX is doing.
Part II: AMMs vs Open order books
Aditya — Here at DerivaDEX, we’re pretty pro the open order book model. What are your thoughts on that dichotomy, order book versus AMM?
Haseeb — It’s been fairly clear that especially with gas prices becoming as onerous as they are today, order books have really, they really failed relative to AMM model. So the AMMS are now completely dominant when it comes to DeFi trading. And part of the reason why AMMS have worked so well is that they’re incredibly simple and they’re very very gas optimized, and that allows them to basically be trading in real time without having to deal with all the complexities of an order book, which requires cancellations, which requires trade collisions and all sorts of complicated stuff that generally speaking leads to a not so pleasant user experience.
The other thing about the order book model is that with AMMs, people are trading assets that are relatively low volatility. They’re trading stable coins. ETH for example, is relatively low volatility, at least compared to where it was two, three years ago, and that allows the AMM model to really work. When the underlying isn’t moving around too much, a really dumb constant product algorithm can do a pretty okay job.
However, if you’re thinking about a world where you’re trading very highly leveraged positions, then suddenly, the error rate in pricing and in execution, really, really compounds when you increase the leverage rate. And that’s one of the reasons why, and the other problem of course, is liquidations are not straightforward to do on-chain at this point. And so, because of that, it’s very difficult for people to do leveraged trading in DeFi.
And it’s very difficult for them to price volatile assets, especially leveraged assets. Once you start getting into much more sensitive types of trading, you really need order books, having a very simplistic AMMs doesn’t really scale to these types of markets.
Dan- I find it funny that this is like such a lightning rod, at this exact moment in history. I obviously fall on the order book world on this, because I think you can generally recreate the AMM liquidity provisioning model on top of an order book fairly simply. Absent, costs and gas and tech-related stuff, in a world where you don’t have to think about that, I refuse to believe that the AMM is the dominant model over an order book that you could functionally make the exact same thing as the AMM, if you wanted to. Like you could create that full thing, plus all the additional features like placing orders or prices and things of that nature.
AMMS are super useful because they’re quick, they’re easy and you can get going with almost no barrier to entry. What I really think people got to take from that is you need to make that version, that simplistic view of it and slap it on your order book thing.
Uniswap just gives you a very simple interface. That being said, simple for DeFi. It’s still complicated. But from a market’s perspective, it’s incredibly useful to people that don’t want to know the intricacies of what’s going on behind the scenes. But anyway, like I said, at the end of the day, I think the order book model is all encompassing and the right sort of way to support going forward.
But man, is it a polarizing thing to get involved in that fight for some reason. And, it’s very odd to me that people have drawn lines in the sand of what side they sit on. It’s just, it’s very weird to me
Niraj- I just don’t think we’ve seen a really strong offering of an order book model for a DEX yet. Obviously order books are extremely strong in centralized venues. There’s still a massive delta between centralized order books and decentralized ones. I think Uniswap has enabled a lot, but that that model is not applicable to everything. And so if the tech is there to enable professional market makers, of which the category is growing a lot within DeFi, I see that this open [order book] model being really, really strong, especially for a derivative, where the AMM model doesn’t really scale as well in this case.
Aditya — As I stated at the outset, we’re very favorable and pro the order book model. AMMS, have and will continue to evolve and it will be exciting to see how that progression takes place. In their present construction, the way I see it is that they’re fairly fragile systems. Inherent to their design, largely they don’t work too well in two very important scenarios. One is large order sizes and volatile markets, and those are two hallmarks of the crypto sphere. We’re excited for DerivaDEX to create a order book experience that safely supports liquidity providers and market takers, both institutional and retail sizes, in a way that gives them the capital efficiency and the familiar experience that they expect.
Part III: Governance and token models: What’s working? Whats not?
Niraj- I think we’ve gone through some really rapid iteration in the last six months. I feel like we’ve been talking about DAOs for a really long time. It’s gotten a lot better, stuff is happening a lot faster. I think what’s working is that people are using it. Governance is actually an important part of these networks, especially in this context now of this capital formation innovation, which is paying out users for participating in a network, whether it’s providing liquidity or it’s trading or it’s providing cover on the insurance fund, which I know we’ll talk about later. I think there still is a lot of work to do on how we actually scale these systems and like what the future looks like for governance. I think a big flaw of a lot of these DeFi platforms is this negative feedback cycle where there are professionals that are just kind of mining the shit out of these projects. They’re not really adding anything to the protocol other than this like liquidity that’ll be there for a little bit, but it’ll be taken out as soon as they get the coins. And it’s totally rational for them to do that, you know, in a trading sense, but, net to the community, it’s not the best long-term incentive. So I’d love to see more models around, you know, what is it, what does a carrot or stick model look like for you know, incentivizing people to lock coins in or keep rewards for longer time and have incentives more aligned.
Dan- Look, I’d love to see a world where owning a chunk of the governance, really mattered. I think there’s a lot of ways you can extract value as a trader from owning a significant chunk of the governance, if the governance really matters. The issue that we see with this so far is that a lot of this is “in name only” kind of stuff, right? If there’s like a massive chunk of it, that’s just owned by a certain number of groups that can just dictate the fate anyway, then what’s the point if you own some chunk of the governance outside of that. So that immediately kills that plan. I think about this in like the evil sense immediately. If you could like own enough of this thing, how could you like wield this thing in your will to make money for you? That being said, like, I think it’s moving in a better direction. I do believe that there a real edge in being able to like have a governance token that is strictly tied in the sense that like, what gets proposed is what like actually happens via a real consensus amongst the community of people that are operating with the system. I think that’d be really useful and it will accrue value.
I don’t see it yet where like anybody’s tied all that stuff together. I’m a buyer of the thesis that this will get better. I think it’s going to be a thing, going forward, people love it, right? This is Compound did this what, eight months ago now. It’s basically been just off to the races ever since. So there’s like a clear market demand for it.
Aditya — Haseeb, I know you’ve also written a lot about liquidity mining and what users are ultimately getting their hands on this token and how that may not be ideal. Would love to hear your thoughts on this.
Haseeb- When it comes to liquidity mining, I think it’s another hot button topic. I think people have very different opinions about how effective it actually is. I think, as long as you understand what you’re buying with liquidity mining, then it is very sensible. If you think of liquidity, mining is going to magically win you long-term adoption and give you sticky users and give you true believers, it’s just the wrong tool for that. As much as it’s ever been, the only way to win long-term users and to really get a stickiness is to build a great product and build a great community and get people excited and animated about what you’re doing.
Liquidity mining is to my mind, very, very analogous to like having some kind of very attractive referral program or a bonus program or subsidizing the upfront cost of using your protocol. Once the price has gone back up, will they stay? It really depends. If you don’t give them an experience and a community that they love and want to be a part of, or a product that they are really enticed by, or that actually is something differentiating it for the rest of the world, then yeah, they’re going to go elsewhere.
Aditya — Our DAO, what we call it DerivaDAO is particularly powerful as well, because it allows users to participate in all aspects of the exchange. So this would be in controlling the fee schedules, the products that are listed, the roles that other stake holders and participants play such as who are the whitelisted set of operators, the matching engines of liquidation agents. It’s very empowering for every trader and user that’s participating. With respect to the token, liquidity mining doesn’t just solve for a deficient product. It doesn’t just magically make all of your problems go away. But combined with a strong product, a strong token model can actually keep the flywheel going.
Part IV: DerivaDEX Insurance mining
Aditya- For those of you who might be unfamiliar, insurance funds exist are common to centralized exchanges, and are necessary when it comes to leveraged exchanges. They serve as a protection against adverse liquidation scenarios in order to cover losses and pay out to traders. If you see insurance funds like a BitMex or FTX, these are some of the largest whales on the market. They’re very overly capitalized opaque processes.
The DerivaDEX insurance fund is a very new, radically transparent and different model. In addition to liquidation spreads being added to the fund, we are also capitalizing it by way of fees and also users can stake directly into the fund. The fee concept is very different from typical exchanges. All of that goes towards this decentralized insurance fund.
The insurance mining program, where users stake to the fund, is going to last a year. Out of the gate, users can participate by staking or depositing any number of supported collateral types upon launch. This is going to be USDT , USDC, and then the Compound and Aave C and A token variants of those tokens that I mentioned, in addition to HUSD. Users receive DDX continuously for staking into the fund, and this DDX is realized to their on-chain exchange wallets. They can also participate in governance without withdrawing or transferring these DDX tokens to their own off-chain personal wallet. It’s worth noting that the exchange isn’t live right out of the gate. So there’s actually going to be a period of time at which there’s actually no liquidation risk for users participating. When the exchange goes live and the network goes live, then, as I mentioned, liquidations spreads and feeds are going to be added to the insurance fund as a buffer to protect people that are staking into the fund and bootstrapping it upon network launch.
We’ve shifted the paradigm here so that it goes well beyond covering liquidation losses, but really does promote the full exchange experience for all, all stakeholders and traders involved.
Niraj- I think it’s really interesting to see this kind of system constructed from top to bottom, it’s not just the exchange piece, it’s also the insurance fund piece. You know, users can actually have a piece in this network, which I think is really interesting. In centralized exchanges, you’ve never been able to have a claim on the insurance fund. This to me is exciting because I don’t think we’ve really seen anything in the derivative space yet that enables a DAO like this at scale. I’m really looking forward to the launch of the project.
Haseeb — I largely agree with Niraj in that I like the idea of giving people direct ownership over the insurance fund. One of the big complaints that you hear, again and again, from people who are dealing with some of these exchanges that just have massively, massively overfunded insurance funds is like, well, what is this thing? [Derivadex] has a more transparent and more equitable way to think about what the role of the insurance fund is how each part of that protocol should be packaged together and owned by its users. I think what you guys are doing is super innovative and I think there’ll be a lot of excitement for it.
Dan — There’s two worlds, right? There’s like, OKEX and Huobi’s, early days where they were just wrong on the side of being too shy with the insurance fund, and people are taking massive haircuts. And then there’s BitMex on the other side, which is way too far in the other direction where it’s just accruing constantly.
In DerivaDEX’s model, you’re letting the market figure out the insurance fund strategy. And I think that’s a much better solution. Users can’t just be passive towards governance, if they’re going to keep getting haircuts because they’re not paying attention to the real risk being set on the liquidity parameters for the insurance fund. I think this is a great system to get that model right, in the sense that it’s not going to be overtaxing, it probably won’t be under taxing. Who knows how it’s going to work out of the gate, but over a long time I’m pretty bullish that this equilibrium is a much better sort of system for that liquidation risk.
Aditya — We are also really excited to see how this plays out when it goes live. This has been an awesome discussion. Thank you all for participating. We’re building this to be a very community driven exchange, and the community will be a crucial part of the DerivaDEX and DerivaDEX story from this point forward. Thank you all again.
