On The Brink with Castle Island - Uddhav Marwaha (Friktion Labs) on Building Durable Portfolios through DeFi (EP.340)

Episode Date: August 15, 2022

In this episode, Uddhav Marwaha, co-founder and CEO of Friktion Labs, joins us to discuss Friktion's approach to a full stack portfolio management platform in defi. We cover a plethora of topics, including: The opacity surrounding defi yields and distinguishing fleeting opportunities from sustainable sources of income Using options overwriting and basis strategies to generate more durable returns and provide portfolio protection across market conditions In-housing core tooling such as Friktion's RFQ system (Channel RFQ) and an exotic options DEX (Entropy) to unlock time and cost efficiencies   Friktion's emphasis on demystifying complex products and deconstructing sources of risk and yield in crypto differentiate it from the pack. To complement the data and analytics within the application, the team frequently publishes research breaking down complex topics.  

Welcome back to On the Brink. In this episode, Udav, co-founder and CEO of Friction, a SIV portfolio company, joins us to discuss his vision to build out a full-stack Defi-native portfolio management platform. We cover the different ways to generate yield in defy and the lack of clarity around where yields come from and the associated risks of the different sources of yield. We also talk about how friction has leveraged options overriding and basis strategies to create more sustainable forms of yield that together can help protect portfolios across different market conditions. We get into how friction is expanding to other end user groups, notably DAWs and more traditional institutions and product types such as under collateralized loans to offer a more comprehensive product suite. We also touch on the team's decision to build a native RFQ system with an auction mechanism, which provides a faster and more sustainable way to bid on, provide liquidity for, and price options on chain. And we cover a lot more in-depth color on options, derivatives, and portfolio and risk management in DFI more broadly. So without further ado, here is our conversation with Udav.

Hey, everyone. Welcome back to another episode of On the Brink. I'm Ria and today I have the pleasure of speaking to Udav, who is the co-founder and CEO of Friction Labs, which is building a full-stack portfolio and risk management protocol on Solana and is also a Siv portfolio company. Udav, thanks so much for joining us today. To kick it off, let's start with your background. What were you doing pre-Crypto and what was your journey into crypto? Hey, Rhea, thanks for having me. I'm excited to be here.

and chat today. Before a crypto, well, I guess I can give a bit of my background. So I grew up in India and then I moved to the States when I was like 12. So I've been here about half my life now. And early on at university, I was, I studied Petroa, engineering, national security and math. So not really like a traditional, I would say, kind of CS crypto upbringing to that extent. I mean, I really studied how to get flow from hydrocarbons in rocks out of the ground and then how to be creative about it. I spent like two summers working in an oil field in South Texas and West Texas and kind of two of the hottest basins at the time. And then I realized I was at that point I realized I was pretty obsessed with like geopolitics and statecraft from a perspective of being able to build

mental models of the world and really use that to forecast events that may happen in the future. And then as I got more and more obsessed with geopolitics, geopolitics, I realized that it basically drives and shapes behaviors of nations and their decisions to make, to kind of work with each other or not work with it over time, and then started working with this kind of geopolitical risk firm whose really question was like, how can we forecast these risk events ahead of time and then also kind of inform people who are the key stakeholders of those events. So the guy that I worked for kind of emphasized that geopolitics is really just a kind of question of people in place. Place defines the kind of actions, fears, and then the limits of what people can do. So both their kind of

capabilities and vulnerabilities, i.e. like the difference between living in China versus the U.S. versus Nigeria, and then people being what creates communities, which create nationalism. And nationalism is kind of like the modern form of love for the things you were born to, i.e. your country, which ends up being like a large driving force behind humanity. So if you just study like a couple of dynastic cycles, you'll find that like you can basically predict many events based on where people either are or what those communities look like. So what does this have to do with me and crypto? Well, I found this so exciting and then realized in like April of 2017 that I needed to apply some sort of like dimensional reduction to this view of the world that I had and particularly got sucked into markets because I realized that they are a very efficient way when designed well. went open and truly open to be able to price an event. And I think the older book is just something

that's still quite impressive to me as an innovation in the world. So as like a junior in school, I joined this kind of small but heavy hitting quant trading group at UT, the University of Texas, a bunch of options, market makers and traders, quant researchers, and then a couple of banking folks. And then I came in with more of a fundamental tilt. So I was still like kind of super obsessed with geopolitics. I spent my time studying like regions of the world using open source Intel tools to track and forecast events. My specific areas were Nigeria, Venezuela, Iran, Colombia, Saudi. And I would focus on being able to kind of being like the kind of subject matter expert in those regions and really be the one to identify how risk would play out there.

So as I was saying earlier, I needed a way to reduce the dimensions of all of this into something that was like easy to, easy to understand and easy to kind of make something out of. and quickly realized that I wanted to be in a path to taking risk, which kind of is the definition of a trader. And I wanted to find a desk which to work on that had some meaningful amount of edge. So there's many forms of edge in trading. The edge that I was particularly enamored by was that of a physical market maker or physical trader, meaning someone who has assets to trade around and you can use those physical assets like a oil field, a refinery, a pipeline for some embedded optionality.

And as I was studying hydrocarbons, it kind of made sense to go into commodities. And then with the Geopal background as well, was really exciting. So after college, I joined BP, which was a very much fundamental physical-based oil trading shop that had a... I viewed their edge as kind of fundamental information asymmetry, meaning they can predict supply and demand by using the assets they have and the trading flows that they get. So on the supply side, we were able to basically track barrel movements, across the world on land and water, whether this is like an oil product, a refined product, or a carbon product,

and then have a really deep historical understanding of supply events and what caused those events. And then what the effects of some of those events were had a literally like infrared camera on every major refinery unit in the US and kind of some of the major ones across the world to track pipelines to basically be the first ones to get information about changes in kind of market supply

and dynamics. And then on the demand side, like right as COVID was hitting, I was about a year into the job. We spent a bunch of time analyzing transportation activity for every city in the U.S. to be able to forecast gasoline consumption. I was sitting on the gasoline desk at the time. And we found that was super useful stuff. Got pretty obsessed with the idea of being able to use markets as a voting mechanism.

And then, yeah, that was kind of like post-college. I had gotten sucked into crypto in like April of 2017 when I bought my first Bitcoin. And since then, I realized there. There's better ways to make markets that don't really exist in defy today, particularly with the idea of creating risk management. That's something that I hadn't seen anyone tried to tackle. And I thought it was a perfect place to build. So that's a kind of long-winded story of how we got to friction.

It really amazes me the diversity of talent that we have in the crypto space because I think leveraging a more diverse background and set of talent just gives. rise to more vibrant and well-thought-out products. So that's super interesting. Thanks for delving into that. You kind of hinted at what my next question was going to be. And I'd love for you to expand on that, which is, you know, as you surveyed the defy ecosystem, what gaps did you notice that kind of inspired you to build friction? And yeah, what areas did you see friction feeling over time? Yeah, so I think the point of Defi that really got me hooked was right before Defi summer in 2020 when we started to see just like new forms of yield generation. And yield is blatantly overused and misused term in crypto, which hopefully we can chat a little bit out that later.

And the thing that I saw was there's kind of five ways to make returns. I'm just going to call them returns instead of yield. So you can lend into an over-collateralize lending platform, which has not really most of the capital efficiency that you would expect. from being able to lend in like an open money market. You can liquid stake to validate a network and kind of earn some rewards for that, which a lot of people compare that to kind of the risk-free rate of crypto, but that inherently has some of its own risks as well. You can provide liquidity into automated market makers,

which was definitely the rave during DeFi summer after a kind of compound kickoff, which is a lending platform. But you can start to earn these liquidity mining incentives, which are essentially a form of customer acquisition in order to bootstrap some early stage user growth for a platform by distributing, by that platform distributing some native token. And then as DFI summer matured, we started to see kind of the rise of things like derivatives arbitrages and volatility trading and using options overriding strategies or basis strategies

to generate some market neutral or some lower delta risk than you would expect from any traditional strategies. And the big thing that got me during all of this was there's no way to do this in kind of one place that both measures your risk across every strategy you deploy a risk on and also can estimate what your return will be with some degree of confidence. And there's so distributed the most of the incentives and most of the yield really was coming from inflation or the kind of creation of new tokens to boost the ecosystem, as I was saying, in early stages. And then I realized we could make something very interesting out of friction, which would not rely on any of these, you could call it

inflationary strategies as the main source of return. And if we could do this properly, then we could build something that will be able to survive in multiple market cycles, which Defi has had a challenging time of doing because we haven't really seen, well, even crypto broadly, we haven't really seen how crypto does in kind of a world in which quantitative, quantitative tightening is the standard and money is differently treated as it was for the last 15 years. So really excited to try to build something that could survive in these kind of crazy macro times too. So let's double click on like yield and just unpacking what yield in in crypto is today. When people hear the term yield in defy, what is that typically coming from? Yeah. So the iconic thing in

defy is you see an APY number. And an APY number has it's very loaded term because if you from like how you think, how would I think about risk is every time you deploy a dollar, there's like some max drawdown. there's some standard deviation of a move that could happen based off of the position that you have on, and then there's some expected return of the strategy. And without knowing all of these things, you shouldn't really expect someone with large amounts of capital to be able to allocate into the system. So to break down a little bit more, the first form we talked about was lending. So if there is demand to borrow an underlying asset at a variable or fixed rate at a fixed term or open term, you can lend into a pool of assets and earn some interest rate.

Generally, this is not incentivized. I mean, nowadays, obviously, you've seen that change a lot as like compound launched in 2020, and now optimism is incentivizing AV3, which has caused total value locked on optimism to basically double in the last week. And then you have liquid staking, which is essentially just delegating the native token of a proof of stake blockchain to validators or stake pools in order to earn some, again, inflationary. block reward. And then there's liquidity provisioning. So if you find, well, one of the coolest things about crypto that also got me in the beginning was this, the idea of an AMM using a pool or a reserve to balance two or more assets to allow essentially anyone to become, take the role of a marketmaker and a traditional market and earn the transaction fees from that. In addition to those transaction fees,

these AMM platforms or they incentivize their users with some amount of native token as well to really bootstrap a bunch of liquidity to these, we can call them long-tail assets. And we've seen the AMM model evolve a lot from a simple kind of constant product in UniV-2 to now what really looks more like an order book in UNiShop v3, which is, they call it concentrated liquidity, which essentially you can place orders in a range. And from anyone coming to crypto from the traditional world, you can see many similarities to a central limit order book, which was actually one of the reasons I found Solana quite exciting because it had serum. And serum, I think so far is the most successful example of a central limit order book

existing on chain and being able to work on some scale. And then as Defi evolved, we started to see creative ways of putting all of these things together. So because in Defi, you can access tokenization, you can take that token that represents an ownership position or a liquidity provisioning position in AIM, and you can lend that into a money market to earn some more yield on top of that. But one of the most challenging things for like newcomers and veterans, as we've seen in like the last six months, is the risks of these strategies are not very interpretable. So you have to be able to unravel the design mechanism, which involves maybe sometimes interpreting and understanding the program and then actively monitor the price risk. So if you're farming on curve, which is a stable swap invariant of a concert product market maker, then you need to be kind of vigilantly watching that pool, especially since that's where oftentimes the largest amount of assets like stable,

state ETH versus ETH trade, or you saw the kind of tri-pool with various stable coins. So for most people coming into D-Fi from the traditional financial world, it makes it very challenging. And during D-Fi summer, particularly in like the six months afterwards, we kind of saw this mass drive of manpower and capital coming into to capture these opportunities at scale, which we call them like institutional yield farmers. And when they entered the market, they were like the first systematic liquidity providers to earn both the transaction fees and liquidity mining incentives,

and they quickly sell those incentives because they booked their P&USD. So it created this really cool effect of we can bootstrap almost anything by giving away some just to call it a free token that has some value, arguable, like what the actual price of that is or what it should be in the future. And we use that to bootstrap a lot of initial stuff in defy. The challenging part, though, is in a bare market when many things go to zero, how do you create a world that can survive without that? And really the only way for us to do that, we think,

is through taking advantage of the risk properties of assets. So one of them being volatility, i.e. how much can a underlying asset move over some period of time and in the future? And then using that to generate returns is one form, using mispricings between spot and perps or spot and futures. And then finding creative instruments like the power perpetual, to harvest yield in a way that doesn't require emissions.

So that's kind of like how we thought about conceptualizing friction from day zero. Yeah, I think this idea of user acquisition in crypto is relatively easy with some of these strategies that we've seen deployed over the last like two and a half years maybe, whether that's across defy, Dow's, NFTs, but retention is much harder. So, yeah, it's exciting to see platforms like Friction, building more sustainable, durable solutions across different market cycles that allows you to kind of retain these users over time. So, you know, you've described Friction as this full-stack portfolio management platform solution. Can you break that down?

What does full-stack portfolio management in Defi look like? Yeah, I think this is a really, it's a challenge. challenging phrase, I think full stack and portfolio management, and I've never been put in the same sentence together. So I can describe this in a couple of axes. One is when we say full stack, we mean both on, we really like two spectrums, so there's risk and reward. So if you think about defy as having its own efficient frontier,

we want friction to have strategies that sit everywhere on that efficient frontier. And we call these native strategies volts. We have currently four volts. We're launching two more here in the next quarter. And each volt has a very specific, role in a portfolio we think. Traditionally, the first two volts, the first volt is a call overriding strategy. The second one is a put override strategy. So these are short vol in one direction.

The third one is a crab strategy. So it sells a delta hedged power perp in order to harvest ball yield from a Bitcoin squared perp, which we can get into the weeds of how power perps work at some point. Yeah. And then the fourth strategy is a basis position. So kind of one of the more lucrative strategies in late last year or most of last year was just harvesting basis. And then the fifth one that we're about to launch is what we call capital protection. So it's a combination of fixed income or rather interest from a lending platform. And that interest is used to buy vol to provide portfolio protection. So we think about this as kind of the first principle protected and long volatility product in defy. And historically,

long volatility has not been easy for defy users to access. as we have seen continuously traded options, liquidity, pretty tough to provide without giving away too much edge to marketmakers. And then Volt 6, which we're super excited for, is going to come out here in the next couple of months. And we'll be really the friction's first fixed income or credit product to allow under an uncollateralized lending between some of the people on the platform, which probably should give an overview of kind of how does a platform look,

who comes and engages in the platform and things like that. For borrowers as well as lenders who are looking for fixed or open term, fixed rate or variable rate, and I think this unlocks a ton of capital efficiency for friction because we already have some of, I would say, the industry's leading market makers and seekers of leverage in the ecosystem who can manage this risk much better than, I would say, a prop trader on average, already on the platform. So opening up this opportunity for them, as well as our users to able to access, that form of yield is quite exciting. And what does the process for launching a Volt look like?

Yeah, it's pretty cool. It's a lot of fun. I would say the high-level overview is, well, what is the strategy, how does the strategy look historically? So always starts with fundamental, like, okay, does this strategy make sense? Is it taking too much or too little risk? How do we tune the risk parameters to make it, we can call it perform it over multiple condition over multiple vol regimes, let's say. And then what, what is the liquidity profile of this

asset? So since we launched in December, we've traded around two and a half billion of options, volume on Solana, and that covers about 25 assets. So as we've added new assets to this, the big question has been, is there a deep enough market for these options? And then how do we execute this efficiently? In terms of execution, we built this, and also this answer is a question about the full stacking atrial friction, we built this thing called Channel RFQ, which is essentially a request for quote venue in which any maker can come in and bid for this option through what we run, which is a gradual Dutch auction. And this runs on-chain powered by Sierra Mansellana. And through this process in a concentrated window of time, makers can come and bid for the options that are being traded, which allows us to have this very smooth process of trading between 50 and 200 million of options a week in kind of less than an hour and a half, which is all done in kind of inefficient style that you would expect

to see during like a Platts window auction commodities or a NASDAQ closing auction. And that's, yeah, that's that's one part of it in terms of the liquidity profile required, how much can it trade? And then demand is always a question? So is there demand to sell three to 15 Delta sole calls inputs? Definitely. Is there demand to sell like some random alt-coin calls inputs? probably not. So while we are very keen on kind of being the broadest market for

defy options, we think it's pretty challenging to get all the altcoins on there, especially because there's not a lot of natural, historically there's not been a lot of natural vol buyers. And we're seeing that change structurally over the last year and a half. I mean, Deribet had been the first mover for Bitcoin and Ethereum calls and puts and purpose as well. And a couple of months ago, they added Solana. And then we've seen like there's just a ton of demand for retail to use their native tokens for altcoins to earn some sort of return. And they're moving away from these yield farms towards, I would say, more predictable strategies. One thing you mentioned was Channel RFQ, which is your kind of internal, natively built

RFQ system. Why did you feel the need to build Channel RFQ? What motivated that? Yeah. So the team we were working with for Channel in February was just kind of hacking on. ideas about how to get better liquidity for friction. And our team is, as I was mentioning, I don't know if I mentioned much of this earlier, but a lot of our team comes from some trading market.

So I come from commodities. Alex, my co-founder, comes from treasuries. Alex and our team comes from systematic equities. And one common underlying theme in order to get large amounts of size done in any market is that you need to get everyone at the same place at the same time if you want to trade at close to fair value. And historically, this has been really challenging. for defy, especially for continuous liquidity, because you can't ensure that every marketmaker is

always going to be showing a really deep book when it's time for this auction to occur. So the idea behind channel was that we have to take advantage of doing this on chain, but we also need to run a compressed auction period in which we, and which everyone knows that this is where they come to trade slana and Bitcoin options and whatever options we may be offering that week. And since launching, like this has become the largest venue for salon and Bitcoin options in crypto or particularly defy. Since I think a lot of the auction idea is,

it makes it a lot more convenient for any market participant. And we're going to make this thing permissionless pretty soon. So anyone will be able to come in. We just need to figure out how to operationally kind of keep the system as efficient as it would be in a kind of contained environment. But yeah, I think this is the fastest way, and it's a more sustainable way than continuous liquidity for options, particularly because they're quite hard to price, especially on chain.

And the amount of edge that would be given away, in a way for someone just market-making options is quite big for retail. So you mentioned a number of stakeholders that are participating in friction as makers. Can you talk a little bit about the other side? You know, who is coming to friction and actually deploying capital into the vaults and seeking to be sellers of these options and structured products that you're creating? This is a really interesting question. We're actually writing a piece on like what is a crypto institution right now.

and that'll try to decompose what some of these like roles of depositors or users of the volts are. I think so we covered the, we covered the, let's say the supply side, which is the makers. So on the demand side, I would break it down into like four categories, three or four categories. So there's this people who are defy native that have used another chains, defy like sweet before or come to Salana and been using that before, who are pretty familiar with the strategy, pretty familiar with the payoff. the amount of education required to get them over the hurdle isn't super high. And then the next category is someone who's never really touched crypto before. It doesn't really know how defy works and needs a lot of support during the entire process.

And that's where a large part of the market is today for us. We think, like to capture in the next couple of months and years. And a lot of what we're doing is to be able to tailor to that audience. So this could be something like a centralized lender or a centralized exchange that's looking to offer an earn program to their users. And the users don't want to think through the complexities of being able to put on this position. They want to see a product and they want to make a decision on whether or not they deposit to that product. So that's a distribution channel that we're spending a bit of time on now as we see kind of

a lot of risk taking slow down from these lenders and exchanges to look for places that they don't need to be constantly selling their emissions and they don't have to be worried about the exposure they're taking on through things like impermanent loss. The next type I would say is Dow's or foundations and treasuries. So you could break those down into three more categories. But I mean, DAOs fundamentally are supposed to be organizations that are owned and governed by many people to meet this common goal.

And it's built on this idea that the smart contracting program can govern decision-making. And it's often used by protocols and applications that are built on various chains to manage their funds. And most of these funds also come from the same places. So this can be fundraised capital. there's some amount of token sale or venture capital, could be ecosystem grants or some revenue. And as these DAOs come in, they're really looking for a place to deploy risk,

but also understand that they have a lot of working capital that they need to hold up to for the next an amount of years. And for us, the strategies you work with them around are really on minimizing risk while being able to get them a consistent stream of cash flow. And it really depends on their risk profile. So we have this program called circuits, which is essentially Dow portfolio management,

So if any Dow or Treasury Foundation is looking to deploy risk into Defi or manage their treasury in a more passive way, we build a set of custom volts. So the way that process works is Dow can anyone can go to Frickson's website, click on the circuits page and fill out this really base level risk assessment, which asks you a couple of questions. So what is your target return range? What's your drawdown tolerance, various risk metrics, liquidity profile? And once we get all this information, we're able to see, well, where does each Volt at various risk levels for each Volt fit on this Dow's, we just call it their efficient frontier? And how can we put a set of strategies together for that Dow based off of what their assessment has said? And then we work with them on deploying in the right places. Generally, these tend to be more longer-term partnerships. And they're really focused on kind of fitting what.

what risk profile that that Dow wants to or foundation wants to take. And it's been a cool journey doing this, obviously through the start of the bear market. Pretty challenging to really outperform when everything's going down for any Dow that wants beta. So another product we're working on is kind of a smart beta product that would be more resilient to two times like this. Yeah. I mean, speak to Dow Treasury Management more broadly, like some research that I've seen published. is just how much Dow's have maintained in their native tokens and how, you know, poorly they are prepped for a bear market and prep for a downturn.

Yeah, talk a little bit about what it's been like working with them and whether you see them starting to shift and kind of learn from prior cycles or maybe learn from this cycle so far and become more balanced and diversified in how they manage their treasury. Yeah, this is a really interesting topic. I mean, there's about 12 to 20 billion of, we can just call it, like, Treasury AUM amongst DAO's or general like crypto organizations that are public about how much they own on the balance sheet. And as you said, most of these organizations are sitting on anywhere between like 80 and 95%

native token or the token of the L1 that they're on. So most cases, it's ETH, since they may have raised funds or they may collect revenue or there's some sort of fee structure that comes with that. I think the big question for Daos is understanding this now is the concept of diversification. It's going to take some time, I think, to convince Daos that diversification is important because for most DAOs that are sitting overweight native token, it may appear bearish to sell their native token. But in reality, it's an option for, I think Hossu was saying this at some point he's like, it's, It's kind of like an issued or authorized but unissued share in the traditional world. And it makes it very challenging because they need to both show that they're long-term convicted, but they also need to be able to grow their organizations.

And they have a ton of constraints that they face in doing that. And one thing for us is options provide a really interesting way to trade correlation. And if you can price an instrument to take out the correlation between a liquid DAO's native token, or Dow's Liquid Native Token and maybe the underlying L1 they're on, then it can reduce some of their portfolio sensitivity to massive market drawdowns, which most other instruments really can't do. And if they want to survive and thrive across multiple bear markets, they really need to understand the risk inherent in their holdings

and do a lot of work, which I think in the early stages is just education and information about how to build a portfolio. And then how to think through the risk analysis process, the design process, how we think about, stress testing and understand what factors are driving their treasury. So for us, a big thing is a lot of just education and conversation because these most DAOs, I think, are very heavy on, like, DefiDows, particularly, heavy on like the D side, not enough on the Phi side.

And a lot of this is just getting up to speed on maybe what are some like fundamentals of markets that everyone can come to. And I think it sounds a little bit strange, but in reality, there's a lot of room for DOWs to get creative, but just start with the basics in how they think about their assets. On the topic of DOWs, too, I think, you know, as DOWs took off a number of treasury management intermediaries popped up. I don't know if you've interacted much with these intermediaries, but if you have, how effective do you think they've been? And do you see them as maybe an interesting distribution channel for friction to access more DAOs?

Generally, I think that there's a challenge in a DAO in which you have to either make a decision on whether treasury management is done internally. So you bring on someone full time to think about risk management, which I don't think every DAO can afford to have like a portfolio manager or a trader sitting within their function given it may not be close to what their expertise is. And even evaluating whether that's the right person for the organization may not make sense. So often it leads to kind of an awkward place, an awkward position, awkward place for for someone who wants to take on that role, which is generally why I think it's likely going to be better

done in the long term through working with external protocols like ourselves to be able to price these things, to be able to build the strategy set and to do some of the research that I think we're good at and we spend a lot of time thinking about. And in terms of, I mean, intermediaries generally aren't super constructive for crypto ecosystems. I tend to think. I think there's a question of like, well, what are the options you have to do with this treasury? Like all the forms of yield generation earlier, they're just what you can do if you have some token that may be useful for a yield farm at some point or maybe useful to lend. But what do you do when interest rates on AVE compound and large lending platforms are now like in BIPs rather than in percentages? And your better

source of return could come from just like a bond or an off-chain asset. I think it makes these like really challenging questions for DAOs. And it also, it's cool because it's evolving into DAOs need to also have this little sales function within them, or at least from our perspective, in order to have the ability to kind to make relationships with new DAWS or protocols, treasuries, however you want to think about that. You need to have some sort of institutionalized sales process that can communicate with these folks in the DFI native way.

And so far we haven't really seen defy institutions really take off because it kind of sounds confusing. Like, you would think a defy native institutions just at DAO, but in order to, like, speak with defy institutions, we're building this team we call friction institutional essentially is responsible for everything from the risk assessment to the origination to the design of the structure of the products that these DAOs may access. And we're trying to do this in a super defy-native way, which is pretty challenging. We haven't really seen someone take a stab at it before and pretty excited to mature in some of these conversations we're having with DAOs who are sitting on whether it's like 50 million of USC or 50 million of their native token and have watched

that thing fall like 25% a week, just like 180, 200 ball asset. And I think you just don't want that type of exposure in the long term because it just doesn't enable your protocol to grow. So I want to take a step back. And, you know, one thing that you've mentioned a few times is just building a platform that basically equips people to perform in different market conditions. can you talk about how some of the vaults have performed relative to your expectations for them and then maybe relative to defy more broadly? Yeah, really good question. So the four vaults we talked about so far, the first vault, which is, and also all of this is publicly available at

apdot friction.5 slash analytics, you can see the performance of every underlying asset and every strategy. The most popular ones so far has been the Solana call overriding strategy. that's up around 15% since, or 18% since it launched in like December of last year, so about eight months now, seven, eight months now. And that's generally a strategy for people who are looking to add sole exposure and tends to be something that works during bear markets, as long as there's not a big rally up in one week.

And the way this strategy works is it's selling between a five and a 15 delta call every week. It has two voltage levels. Voltages are term for risk. So there's low voltage, which may be selling like a 3 to 8 delta, and then there's a high voltage, which may be selling a 10 to 15 delta call on a weekly basis. So those strategies have done quite well. The put strategies, however, which are essentially just selling downside volatility, have not done super well.

They suffered from losses when market gaped down, both in kind of May 9th, the Luna weekend, and then middle, early June with Celsius through AC. But the idea for us with those is like they're specifically made to perform in a time. Like that's generally more bullish strategy than the call overwrite one. So tends to be as long as it's following what we think it should be doing in terms of performing, as the strategy said, we're pretty happy with that. The crap strategy, which is a bit more interesting, essentially makes a view every week that spot stays within a range. And that range can be defined. And that range is defined every week based on what the funding rate for the Bitcoin squared perpetual that trades on entropy.

our kind of exotic purpose decks is. And that profit range, that strategy has been quite profitable since it launched. It suffered like 2 to 3% drawdowns during May and June, but it's still up around 10% since it launched in April. That's the Appdorfiction.5 slash crab. The basis strategy initially launched in a super negative funding rate environment and the strategy was just long basis. So collected a lot of funding during that time. since then funding has normalized on the decks that it trades on, which is mango markets on Solana. We're working on expanding this to a variety of centralized venues through our partnership with Paradigm,

which is kind of the largest RFQ off-chain, off-chain RFQ layer for options. And we're excited to kind of bring centralized exchange liquidity to a tokenized basis strategy. Obviously, an interesting time because basis generally correlates with its leverage in the ecosystem, and it's not quite high right now. But it'll be cool to see when this pops back up again. Something you've mentioned a couple times during this conversation is the power perpetual product that friction launched. And in the past, you've talked about how this is an instrument that is net new and unique to defy. Can you expand on that?

Yeah, I think power purpose, we were super excited about when we were getting entropy up and running in March as we realized that there is no real similar innovation in really anywhere. I mean, there's like a variance swap, which is a commonly traded or more. exotically traded options instrument or options position. Initially, power purpose were kind of cooked up in August of last year by the team at Paradigm. And then they kind of implemented this through Open on an automated market maker on the on an ETH contract called Squeeth. So this was an AMM that traded ETH squared and it paid funding in ETH.

So our different towards this was our contract and our product was on entropy was built on a central limit order book. So using the infrastructure from Mango. And that funding is paid in USC. So it very much looks like a normal perp that you would trade on like a centralized exchange, except it has this interesting power of convexity. So to unpack that a little bit, when you trade a normal perp, again, perpetual perp is a perpetual swap. It's an instrument that's crypto-native that allows you to essentially get non-expertory exposure to an asset. So it's like a future that never expires and never has to roll. A power perp is unique because instead of paying off, such as returns in a normal perp would be 1 plus R where kind of R is the return in percentage,

the returns on a power perp would look like 1 plus R squared. So you can expand that to R squared plus 2R plus 1. And then the 2R term, you can view that as just a 2x levered version of a normal perp. So you can hedge that with your delta from a normal perp. The R squared term, which you can think about that is like the expected variance of the instrument. So let's say you're trading Bitcoin. This could be the expected variance of the Bitcoin, which is similar to the concept of gamma and options or convexity. Essentially, it means that your upside is less bounded than your downside on an instrument like this. So to give it an example, let's say spot moves up 10%, you make 10% return your perp, spot moves down 10%. You lose 10% on your normal perp. But in a power perp,

if spot moves up 10%, you're making 21% return. And if spot moves up 10%, you're making 21% return. And if spot moves down 10%, you're losing 19%. So it has this really cool property of convexity, which for anyone interested in reading more, we publish a lot of kind of research and content about this on friction. So definitely check out the blog if that's more interesting. But essentially, like one cool property here is the power perps will always have a positive funding rate because the longs need to compensate the shorts for having that asymmetric payoff we were talking about.

And this is very cool because now this power perp basically looks like an option that never expires and has no strike. It just pays out funding on whatever window it's designed by. And we thought this was very exciting because you can now harvest what looks like an at-the-money fixed strike straddle that rolls, let's say, every week based on the rebalance window of the crab strategy to collect a volatility premium every week from an underlying asset like Bitcoin. And generally, crypto-valls are much higher than like triad-fi asset falls. So it makes it interesting to try to harvest during times of high vol and also see how it performs during times of low vol. So the high level view here was like, this is the first time you can harvest ball premium at scale with an asset that doesn't exist or a product that doesn't exist in Tradify.

And one thing that's obviously a challenge here is this discretion of scale. So how do we take this kind of confusing instrument and how do we get a lot of people to want to provide liquidity to it? Well, we took a kind of more difficult route with the club or a central limit order book, rather than an AMM where people can just deposit into it and set and forget just like a normal AMM. So doing this is challenging. Finding people that not just want to price a power perp, but actively traded and put positions on has been fun. I think it's a lot of working with some of the smarter market makers and getting people to spin up their own bots to try to market make on entropy, which if anyone else thinks this is interested in marketing and entropy, we'll always in Discord talking about this stuff.

So drop us a message. Talk about entropy and the decision to kind of build your own native decks for unique, unique products and options directly within friction. Yeah. So entropy is just a venue to trade exotic perps. And it trades Bitcoin squared perp. It also has this really cool product that we were super excited by hasn't gained a ton of traction, which is an implied vol perp. So essentially it's taking a Oracle from a centralized exchange to calculate the, let's say, seven day or one day at the money implied volatility, just a number. And it's allowing someone to trade that number as a perp, which is kind of weird to think about it because in defy, you can just like take an index and turn it into a perp. Obviously, can't do this in Tradfai. So the idea was, like, we would create these venues because we just wanted to get this product to market and we needed a place where people can exchange these power perps.

And it didn't exist on Solana when we were kind of building it. We really wanted to have it. So we just kind of launched it to get it out there to able to use as the settlement layer for this. Obviously made super possible by Mango, who has an amazing, or the Mango v3, which is an amazing open source perps decks that you can essentially plug in various oracles and turn them into these different feeds. So we were pretty excited just to get that up and running. And now the challenge for us is to scale that and see if this is a really popular product over the next couple years in crypto. I want to talk a little bit about just building something like friction on chain and what you have identified as some of the biggest, you know, advantages of having something that's on chain, transparent, automated, and how that's helped attract different user groups. And maybe this is also a good segue to talk about your kind of like institutional strategy. and why you think something like friction is attractive for institutions and why you're kind of going after that segment. Oh, yeah, this is an awesome question.

I think about us a lot because in the last three months, we have seen a massive de-leveraging, obviously, in crypto that was largely credit-driven. So a lot of people say it's more of a credit winter than it is a crypto winter. And a lot of the reason this happened was simply because people were not talking to each other. and the 10 people that needed to be in the same room weren't in the same room, and they weren't able to see each other's positions across various places. And I think it's easy to quickly say, like, Defi solves these things, but in reality, this is just a problem of, like, human coordination. And I think one answer is Defi because in our view, like all of our positions,

our auctions are held on chain, the underlying assets are sitting on chain, so you can track the flow of them whenever anything's happening. We're highly transparent about the performance, historical, whether it's positive or negative, and future expectations, where we see the strategy going. I think that's just one awesome advantage that Defi will continue to have, is any institution that wants to come in should be able to have a base standard to vet what the integrity of the information they're getting,

what the risk of counterparty or smart contract that they're taking. And a big job for us with friction institutional is to demystify some of these things. Like, what is counterparty risk in Defi? So it's still a hazy question, right? And counterparty risk in defy still exists. Like I don't think it completely goes away. I think it's just a different type of counterparty, and there's a different way to measure it.

And you need to be much more plugged into a developer-friendly, a developer ecosystem rather than into kind of getting on the phone and talking to the OTC guy. So to answer your first question, our approach is always going to be kind of much more transparent risk management and opening up everything we can to avoid these type of opaque de-leveraging cycles and do what we can from our front. I mean, our products are attended to be very retail-native,

as we were saying in the beginning. Another big push forward for us is there's two parts of kind of institutionalization for us, one being KYC and custody. And in that part, really getting all the custodians so you can think like the fire blocks, copper, anchorages of the world, comfortable with friction as a platform with our contracts and making sure that the security is like top notch. So we believe in like kind of quarterly auditing processes. We finished our first audit in June and our focus is really keeping that quality up as we launched new volts and we may take risk to various other smart contracts and then in terms of kind of the know-your counterparty know-your transaction it's quite important for us if these institutions want to

come on board that they're comfortable with the person on the other side of the transaction and even if they're still using defy rails to maybe take a fixed term or variable term variable rate loan that they can see that person and they can potentially talk to the on the phone to that person. Otherwise, it kind of defeats the purpose of a lot of what DFI is supposed to be about, I think. While in the early stages, everyone was like, oh, you never need to talk to your counterpart, you should never really know what's happening. But in order for defy to, I think, get scale amongst institutions who are already comfortable in a system where they can talk to people or they can understand who's on the other side of this trade or this position

or deal. It's quite important for us. I would never have thought two years ago that if I was building something in defi, we need an institutional team. But today it makes even more sense, given like how much of this last six months has just been like relationships or leveraging relationships between counterparties and that tipping the scale for these weird events. And as more regulation comes in to prevent things like this, I think we're trying to position ourselves quite well in terms of using defy as the infrastructure layer for this. Yeah, definitely. I think it's also a shift in thinking about defy as something that needs to be completely anonymous where you don't need to know, you know, the counterparties that you're transacting with. And it needs to be trustless

and permissionless to defy as, like you said, enabling infrastructure or technology that that just leads to more transparency, more accountability and just unlocks, you know, better like risk management, like what you're doing with friction. for institutions as well as, you know, more sophisticated retail traders. So have you, you know, in operating friction this year so far, have you identified any learnings that, you know, you feel you can use to improve your existing set of products or expand the portfolio of tools that you're offering?

There's been a bunch of stuff that we've iterated on just quickly as we've seen feedback. So one really powerful benefit of doing stuff in DFI is like, So we have this community. We call them Lightning OGs. They are, or Genesis wielders. They're our first kind of power users of the protocol, about 800 of them that have been early contributors helping the protocol with feedback, with ideas, and just honest opinions that sometimes as a team you may not, you may look over.

A big part of us has been kind of leading on this community. There's pretty strong community of people to engage when we're trying to think through a new product idea. We're trying to communicate it to users before it really goes out into the wild. And I think we've taken a ton of advantage of this and we've built out this community to be a lot of fun. Hopefully have our first like IRL event. And this is an FD, by the way. So if you're an NFT holder for these OGs, you can participate in a lot of these benefits.

A lot of the iteration has come through that. A lot of it has also come through building out our team in terms of our kind of quantitative research and analytics capabilities. So how can we turn ideas that we may previously have thought were unscalable and defy and like how can we make them more scalable? So the idea of a power perp still isn't proven to be the most scalable thing, but it's something that we're constantly hacking at and seeing like, oh, what are some better ways or is a power per better than not the money straddle, for example, something we just talked about last week and constantly thinking about these things. I think as a team, we tend to do a lot of tinkering, and this can result in like random conversations for three hours at 2 a.m. about why a product is better than another product. And then like the next day, we do a survey in this group of Discord users and they actually tell us what we thought was wrong. So we do iterate a lot as we think about these things. And talking to customers is a huge part of it.

I think we're about to hit our 800th support ticket in Discord. And that's been a lot of fun for us. Like we see questions, new users come in. We saw a lot of users who are calling themselves Celsius casualties, like signing up on our Discord and getting onto friction and saying like, hey, how do I address all these 10 risks that I had with Celsius? And I'm moving over to D5 for the first time. How do I make sure this is safe?

So I would say our feedback historically or growth has been product driven historically. It's becoming more community driven now. And then in terms of users as we mature to more institutions, we'll be taking a lot of feedback and input from them in order to scale out the credit product. And then just shifting gears for a second, what are Frictions main sources of revenue today and how do you expect that to change or not in the future? Currently, we have a performance fee on the Volt. So when the vault performs within an epoch, so epoch is kind of the duration in which the strategy of a volt runs.

If the performance is positive, we take some 10% performance fee and there's a withdrawal fee, kind of like a switching cost mechanism for those volts. We're getting kind of consolidating feedback from the community and the users right now of whether they think a generic kind of holding fee or management fee is better. I'm personally not the biggest fan of the concept of that. I think DFI is not really made for management fees. I think our long-term fee structure is going to be more incentivizing users to be in a diverse set of volts so that their portfolios are resilient. And there's no point making money from a user who may actually be net losing money from the platform. Our goal is we make money when the users make money. And historically, most of our revenue has come from the first two volts.

So first year to date, I think we're at around 700,000 of revenue kind of on track for like 2 to 3 million air. as a protocol. And then this would be something that the Dow Treasury eventually makes a decision on how to allocate and how to redistribute, whether it's reinvested into the protocol or a bunch of options to be done with it. I think it's kind of nice to be in the pre-token world where we're still thinking about these things rather than having this be like a top of mind concern for how do we optimize the tokenomics here and here. You're operating as a centralized team and platform today. And I imagine there will always be components that are centralized, right? So what components of friction do you plan to decentralize?

And then why do you think decentralization is necessary for those specific components? Yeah, this is a really interesting question, kind of philosophical. I think when you think about it, you can't really decentralize labor super well at scale. it becomes hard to coordinate. And by definition, like coordination requires centralization, for people at least. So I think the part that we continue to have to think about, and there's a lot of people to learn from here, I think like the way Uniswap Labs has done things, the way NOSIS has done things, D-YDX, all these organizations that are like 50, 60 plus people that have found ways

to build protocols at scale. Like when Uniswap decentralized or made the big push towards decentralization, like the final version is just sitting out there and people can contribute to it. In our world as well, I think once we get to Volt 7 through 10 at some point in that range, it'll be the point where we're like, okay, the volts are now permissionless, anyone in the community can spin it up, but there's a base requirement for how much liquidity needs to exist in a Volt in order for that auction to be held. So that's a bit of the tradeoff between how much it can be decentralized. And a couple of other things are just like decisions. Generally, I believe in like a more council.

based approach. So some sort of kind of structure, I think like there's like many countries and this kind of goes back to geopolitic stuff, but like there's many structures to learn from in terms of who you want making decisions and like how intelligent they are. Right. Like we've seen this consolidation of voting or governance in defy or crypto generally has has been largely dominated by like early token holders, whether it's like VCs or early contributors. And it for the most part, it's, I don't think what the end state of crypto looks like. I think some of that, um, people just need to bootstrap decision-making and get things across the line. But I think we're going to see this within the friction community as well as, like,

people stand up to be these kind of, I would say leaders within the community to make some of these decisions. And when it comes to parameters, I think it'll always be tricky because there's going to be a subset of users. And maybe some of the core members, maybe some like very smart market makers or traders on the platform who have thoughts that should be weighted more or less than other people who don't have those. backgrounds. So I think building out these, I would call them like task forces within friction. A good example is like Rari. Rari started off with task forces and then task forces turned into like their own protocols. I think like three or four protocols launched out of that. So as we as we develop that, which it's going to take time. I don't think it's going to be fast.

And if it's too fast, it may end up resulting in like a massive coordination problem that probably can't be solved. So we want to take a gradual approach to this. You guys are clearly super ambitious and have a lot of moving pieces. How do you think about balancing breadth versus depth? And how does that factor into friction's longer term roadmap? So I think with the idea of full stack portfolio management, there's like a universe of maybe 10 volts that can cover, let's say, like 70 to 90% of the strategies that someone wants in their basket.

And it may seem that like we're going. broad to get these out there. But once we get these out there, like our focus is now going to every Volt and getting deeper and saying, how can, now that we have this like simple market for, if you notice, like, we just didn't launch a Tand Delta option selling product as a Volt 1 because we thought it had some, like, structural inefficiencies, right? We launched various risk tiers, so users can tranche their risk to some extent with the Volt, the option overriding strategies. So for each product we launch, we want there to be something that differentiates it from both, like, what's in the market, but not just for the sake of differentiation, for the sake of having a better

product. And I think it's having something that users can toggle their risk with. And once we get to, I would say like seven through 10, we'll see a lot of our efforts consolidate into, okay, well, how can we scale the crab strategy? How can we get access to basis on every centralized exchange in the world through defy rails? And how can we get, how can we tap into the credit markets from institutions that don't even want to think about crypto credit today? And I think once we get these out there, it really, I think the bear market is a perfect time to do this because we, we, we, we, we, we, find segments of users that have demand for these things, and we use that to bootstrap the product. And then we grow teams around this, right?

Like friction institutional kind of comes in as this umbrella team that says, hey, we're going to take these products to market in a place where the current friction team couldn't do. How can we get this to an audience that doesn't think about crypto today? So that's kind of how I think about breath versus depth. Well, Uda, it's been awesome to have you on what you have built and our building at friction is truly impressive. I want to emphasize that your focus on kind of U.X and user education and providing as much data and insight and analytics around, you know, what users are actually putting their capital into, how their overall portfolios are performing. Like, it really stands out relative to some of the other DFI protocols that we've seen out there.

So highly recommend users to, one, check friction out or listeners to check friction out, but also, also read a lot of the research and education that you put out there to make this something that is accessible by a really wide set of users. So thank you for coming on and answering all these questions and for all the great work that you're doing. Yeah, really exciting conversation and glad to be here. I think on the education front, one last thing to throw in is we're launching this thing called Friction Learn, which is essentially a Defi Academy.

So you can go from zero to one in your defy knowledge by hopping on this platform. And without having to take risk, we think you should be able to understand all of these forms of field generation we talked about earlier. So really excited to get that out there and looking forward to it.