Lucas Schuermann – Swapping Out Perpetual Futures (S7E34)

Hey everyone, Corey Hofstein here. Before we get started, if you're into capital efficiency and smarter portfolio construction, mark your calendar. The second annual Return Stacking Symposium is coming to Chicago on October 28th, 2026, headlined by Cliff Asness, founder and CIO of AQR. It's a full day of conversations on how to stack return streams, put your capital to work more efficiently, and rethink the way you build portfolios. Whether you're an advisor, allocator, or just curious about return stacking, this is the room to be in. Head to www.returnstacks.com slash symposium to learn more and grab your spot. That's returnstacks.com slash symposium. Now on with the show. All right, Lucas, are you ready? Yep. All good. Let's do it. All right. Three, two, one. Let's jam. Hello and welcome everyone. I'm Corey Hofstein and this is Flirting with Models, the podcast that pulls back the curtain to discover the human factor behind the quantitative strategy. Corey Hofstein is the co-founder and chief investment officer of Newfound Research. Due to industry regulations, he will not discuss any of Newfound Research's funds on this podcast. All opinions expressed by podcast participants are solely their own opinion and do not reflect the opinion of Newfound Research. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of Newfound Research may maintain positions and securities discussed in this podcast. For more information, visit thinknewfound.com. My guest in this episode is Lucas Sherman, co-founder of Variational, which seeks to bring the trillion dollar OTC derivatives market on chain. This is my second episode on Variational. Little over a year ago, I spoke with Lucas's co-founder, Edward Yu. In that time, the mission hasn't changed, but a lot else has. We begin with a reintroduction to the business and its retail-facing platform, Omni. Unlike a central limit order book, where a crowd of competing market makers quote against unknown flow, every trade on Omni is quoted by a single, internal liquidity provider, the OLP, through a request-for-quote model that segregates flow. Because OLP knows who it's trading with, it can price retail flow as non-toxic, netting offsetting positions against each other, warehousing what it chooses to keep, and hedging only the residual risk externally. It's a model that lets Variational list the long tail of crypto-native assets at competitive spreads, even where on-platform interest remains thin. The surge in demand for real-world asset exposure on-chain has pushed that model in two directions. The first is widening where those hedges can go. Rather than trying to rebuild 40 years of traditional market depth on a crypto order book, Variational has built a global network of dealer relationships to tap into it directly. The second is questioning whether the perpetual future is even the right instrument to begin with. Perp funding is driven by where the contract trades relative to its index, which can make it volatile and hard to forecast. Variational's answer is a swap, a price-return leg plus an explicit financing leg priced off short rates, a far more predictable cost of carry for anyone who wants to hold levered exposure for the long haul, and there's currently over a billion dollars of dealer capacity behind it. I hope you enjoy my conversation with Lucas Sherman. Lucas, welcome to the show. Excited to have you here, talking to you from across the world, late night here, early morning there. I know you are super busy. Extremely busy summer for you, so I appreciate you carving out the time. This is going to be a fun one because it's in many ways a follow-up to a conversation I had with your partner, Ed, over a year ago. So this is kind of a part two for listeners who are just seeing this episode that haven't heard of Variational before maybe, or didn't hear my episode with Ed. Please go back and listen to that one. Hit pause here. Go back, give it a listen. It's going to give you some great background as to what's going on with the platform. And for those maybe who aren't going to go back, Lucas, maybe you can... We don't need to go into a full formal introduction of what Variational is, but maybe you can give a brief reintroduction as to what the Variational platform is and who you're trying to serve. Yeah. A brief reintroduction. I'll squeak in about myself as well. I like to joke sometimes on the discount version of Edward. The other side, the other co-founder, Ed is a quant by background. We've been working together for the better part of a decade in hedge funds, in market making, in HFT, and in crypto. I sit more on the engineering side by background, but these days I'm more on the business side of the protocol as well. Skipping forward to just a very brief introduction about Variational. Our main goal is to bring derivatives trading on chain, and that takes two main forms. One is retail derivatives trading. We're building a broker-like model as opposed to an order book or an exchange, using RFQ to match versus our users, and offering, because of this different model, some very interesting benefits in terms of liquidity aggregation for RWAs direct from TradFi, zero-fee trading, hundreds of different pairs, some of the best liquidity you can find, and a one-click interface where you can trade all of this from a single account, a single USDC balance, so we can get into that more. That's been very exciting, and that's been growing incredibly rapidly since your last conversation with Ed, so we've been very happy about that. Our longer-term vision is not just stopping with retail, though. It's pro and the institutional side. We come from this background in OTC trading and crypto and beyond, and OTC options and structured products, and that's a very interesting segment to us, asking the question, how do we bring institutional trading, settlement clearing on chain as well? I liken that to the ambition of Stablecoins for FX settlement and remittances, where the other side of the coin, we're saying, hey, that's working really well for fiat, moving that on chain. How do we move the derivative trading piece broadly on chain? That's our mission. As you mentioned, you've seen incredible growth in that retail side, the part of the platform that you call Omni, so we'll focus most of our attention there. Maybe we can turn to the institutional side a little bit at the end. With Omni, this is a platform where you are charging zero maker and taker fee. As you mentioned, it is an RFQ-based platform, which is very different than the central limit order book platforms that we have traditionally seen, both in the centralized exchange and the decentralized exchanges that have come to market. Now what's interesting here is it means that the revenue that you're going to generate because you're not generating maker or taker fees is entirely in the spread, but your pitch is that that spread is going to be a tighter all-in cost compared to other venues that are going to charge fees on top of their spread. Can you walk us through maybe how these things can coexist? How can you charge a tighter spread than a venue that is going to have those fees? I think there's two questions really to answer here. One is the economy scale of market makers and why that's an amazing flywheel. The second is walking through the life cycle of the trade. Maybe a common misconception that you're asking about that we can blast through to start with is that because we aggregate liquidity, because we hedge, which we do, we don't counter trade users, we don't V-book, that our execution quality on these venues must mean always crossing the spread and the taker fees and therefore that our net execution cost should be superset or higher than that. That's not the case for the same reason that if I trade against Jump or Jane or Hudson River Trading or in crypto, Wintermute or Selenium and beyond, you can imagine that if I do a large order with them, they might be aggressing to offlay the risk. They might be trading on a variety of different exchanges to build those hedging positions on the other side. Let's say I sold $100 million of BTC to Wintermute, but does that preclude them from quoting me very, very, very tight on that order, tighter than the taker fees on Binance, tighter than we could possibly walk the book at that size on any of these exchanges? No, it doesn't preclude them at all, right? I mean, that's very much their core business. So when I think about how does OLP aggregate liquidity, we use aggregation as one way to describe kind of our model, this intermediation of the flow, this principal brokerage style model. But another way to think about it is we act like a market maker. So our goal is to build a large economy of scale through internalization of flow, through broad connectivity to hedging venues, through economy of scales that we have at the hedging venues and beyond, such that just like a large market maker would, you know, we're quoting better than the sum of our parts, both in terms of fees and in terms of spreads externally. So the short answer to your kind of secondary question, which is how can we provide a better price than one could get executing that whole size immediately onto, say, a Binance or immediately onto Hyperliquid or whatever have you, it's for the exact same reasons as any market makers. And we can come and pack that in a bit, but that's very much the economy of scale and kind of the whole point of this intermediation that we're doing. On the former side, on the flywheel, I actually just wanted to briefly kind of contemplate what are the drivers that allow market makers at scale or our system at scale to offer such tight spreads, but to still generate quite a lot of revenue. And again, I'll liken this to our comparable in traditional finance. Let's talk about it, that Jane or the jump, et cetera. They generate billions of dollars in profit every quarter, maybe even every month, depending on who you're at in the market. So these are amazing businesses, but does that mean that these spreads that these guys are providing onto, say, retail brokerages like Robinhood or onto the actual liquidity that exists on NASDAQ or NYSE or CME, does that mean the spreads are wide? No, no, not at all. The bigger the economy of scale, the cheaper they can be because the lower their fee tiers are, the better connectivity they have, the more internalization they have, the better short-term health that they have, et cetera. This is our broad mental model, and it's why I'll make an even stronger statement than your original question. Our execution costs are some of the best anywhere on Omni because we have zero fees plus very tight spreads, but we expect those to actually improve over time, but mostly due to the economy of scale here. As the system gets bigger, just like any other market-making system, we're able to drive a larger economy of scale, higher internalization rates, et cetera. We can actually simultaneously expand our margins, which is good for the protocol, and reduce our spreads, offer better size to our users. So let me continue to push on the first point you made a little bit because any other market maker at scale is also going to be implementing cross-venue hedging, right? So if I go to put an order on Hyperliquid or Binance, large, sophisticated market makers are going to take that order, and they're going to be able to hedge it across a multitude of venues as well, as well as internalize the risk as they see fit. Now, that's the same way in theory that you can. What is actually structurally different about the RFQ model that allows you to potentially create tighter spreads than a market maker could in a CLOB model? There are a few nuances here, and I'll unpack probably three of them, but let's start with the big one, which is called flow segmentation. The idea here is that there is a huge difference between us taking the other side of a trade on an omni-platform, more like, in your example, someone trading on a single-dealer platform or with a known counterparty in an OTC block trade, quoting via RFQ. RFQ is just kind of the matching primitive. I think it's interesting. Like, of course, we like to talk about RFQ as a clear differentiator versus limit order books. It's easy to understand. But really, the biggest difference in our model is having this single system, this principal brokerage on the other side and being a broker-like aggregator as opposed to having an order book for price discovery. So the point being, RFQ versus order book is a bit of a semantic way in which we do that matching. The bigger difference is actually, again, we're on the other side of the trade. We have full information about who's facing us. We're getting the quantity and the size all up front, this information. And what it allows is essentially safety and information leakage, or let's call it an adverse selection on the flow. So let me unpack that in a more specific case. Firstly, you know, in my $100 million example, let's make it a little bit more reasonable, let's say $10 or $50 million. No one in their right mind is walking the book on an order book. Firstly, these businesses like Jump and Jane and everyone else who's posting liquidity there, they're not charities, and they will very quickly see that someone's walking the book or even if you're trying to TWAP it on kind of a public exchange, they're very quickly going to be moving the market price, right, because their goal is to maximize profits per dollar traded. So that's number one. Number two is as they generate more risk on a public order book, you don't know who the other participant is. I could be sweeping the book as a taker with a lot of risk because I have short-term alpha and I'm sitting at Tower or HRT and Jump and everyone else is going to be scared of that. So the idea of providing a good all-in price depends a lot on the segmentation of flow, or in other words, what you would assume is the toxicity of the flow, because taking on this size, this amount of inventory and then going and hedging it, a lot of the game is in understanding how long are we okay holding that inventory? How much will we be able to internalize it versus other participants that are coming in? And then indeed, what is our connectivity and our view on the hedging venues and where theirs are going and how much we can size up? So the short answer to your question, pillar one is flow segmentation. And this is the biggest mechanical difference between trading via RFQ or more particularly via what our system looks more like a single dealer platform in many ways. And again, like trading on an order book is that we don't have to be taking down flow or posting quotes that can be sniped by anyone that can be front run by high-frequency takers that might be trading into the public void of who knows how smart the guys are at the top of the book on Binance. And I'll give you a cheat code. They're very, very, very smart. So you have to be very defensive on that. It's a certain amount of trepidation, whereas for us, we don't have an API yet. It's on the roadmap. We can come back to that question later, but we're optimizing for what's the best possible price we can show to retail order flow when we know that it's retail order flow, when we can see who's on the other side of the trade, and when we can get comfortable with quoting very big size, because this isn't adversarial or front running or arbitraging. These are guys just trying to get the best all-in price for their block trade or their entry into a mid-frequency day trading position. And that's a completely different ballpark in terms of how you can quote it. To put this all in a summation form, our liquidity equality does look like what it would look like if you're looking for an RFQ from, say, I mean, a tier one desk, right? Now I'll pick on, again, a jump or a winter meet or something like that, but I would expect it to kind of be in the same ballpark. But the difference is that's not at all what's happening when you're matching on a public order book, but that is what's happening if you are, say, a $10 billion institution onboarded onto an OTC trading relationship with one of these guys. So one of the ways I like thinking about what we're doing at Variational is democratizing access to that quality of execution and that style of trading, which is massively usually better for the end client than having to trade directly on an order book. There's a few more mechanical things, but in the interest of time, let me pause there on your question. I think that's one of the big lion's share reasons. And maybe my next question will get you to continue to extrapolate, because let's give you the benefit of the doubt and suppose that this all works as you've laid it out, which leads a majority of non-toxic flow to move to Variational, leaving largely toxic flow left at all the venues that you would be hedging on, which means your hedging footprint is likely going to grow larger and larger, implying larger spreads, and again, correct me if I'm wrong in all this, it strikes me that it would become increasingly difficult for you to hedge that flow on the venues, given that the liquidity that they're seeing would be decreasing over time. I agree in part with the last part of that statement, which is actually asking the question over time, will we have as many venues for hedging crypto and is that necessary, especially in a world where most retail trading, which we call the soft non-toxic flow, moves to broker-like platforms, namely ours, we're kind of defining that category. This is what happened in traditional markets, by the way, how many different brokerages do we have? Many, and we can get into why, why I think there'll be much smaller numbering, mostly us in kind of the crypto markets and on-chain revolution that's happening right now. But there is a reason why we don't have all this fragmented liquidity and so on. But I'll strongly disagree with the first part of the statement, which is that if we kind of think through a bit from first principles, let's say variational is soaking up more and more and more retail flow, which is what's happening right now and is very much our goal. And this is moving from other platforms where it might be currently trading directly onto order books, indeed making them more toxic in some sense, and it's moving more and more towards variational. This is the absolute dream scenario for us. We have soft flow to monetize. We have some cases of short-term alpha or some information signal that's coming from the flow. Soft flow drives a huge portion of notional and nominal activity these days. We have the ability to internalize it. In other words, I like to joke about squeezing the lemon. We don't counter-trade our users and we're quite clear about wanting to always be more of an aggregator, more tilted towards just providing the best all-in price than necessarily juicing the spread. That's very different from a market maker, that's why it's a lossy comparison where I'll pick on maybe Jane Street, like their goal is absolutely to maximize their profit on the desk. They're not a charity. They're not trying to say, hey, how can we make these spreads tighter to build a better retail-facing brokerage product? No, they're like, how can we possibly juice the spreads as wide as possible and still get people to transact? The point is to return to the top. In a world where we have all the, call it soft flow, the non-toxic flow, this is a dream scenario for monetization. It's a dream scenario for pricing because we know our flow is non-toxic and therefore we can quote it really, really, really tight while still monetizing nicely. And it's a dream scenario for hedging. We have all the negotiating power in the world. If everyone else's flow is toxic and ours is not, that's the good thing. That's the best thing because even our aggregated hedging flow is still highly non-toxic and would be very desirable to these venues, kind of in this toy example where you assume the books are getting more and more toxic. But what I would say is like even one layer up, it gives us a huge amount of negotiating power just at that size and scale. Again, if you assume, let's call it non-toxic order flow, somewhat scarce, which it is even in traditional markets at equilibrium, it gives a lot of negotiating power to the guys who are the ones orbiting where that goes. Do we hedge it OTC just to a small set of counterparties? Do we hedge it on what I think will eventually exist in crypto like ECN style platforms like dark pools and one level lip pools? Or does it go directly onto exchange? It'll be an interesting question to see how it evolves. I do agree with the premise that it will lead to a thinning of the herd because you do need those quote unquote toxic, like those price discovery, high frequency venues, the NASDAQs and NYSEs and CMEs are those in the traditional markets. But absolutely, I think it makes sense that those will eventually look a little bit different. And it brings me to kind of a final point here, which is, I like to say we're not a direct competitor to hyper liquid or a lighter and we really respect their businesses. I think many of those will succeed both in crypto and beyond. We're similarly not a competitor and in many ways like a downstream client or beneficiary of CME and NASDAQ and NYSE and so on, and the same way that traditional market brokerages are. But I do think that we have to ask the question, exchanges are infrastructure. They are for order matching, they are for price discovery, they are for institutions hedging against each other and HFT. I do think eventually retail flow will move to a broker like model like ours. You mentioned the potential introduction of APIs. How does that change your thinking about toxic versus non-toxic flow that you'll be facing on the platform? I'll give you a surface level and then I'll give you a nuanced answer. The surface level answer is beauty of RFQ, beauty of our model, broker like platform. So we have perfect information, perfect flow segmentation. A quick sub question of what you're asking is how do you avoid getting run over? What if you do have a adversarial counterparty connect to this API? Isn't the whole idea that we're able to quote really tight because we have the soft flow on the UI, et cetera? Yes, but the reality is to put it incredibly reductively, it's quite trivial to protect ourselves. We don't want to service real retail and that includes guys who might be using cloud and building APIs, strategies and so on. What we don't want is truly high frequency takers like indeed the tier one HFTs we were just describing to be trying to arbitrage our prices. And that goes back to your earlier point on the order books. By definition, if we allow that to happen, we have to widen out prices for everyone. That's not what we want to occur. So instead what we'll be building is ways to detect and monitor that. on a individual by individual basis, let's just say protect ourselves as necessary. And that allows us to still service the 99.999% of the world, both on the UI and even on the API that will be on that high frequency basis, highly non-toxically very normal because that's an incredibly small number of firms that would really be looking to arbitrage or causing issues there. The slightly more nuanced answer right is this is the exact same problem set, let's say, as indeed, like again, just returning my example, like what we'd be facing for building a single dealer platform. And I have it parts of my career at a market maker like a winter mood or in traditional markets, a jumper, HRT or so on. You want to figure out the toxicity and information leakage on every single venue. And that includes the cross connectivity you have to direct retail clients or API clients. And there's a lot of theory that goes into how to measure that, what to do about it when you do identify that there's some toxic flow and indeed kind of how to balance that such that you can still win the flow. We'd like to service API clients essentially up to that ultra layer because again, we're a brokerage dump doesn't trade on Robinhood. That's just a silly premise. From a first principles perspective, it's not our core business. And as you can imagine, there's a lot of tools once you've identified to do things like short of shutting it off, like you can widen, you can add speed bumps, all sorts of stuff to stop the abuse. One of the biggest evolutions that's happened over the last year since I had my interview with the Ed has been the growth of real world asset perps RWA. They were around for years with like gold perps and they were largely a niche product. But in the last 12 months, they have really exploded in depth and volume and open interest and just seem to be attracting a lot of attention and a lot of platforms are figuring out ways on how to list them. What we've seen is a lot of venues have gone to list them the same way they listed crypto that they have an external market makers who are quoting an on chain order book. But now those external market makers have to go figure out how to hedge in markets that live entirely off chain predominantly and sometimes aren't even open. You guys in your series a announcement, you made the point that you can't rebuild and this is a quote from you, I believe 40 years of traditional market depth from scratch on a crypto order book. Talk to me a little bit about why you think that the way other venues might be going about this is overly inefficient, overly expensive and how you're looking to solve this problem differently. Let me unpack that because I think it's a great question. It's something that we think best illustrates why our model works incredibly well and I'd argue absolutely the best for solving our debate trading on chain. A couple observations here, quite a few. Number one, the amount of liquidity in traditional markets and this is perhaps the most obvious statement in the world is incredible. We're talking over decades and decades, global networks of participants, dozens and hundreds of different HFT firms, banks, non-bank dealers, broker dealers, so many layers and we can get into even what it looks like in the US versus abroad but just incredible amounts of capital structure, liquidity, sophistication. When we're comparing even the grand success and we have all the respect in the world for say Hyperliquid and the Trade XYZ team but when we're comparing the depth of like an oil order book there which is one of the most tradable things on chain right now, let's say CL on XYZ, it just pales and I mean pales on its face in comparison to the depth on one venue right like CME in this case but this goes across the whole board where when we think about the aggregate depth in traditional markets, there's so many orders of magnitude greater that it becomes silly to compare it on a certain point. We're talking about markets that trade single digit and sometimes even double digit trillions a day versus markets that trade a few hundred million. These are exponentials, these are orders of magnitude so there's huge differences. We're also talking about at the end of the day a subset and again with all the power to them and all the respect for defining this category first, a subset of 30, 50, maybe a little bit more than that, things that are actually tradable in terms of decent depth even at retail size on chain and outside of that set either they're not listed because they haven't found ways to build good liquidity yet or they're listed but the spreads are laughably wide. Maybe another mental model here is for the more retail folks in our audience, if I go trade Nvidia on Robinhood or Schwab or traditional brokerage, interactive brokers, not even just that which is one of the most liquid markets but anything of the hundreds if not thousands of different global securities and FX markets and so on that we can list, I barely even think about execution. I won't use the word I too heavily here but let's say I'm a prop shop and I have a 10, 20 million dollar book and I'm executing. I can truly get incredible size on a top 300, 500 asset by volume in the US. I can trade a liquid options chain, et cetera, et cetera. I won't belabor the point but just to say when we think about what's tradable, meaning the liquidity is good enough to not have just absurd execution costs in TradFi, the menu is thousands if not tens of thousands of global markets and so liquid that we barely think, I barely think, you know, maybe I'm betraying myself as not the best day trader but I rarely think about market ordering a million or 10 million of say some of these especially more liquid names like Nvidia and I think a lot of people would join me in that sentiment and that's a better user experience, right? I don't want to have to be worrying all the time about, oh, what's the depth of liquidity? Can it really support my trade? If I enter, can I get out? Maybe it's really thin over the weekend. Do I get to volatile? Am I going to get liquidated? Like these are just not things that cross the mind in the depth of liquidity in traditional markets and that's why it's so important because it's not expensive. It lets users forget about this and just place their trade, know they're going to get a good fill, know they can always get out, et cetera. That's what we're comparing to now. Let's look at, again, crypto native liquidity. It's kind of like starting up a new version of, I don't know, CME or CBOE, like let's say offshore, you know, again, to make the point even more poignant. At the time when I'm starting a competing exchange, I have no liquidity in the books. It's like a social network with no users. There's nothing there and I'm going to have to incentivize and kind of beg and claw and give subsidies my way to getting that first little puddle of liquidity into the books and move up from there. And again, this is why we say all the power and respect to Hyperliquid and others who are innovating in this space, that was a real zero to one moment. It's amazing that any of these things are tradable at all, but they're tradable by virtue or by nature of kind of assembling what went from a puddle to a little bit of a pond of liquidity and maybe even on a similar to like oil has moved closer to a lake, but there's an ocean of liquidity out there in TradFi and that difference is just too huge in magnitude to ignore. So to return to your question, I've spent a lot of time essentially just really hammering home the point that there's just a massive gap between on-chain order books and TradFi and I think this is probably quite intuitive, but the question is how do you bridge that gap? How do you solve that problem? The solution right now that everyone who's building order book based exchanges is, as you said, it's full frontal, right? It's just saying, well, we're just going to compete and we're just going to incentivize and claw and beg and commercial agreement our way to get more and more guys on chain, more and more of this little puddle to a pool, to a pond, to a lake, and eventually we'll do that across the board. And the reality is that's a really uphill battle. It genuinely is because we see global markets also moving towards 24 by seven, including in the US and abroad. We see thousands and tens of thousands of global symbols that we'll have to do this for if we're using this mental model. And I just worry or I question the premise of does it make sense to be rebuilding all of global liquidity step by step, brick by brick, something that I think is going to take decades to really reach parity because people have tried this even in traditional venues and it has taken decades to build up new exchanges and even other Asian countries and beyond. Returning back to what does variational do differently? We just skip the queue. It's that simple. We're not rebuilding liquidity. We're aggregating. We talked about that a little bit at the beginning of the conversation. We uniquely have this aggregation model where our liquidity, as I said at the beginning, is more than a sum of the parts. It's our hedging venues and then we add our own market making system on top of it. But you can think of it as our liquidity is a superset of what we're connected to. The unique major innovation in variational as it relates to RWA is we connected directly to TradFi. And when I say directly to TradFi, I don't mean crypto native dealers that are sitting on a little bit of TradFi liquidity through a retail broker. I mean directly to the largest dealers and hopefully in many cases non-bank and even future bank prime brokerages that underpin these in the traditional markets. And I think that is the structural difference because our liquidity is approximately equivalent to TradFi through that. I mean, there's a lot to do there and I'll pause because this has been a lengthy answer so far, but we can unpack. That's like just the start for us. That's the biggest systemic differences. We're not rebuilding the order books, we're just going straight to where they're most liquid, which is on TradFi and we're bringing them on chain by nature of this broker like model. And one of the biggest differences I think about here is when there's a crypto order that flows through, you can hedge on the venues where price discovery is actually happening. Arguably that's finance, maybe 99% of the time. With real world assets, that price discovery lives on ice and comics or it is in the traditional finance world. But as you mentioned, there is liquidity on chain now. TradeXYZ is a great example with Brent. They do have, I think a lake is a good way to put it, like they have to their credit built a significant amount of liquidity there. Are there cases where you get an order and it does make sense for you to lay that off on chain versus going directly to your broker network? Occasionally is the short answer to your question. I think right now where crypto has an advantage is indeed that product market fit on 24 by seven coverage. So that's the shortest mental model answer to your question is those would be the instances where I think there is interesting on chain liquidity short of a venue being kind of like out of pricing bands or maybe in some sense like presenting an arbitrage and so on during kind of normal market hours and normal participation. The answer is pretty much an explicit no. That's kind of the fundamental point I'm making here is that we're still talking orders and orders and orders of magnitude different, which most people don't see or appreciate yet because for retail size you are executing usually at the top of book and so on. But it does matter when we're talking about how do we aggregate at scale so that we can consistently deliver for any order, any user size and type like a great trading experience. I mean, that depth really does matter as we were just describing with our economy of scale market making system or this principle broker style model. But going one layer deeper, my take, and I like to frequently disclaim it was, I've already said a couple of times, I am a big believer in hyper liquid. I think they'll continue to succeed. Your implicit question is whether we also do work with hyper liquid. We've hedged historically on hyper liquid. I think we'll always keep a good look at that as a potential venue for a variety of reasons. But I do think it's worth us keeping in mind that there are limited structural barriers. They exist, but they're being knocked down right now to many of the trad market participants also moving towards 24 by seven, whether on the interdealer networks, whether on the broker side, meaning like the primes, or even just whether on exchanges themselves up to and including listing a bunch of additional single stocks, which is happening on CME, as I think they announced and a few others. I like to say that there's a little bit of transience to the answer to that question. The answer to that question being mostly in the off hours and the weekend, and we'll see how that evolves over time in terms of whether that price discovery and liquidity there stays in crypto or moves back over towards traditional venues at the risk of beating a dead horse here. You described in the past the rollout of RWA on the Omni platform really in two phases. The first phase being quoting these assets against aggregated crypto native liquidity with platforms like hyper liquid and phase two being hedging directly to traditional institutions through your broker relationships that you've built around the globe. Maybe you can talk a little bit about what really changes there when you go from phase one to phase two. And one of the things that comes to mind for me in particular is when you look at multiple venues and I see this especially in the RWA space, the definition of the perps and what those contracts actually represent can be quite different. And so I can imagine that laying off that risk on multiple venues actually introduces potentially a bit of a basis risk that you may or may not have when dealing with your traditional broker partners. And maybe you can talk a little bit about that, but talk about maybe what are the real big switch that happens as you go from phase one to phase two. I'm going to weasel in a point of clarity here just so we're talking on the same page as it relates broadly to even how Variational connects to TradFi liquidity. I would not describe our connectivity as to brokers. I'd describe it to dealers because that's where the pricing source is actually coming from and that's at the end the basal underpinning layer to the liquidity in TradFi. You of course could face these guys through brokers and we're probably splitting hairs at a certain point depending on how we call certain banking and non-banking institutions. But I do think it's an important distinction to your earlier question on like how maybe some of the crypto native market makers are bringing a bit of the TradFi liquidity through a number of proxies, let's say into the order books. We're going straight to the source. So just mincing some hairs on that one. But I say we're facing more like dealers in TradFi or of that type of counterparty. Primes is another good example longer term. But to answer your question more specifically, when we started with the perps and you asked about basis, that's another one of my favorite questions. When we started with the perp side, what we wanted to show is what our existing model, which had been built out and scaled out in crypto native liquidity aggregation, what it could accomplish for RWAs. In other words, flex our muscles and show, hey, let's match the market at least where it is and in my argument, build a substantially better product than most of the rest of what was available in crypto at the time outside of hyper liquid and even competitive with hyper liquid. You can kind of see this in the numbers we've taken market share. I think we're at 12 or 13 percent of RWA open interest in crypto on the crypto perps by open interest share. Hyper liquid, I think, is at 70 plus. Most of the rest of the guys are at like one or two percent. So this is something we've been really excited and happy about. But the idea there was we want to match the products as they exist on these other platforms. So a few design principles that were important there were number one, we wanted them to be perps and 24 by 7. That is matching what is interesting about some of these other markets. Number two is we wanted it to follow our same way that we've hedged all the crypto flow. Go on to these other platforms, be more than a sum of the parts, offer zero fee trading and a great economy of scale and great aggregated liquidity. But again, matching kind of an instrument set and listings and sophistication of where we could source that liquidity. So that was the starting point. There were some benefits in my view, and the numbers show that in terms of adoption to trading, these RWA perps on our platform, but yeah, they were hamstrung in a variety of ways. Number one is we're heading into crypto native venues, and I just spent a lot of time talking about why that's pretty limited liquidity. So it was always our intention, indeed, to bring this chat file liquidity on as the main kind of way that we hedge. So that's very much what we're doing right now with phase two and swap. So we'll get to that in a second. Number two is funding rates are not great. These perps are not fungible. It's a question I get all the time of why is it hard to build an aggregator? Why is it hard to run this marketing system, this OLP? Because not just in RWA perps, but across the board, funding rates are variable. The definition of index and mark prices are different, especially in RWAs. You have very different ways to determine the index price and the roll schedule across the dated futures that might be being used, et cetera. I won't get too much into the technical details, but they're all very unique, and this is a pretty hard set of quant problems, even that we solve just to aggregate the perps liquidity. The interesting kind of upshot of this is as we move towards the phase that's happening right now, phase two, swaps launch, and more broadly bringing this track file liquidity fully on chain, we've taken great pains to think about what does it look like to align the instrument that retails trading with the hedging venues so that we can make full use of that liquidity and not have to be pricing in protection or repricing around the basis or the difference between all these different underlying, let's say, index and roll schedules. That is fundamentally one of the most important innovations in what we're doing. Innovation one is bringing track file liquidity on chain through this aggregation, but innovation two is having the gumption to say, hey, there's downsides to perps. There's downsides to having a variable funding rate. There's downsides to such a fragmentation of the instrument definitions. What if we redefine a new product category? What if we allow retail to trade a perp-like instrument, but that has a flatter funding rate, in many cases directly flat, that has liquidity that comes from direct from trad fi because it's aligned on the instrument spec basis that makes a lot more sense for our systems to hedge without having to price in all sorts of conversions. That is what we're doing with swaps. I'm excited to jump into swaps. I want to stick with perps just a little bit longer because I think it'll help provide some contrast as we go into swaps in the latter part of the conversation, but I want to stick again with these real world assets and you started to bring up the complications here, particularly around the role schedules because with these perps, you're not tracking spot. I mean, many markets don't even have spot. What is spot WTI, for example? So you're tracking the front month contract that has to roll. And if I look at the documents that you guys have, are your oil, your copper, your natural gas perps, they all reference dated futures. They all have a blended five day roll cycle. And what you have to deal with here, ultimately, when you have that five day roll cycle is how to set that funding rate in the perps. And so when you have an order book, in theory, traders take care of that themselves. They anticipate the price drift that's going to happen in the role. They push the perp either to a premium or a discount depending on what's happening with that price drift and funding adjusts to deal with that price drift. Again, obviously you guys don't have an order book. The perps price is totally something you ultimately end up quoting. Can you walk us through a little bit about how you guys think about setting that funding around the role to remain competitive? Yeah, that's a really interesting question because in our broker like model, you can think of our funding as being more of an aggregate of the effective funding of the venues where we're aggregating liquidity. We've taken a lot of careful steps to construct predictable funding rates and kind of formulas on our platform that allow us to A, keep it attractive to our users and price the right kind of market dynamics. But B, our pricing is a function of the pricing on our hedging venues. Let's just call it the market pricing. And in some of the same dynamics work for perps here. But to your point, we don't need the funding rate to converge the index price to the mark price on our platform because that's naturally happening in the wider market because of the perp liquidity that we're accessing, let's say on a Binance. And again, it goes the same for RWAs and for crypto should be our mental model here is that there's a convergence of the perps pricing in the broader market. So I guess the very reductive answer to your question is we do not in our current model where we're aggregating from crypto native venues, there is a lot of difference between the venues and that makes it hard. And there is a lot of difference in creating a set of rules that are digestible and understandable to traders in our end that kind of bridge this gap and allow OLP to understand funding inflows and outflows to our users and also funding inflows and outflows on all these disparate hedging venues. And that's very much the so called secret sauce and the hard types of problems to solve here to be an aggregator. But it's not particularly different for RWAs than it is on the crypto side of the house in the sense that we try to carefully align those two things such that the net blended open interest based funding that exists on our platform and the way that might have to see quotes and beyond implies certain things about the net fund flows such that our hedging positions would roughly match. So in other words, the most reductive answer to your question is when we're talking about perps, whether on RWAs or on crypto and kind of these crypto native style hedging venues, ours look more kind of like a sum of the parts. So you can kind of observe this in the data. We in some sense push that complexity down to be solved in sometimes different ways by the team at say XYZ or by the team at say a Binance and beyond. And we don't comment intentionally in specific detail about exactly where all of our hedging flows go in and so on for obvious reasons. You can imagine just as two examples that these are teams that are trying to solve that problem in slightly different ways and then we have to do a lot of math on our end to back that out and then we have our own funding mechanisms on our platform, but they don't necessarily at least for these RWA perps relate to yet a third definition of say role schedules and so on that exists in our platform. You can imagine it's kind of that sum of the parts. We've alluded earlier in this conversation that a lot of traditional markets are starting to move 24 seven, but they're not there yet. And when you're talking about an audience of retail crypto native traders who are used to crypto markets that are trading 24 seven, a lot of times they do want to trade these real world assets 24 seven. And so we're in this bridging period, I would call it where the expectation is a lot of these traditional markets, the liquidity for 24 seven will come in the next couple of years, but it's not there yet. And you have to make these design decisions. So if I again go look at the documents and specifications that you guys have laid out, you have all these different index change modes that happen when the underlying markets close. So you have exponential weighted move it averages when to smooth overnight prices on weekends, equities freeze at the last value while commodities switch to a price derived, I believe from your own order book is my understanding. Maybe you can talk a little bit about those design decisions and where you are making differences between like why do equities behave one way versus commodities behave another and the different treatment that they get. We're taking a lot of inspiration from how these things are handled on other venues. It's funny, I think Edward is the best guy to be asking about the nitty gritty about some of these things. So I'll direct as a blanket, both to the previous piece and to the future piece, sanity check me versus our own docs. I'll give you the broad strokes of it, but for anyone taking this as gospel, but in short, it all relates to kind of the same idea, which is when we're constructing these markets, and we're also looking at how our hedging venues are constructing these markets. When we're talking about crypto native perps, what does it look like if the index price or like the underlying, let's say it doesn't have a marketable market, right? Like over the weekend in the case of equities and so on, how does that bound price discovery for the perp? How does it bound like the gradations of the funding and how much basis you allow in terms of slack before it really starts kicking in and how do you protect these retail traders? But I think it's all fundamentally actually in the same service of a deeper question, which is, it's really hard to map perps onto these markets. Again, I think XYZ did a lot of innovation here, Binance did a lot of innovation here. There's been different models tried and there's other platforms too. But when you end up trying to make something 24 by seven that doesn't have good spot market price discovery over the weekend, or we end up with things that can be incredibly volatile and thin books and lead to cascading liquidations, or I mean, some of these times in the early generations of these products before some of these things were narrowed in, you know, we'd see funding rates go to thousands and tens of thousands of percent over the weekend. And sometimes this was an intentional trading strategy by some individuals that we saw. Slowly we kind of moved into just like there was a little bit of iteration with crypto perps like, okay, what if we want to put the funding rate on some type of moving average or EMA? What if we want to smooth these things? Or what if we want to put min and max bounds to avoid people playing too much of games or benefiting too much from just directly pushing around the price and delays and all sorts of other things. They're all in the same service of solving that disconnect problem, which is that it's actually really hard to keep this synthetic, right, like a perp track to an underlying that in some sense becomes itself either fixed and static. I think QFX and a few others did that versus kind of an EMA of the last price. Hyperliquid was experimenting with this even since the early days of pre-market perps, first for crypto projects and then also the pre IPO perps, this kind of fully synthetic index price underlying price discovery. So I think there's downsides to all of them. There's upsides to a few of the designs in terms of the market has really been efficient and iterating and settling out. And again, that is a big shout out to Hyperliquid and some of the other teams innovating there. But I just want to make the point again, we kind of stand on the shoulders of giants when it comes to our broker, like our aggregator model, because we're not trying to be the price discovery venue because we're not trying to necessarily set the market price per se for these things, meaning in the broader sense of the market, we can take a lot of cues and carefully designing our systems around funding and swan more to look like how do we align them with the aggregate or the average of all these venues such that OLP can hedge on all the disparate venues with their disparate rules. And yet we can provide a pretty predictable or understandable funding rate definition to our users, but also not necessarily take a bunch of basis risk or loss when we might have big differences that emerge between the different hedging venues. So that's kind of how we think about it. So let's talk about this new instrument that you are introducing, what you guys are calling swaps. As I understand it, it's a price return vehicle plus a financing leg, cash settled, close cousin in traditional markets might be sort of like the retail CFD or the institutional total return swap. Maybe you can talk a little about what's the same versus what's different and novel here, both with respect to perps, with respect to maybe the CFD, with respect to a total return swap. Maybe you can compare against those models. And maybe most importantly, like what are you trying to achieve in introducing this new instrument? Yeah. Let's start with the last. What are we trying to achieve? Because I was alluding to it before when we were going kind of all through the complexities of perps and basis and how do you define for, you know, paused market, like what the funding rate should be in swan. This is the problem that no one talks about with perps. And I think it's actually a big problem that we shouldn't ignore, especially for retail traders. The complexity and unpredictability of the funding rate of the basis. If I'm a retail trader and I want to long Nvidia with some leverage, whether or short even I want to take a Delta one position, perps are great and they were popularized in crypto versus options and more complex instruments specifically because they're supposed to be simple. They're supposed to give you that leverage. They're supposed to give you that one click execution and understanding of the underlying. But now for RWAs, especially, you don't have to worry about, I don't know if I want to hold that position over the weekend. The basis could go crazy. The funding rate could go crazy. I don't know if I want to hold that position for a few weeks because a lot of these platforms don't even pay me dividends. Am I really benefiting from using this? What I now see is kind of a weird derivative versus holding the underlying. Not quite as clear. So these are the problems with perps, not to mention the fact that, again, for all the reasons we described, they are unique instrument versus how you could hedge these in TradFi. And our goal is to hedge on the TradFi, therefore we want to align the two. So a few observations to answer your question directly. We're solving the problem primarily of funding rate and secondarily of being able to hedge directly against TradFi derivatives. I would compare the most to a total return swap because our goal is to approximate the return of the underlying as exactly as possible, including passing through dividends, which is a very key point here compared to most perps that we see in the market, but also just broadly like all other types of cash flows and corporate actions. So really it's a synthetic with a funding leg, a carry cost that allows us to embed that leverage. And then everything else we want to track as closely to the true, again, basic mental model of retail trading is the true underlying. When I think about a Tesla perp or a Tesla swap or whatever as a retail, they don't even think or want to get into the details of the financial definitions of the instruments. They just want to essentially be trading quote unquote Tesla. But now they have the benefit of leverage or the benefit of being able to short with a USDC balance collateralizing that. So that's the problem space. So going one layer further, you asked about TRSs versus CFDs versus perps. So versus a perp, they're very different. This system does not have funding and that's not how it tracks the underlying. And we can kind of get into that if it's helpful, just what these types of OTC derivatives look like. But the whole idea is these are not exchange traded or exchange cleared instruments at all. A CFD or a TRS uniquely to our broker model always trades against a dealer riding the other side. So in that case, on chain, the dealer is us, so to speak. It's our OLP system. And then off chain, as we discussed, it's we're hedging that against other large traditional trad fi counterparties. This is the moat. That is the systemic innovation. That is the unique advantage. No one else can pull it off. This isn't something you can do on an order book. But what is the upshot to the user? Well, that funding lag is fixed. It's just a cost of carry, a cost of capital in some sense. We expect it to be around 100 bps to SOFR plus give me a little rope on that depending on exactly where our economics lie as we are rolling it out in the next few days. But suffice to say, it's flat. That's the biggest difference. It's not a synthetic that uses the funding rate to track the underlying index price. It's just priced based on the underlying, trying to pass through the direct exposure of the cash flows as much as possible. And we're essentially just charging for the margin. I would go so far as to say that solves in almost every respect as a super set, meaning strictly better than a perp, right? I don't have variability in a funding rate and I can't get screwed on that. I have predictability. In many cases, this spread to SOFR, in other words, the USD borrow rate, is going to be substantially lower than what we usually see as the equilibrium rate set for perps, like more in the 7% to 8% to 10% per annum. This is probably going to be in that 4.5% to 5% range. And then finally, again, it's just the connectivity. We spend a lot of time talking about definitional complexity and basis and so on. Total return swaps are, I would say, one of the most, if not the most liquid way to trade linear derivatives. And there's a good reason why hedge funds and others are trading this. It's the ESOS I said at the beginning of the conversation. This is our institution's trade. We're trying to bring it to retail traders and just to the broader on-chain market as a whole. Total return swaps are one of the best ways to trade liquidly and with leverage in TradFi. And we can just bring that instrument directly on-chain. So let me pause there. I'm also happy to talk about the differences in that and CFDs and so on. But that's the meat and potatoes of it, at least. I think this next question might be redundant to potentially unnecessary, given that you said you thought maybe swaps were superior to perps in almost all ways. But my understanding in prior conversations with you is that when people search for, say, NVIDIA on your platform, they will see an NVIDIA perp and an NVIDIA swap. And the NVIDIA perp is going to have variable funding and be 24-7. The NVIDIA swap is going to have a fixed carry leg with it. And at least I believe traditionally, we'll only trade traditional market hours. Correct me if I'm wrong there, if that has changed. How should a trader really decide between when to use those? And maybe a second part of that question, if I can ask a two-parter, is do you just expect to eventually sunset the perps and everything would just be the swaps over time? Really important question when it relates to what we're doing with the rollout right now. And I think, frankly, when this is going to be education, like for a while, these swaps are going to be new and weird and everyone's used to trading perps and understands it. I think it's going to be part of our interesting battle and work over the next few weeks and months to educate what are the benefits of swaps, what are the downsides or risks, honestly, I would say very little versus what we're traditionally defining as a perp in RWAs right now. But again, you know, that's going to be fun for us as education and moving the volume over. Yes, we will list both in the near term and they will continue. Users should trade perps if they specifically want exposure to the funding rate or in the near term if they want the ability to trade at 24 by 7. As we all know, the 24 by 7 essentially comes, my whole argument is at the expense of unpredictability and the funding rate or unpredictability, at the expense of perhaps much thinner liquidity, crypto native style liquidity, and therefore getting pushed around over the weekends. Maybe if I did trade over the weekends, I'm having much more limited liquidity by definition on any of these platforms, ourselves included, but also XYZ or Binance and beyond. There's pros and cons. If you're a little bit less cost sensitive to the entry or to the unpredictability of the funding or particularly your basis trade or others, and you want the funding more as part of your broader strategy, those are some reasons. Swaps, the downside, so to speak, is as you pointed out, a limited market hours. And this is definitional for us. We are hedging these exclusively and into TrapFi and that's very much part of the definition of the product. We think that's orders of magnitude, frankly, more exciting when it comes to the depth of liquidity we can show, the hundreds and soon to be thousands of different markets we can show all with TrapFi grade liquidity. Not bootstrapping one new one this week, one new that week, slowly and rebuilding from scratch the liquidity, but just as we turn it on, boom, right to the top of the charts in terms of the quality of execution. The thing that I can't comment too much on timeline wise yet, but let's just say we publicly described it as phase three and is very much coming. It is not my expectation that swaps will remain closed market hours on at least a lot of instruments because as I alluded to, traditional finance itself is rocketing towards 24 by seven trading. And our goal is always to aggregate from as much TrapFi liquidity as we can, including other types of dealers and venues and ATS and ECN and what have you, that will be 24 by seven in the near term. So I foresee a near term future where even that last question that might push someone towards a perp versus a swap, meaning they specifically want to be trading on the weekends, for example, is rapidly going to diminish. And at that point, I don't want to publicly commit that we would be sunsetting per se, but I think there is a world in which we eventually look to kind of merge these definitions or at least make a clear preference towards most flow moving into swaps because I think there's very few reasons to trade a perp on our platform, at least at that point, we'll see what the market has in store. Our goal is always just to list as many things as possible that people want to trade. And I could see also a world where for basis and more sophisticated traders and a lot of other reasons that we can go into, people would want to continue trading perps alongside. Can you talk a little bit more about where that financing rate in the swaps is going to come from? When I think about entering into a total return swap on the institutional side, I can often get a fixed financing rate because there's a known term to that trade. I don't want to put words in your mouth, but it sounds like with the swap on your platform, there doesn't have to be a defined term. So is this actually going to be a floating rate, floating above SOFR? How's it going to differ between different assets and different traders when they enter and exit the position? Can you just talk a little bit about how people trading these swaps should expect that carry leg to evolve? I say it's predictable or flat in relation to a perp, and that's very much varying. In real time, as the market changes, the definition of perp funding leg is changing. That's very much the idea. Here it's flat and predictable, but not necessarily fixed, meaning for an indefinite term. The short answer to your question is it's a spread to SOFR. The main driving factor that would change it in rates across literally everything else in our lives, including our bank accounts and what have you, is the U.S. Fed policy, essentially. I'll put an asterisk there for political reasons, but let's just say in that vein, right, in that ballpark. And this is the same way in which we could say a checking account has a fixed deal or something like that. It's not variable with market rate and changing on a day-to-day basis, but of course, yes, as literally global monetary policy adjusts, there does adjust a cost of dollars, like a cost of margin or a cost of capital. So that's the short answer to your question. The long answer to your question is there are some asterisks here, as I think you anticipated or point out, depending on specific things like overnight holds and rolls and things like that. So the vast majority, I would expect, of swaps will be just that, it's a financing leg to SOFR with a spread, but there's a few where there might be, and these usually are very small in magnitude. Again, compared to PERP funding rates and things like that, there might be a little bit of slack built in to handle some complexities of certain types of ways that global commodities markets trade, for example, and a couple others. But when we're talking about the design space here with retail traders as the audience, we're looking for predictability flat, meaning cheapest cost of carry they can get. These are still servicing both quite well. They're still predictable and near flat, and they're also still probably quite definitely the cheapest cost of margin or carry or leverage that they have access to. So that's kind of how we're thinking about it, but you're essentially thinking in the exact correct vein about how it will work in practice. I want to push into this financing rate a little bit further, and just as an example from my own personal life, I got quoted on some single name total return swaps earlier this year in January in U.S. listed equities and got quoted on some names SOFR plus 100. More recently, getting quoted on those same names, it was closer to SOFR plus 300, and the conversation was really just about balance sheet availability, what was happening with, especially around the SpaceX IPO with banks, and then we just have seen this huge demand for leverage in the U.S. with levered ETFs, particularly in some of these single names that has just, again, made it very costly. My question is, would you expect to see somewhat the same? Now again, it went from SOFR plus 100 to SOFR plus 300, nowhere close to the rates that we have often seen within perps, but still a decent amount of variability intra-year within that spread. Is that a similar expectation that you would expect to see on some of the single names of popular single names like a Tesla or an NVIDIA for the swaps that you're gonna offer? In short, my expectation would be no. We do have certain mechanisms by which, for anomalous reasons in the global markets, there might be specific instances where the cost of dollars is going up rapidly, and again, this is probably a little bit less so on the single name U.S. and a little bit more so on things like commodities and others that have a pretty complex financing stack. With a little bit of rope given here for possible variability, our goal is to always be the cheapest possible borrow across the global markets. I expect that to be true of swaps for pretty much everything, but I do think, just to be clear, the way in which we've structured and negotiated some of the commercial arrangements with the trad5 firms where we're hedging these things do involve some limitations and some understandings on what we would expect are, in many ways, you could call it an economy of scale to buy us in terms of what these spreads we're paying to SOFR are and so on. So this is, again, going back to the top, like one of the beauties of the size and the economy of scale we built so far in the platform and just gets bigger over time. You mentioned you yourself getting quoted, and with all respect, you're building an awesome business, but there's a big difference between you and the quotes you might receive at SOFR plus 100 or plus 300 versus Bridgewater or Millennium versus their prime. Thinking with that in mind and the ability they might be able to, A, negotiate it to be a lot lower, and then, B, negotiate it to be a lot flatter because at the end of the day, these are markups on balance sheet availability from banks and primes and dealers and so on. So I think that's the mental model I'd apply here, and I would expect, although, again, with a little bit of an asterisk that at the end of the day, there are certain fail-safes with these agreements just due to the nature of markets changing, but I would expect ours to be more in the range of not changing too much, and that's our expectation and something we'll be working towards. Now, one of the exciting announcements that you made recently is that you have signed over a billion dollars in open interest capacity with TradFight dealers. I have two questions related to that. The first is, does the capacity sit behind the swaps alone, or does that help with the perps hedging as well on the platform? And then second, as it relates to the different asset classes, can you talk a little bit about where that capacity is most available that you're bringing online? Yeah, so a few points. Number one is that, funnily enough, a billion dollars is just us getting started. This is a first tranche of commercial agreements we did with some early partners. We have more that are in discussions, more that are rolling out, not just to cover hundreds and in the future thousands of assets, but also to expand into different global markets, which is something we're very excited about with swaps. So there's two parts to that question. One is about the open interest capacity, but I just want to take the opportunity to point out that we've had amazing success in getting our model, which again is meeting TradFight where they are, integrating directly with them, arranging commercials between our hybrid off-chain piece in this case and their existing desks. That's really why we can make it work. That's how we bring liquidity in and that's just us getting started. Talking amongst that amount of capacity, yes, we can use it to help perps. There are some nuances there, as you can imagine, like there's now a basis between the definition of the perp and how we might be trading on a swap or a CFD or TRS, et cetera. Secondarily, there's also a question of if we have limited capacity, where do we want to put it and use it? We so far have experimented with using some TradFight liquidity in our perps. We've been doing that as a so-called phase 1.5, you might've seen us announce. But I think we'll be very focused on using TradFight liquidity as much as possible. I know the perps continuing to expand the set of connected liquidity there, but more likely using it sparingly as a way to supplement occasionally. But we want TradFight quality liquidity to primarily be going to that well-aligned instrument because that's where we get the biggest boost to our user experience. In terms of instrument market definitions, yes, like for a variety of reasons with our platform, you could expect the same of any market maker and dealer and so on in TradFight. I would say we don't want to necessarily rip a billion dollars of open interest on just one instrument amongst like the whole set that we might be trading. And there are going to be certain fail-safes there as with any relationship. But again, as with any trading relationship, these are all quite flexible and we're building that understanding, that economy of scale, those relationships in real time. We already have a fantastic set of partners and we're moving very quickly to add more. So as we diversify across more partners, as we bring on more global markets, and as we even just bring on size and balance sheet on our own market making system and negotiate up, this is just a starting point both for instrument selection, for open interest caps and beyond. Well, we mentioned that more so to say, I'm actually quite proud of that, that we've been able to negotiate that much capacity so early on for an as of yet unproven and unreleased product category. It shows the faith that these partners have that swaps are really going to take over and replace perps very quickly. And secondarily, it does go to your question on economy of scale for the financing lag and so on. We're already being treated as a pretty large participant in these markets and we expect to grow rapidly. I've heard you say that you have the expectation that on-chain real world asset exposure will eventually outstrip all cumulative crypto exposure. Say if we look at your platform, the OLP is running internal firm capital. You just announced the billion dollar open interest capacity, but that is a finite amount of capital at the moment. Maybe you can talk a little bit about what do you see as being the binding constraints from getting from where we are today to this longer term vision of real world assets being the dominant exposure on-chain and variational's ability to support that? Is it dealer capacity? Is it regulations? Is it your own internal capital to be able to be the counterparty to these trades or is it something else entirely? How does this get solved? I'd put a few in no particular order, but I'll let tier one and I would start with capital. We do not rehypothecate user balances. We've been very clear and you can see this structurally because they're all isolated in those independent smart contracts. We call them settlement pools or essentially escrow contracts. We have a balance sheet for OLP that's entirely separate, that has to collateralize all these positions. As with any market maker, if we've kind of studied this in traditional finance or a broker dealer, balance sheet is a big part of the equation. We've seen a huge outpouring of interest and support in continuing to capitalize OLP. We obviously have our own large equity balance sheet that is growing every day and is also used to support this. This, I think, is an increasingly solvable problem with scale. So it's a great flywheel. In other words, it's a great problem to have when we're asking the question, hey, we used up a billion dollars of capacity or we used up a hundred million dollars worth of margin for these products. What's next? It becomes increasingly easier to underwrite that because it's a bigger system. It's a bigger balance sheet. It has more flow. It has more profitability for everyone involved. I'd actually argue that that becomes more solvable and solvable over time, but it's important when I call it because it's a moat. I mean, it's not trivial to access billions of dollars in open interest capacity with dealers. I mean, it's not trivial or I would argue possible at all for our competitors to connect to these guys in the first place. We already talked about swaps don't even trade on exchanges, et cetera. So we also consider it a big part of our moat, this idea of balance sheet being necessary for running this business. Number two is connectivity. I like to call this a solved problem and that we've shown that this system works. We've connected it. We will be seeing the beauty of this as swaps roll out and give us orders of magnitude improvement versus what's on markets right now and on-chain RWA trading. I think my goal isn't to stop there. I want tens of thousands of listings. We have some of the best and most liquid partners in the world to get some of those classic indices and G10FX and U.S. single stocks and so on on swaps, but I want the weird stuff. I want everything. So that's the next gating factor is how do we take this global and there's already progress there that I hope to be able to share publicly soon, but I want markets in Asia. I want markets in EMEA. I want not just G10FX, but a hundred different pairs and beyond. There's a lot of really interesting directions to take it and that's the commercial problem. But again, coming at it from a position of strength, showing that this works, frankly solving the hardest problems first, which is going straight to tier one guys in the U.S. and beyond and getting them to work on the biggest markets in the world. We can fill in the blanks from there. Number three is the commercial side. Indeed, there's risk limits and there's everyone getting comfortable and part of these limits are balance sheet. And again, we talked about open interest and so on, but part is on both sides, our own platform included just crawling and walking and running. I expect swaps to go very quickly. We've done a lot of careful testing. We will be rolling them out quite aggressively, but at the risk of stating the obvious, it's always smart to grow things at the very least, let's say on a curve and not zero to one step function. It's as much true on our own platform, our own risk limits and open interest checks and so on, as it is going to be on some of our partners that we will be scaling these over time. But very shortly as this becomes first like a large book of business and a billion dollar book of business and much beyond, those become easier and easier because the pathways are just greased and well understood. We have 500 million in RWA perps right now. I foresee a lot of that moving towards swaps in the near term. We have 1.5 billion on our platform as a whole, and I think it will continue to grow in the RWA side particularly. But more broadly, the on chain space probably has around 10 billion or beyond of open interest that we should be addressing. I think we can rapidly take that in many orders of magnitude of that if you comp it to retail brokerages. So that's really long term where we're going and we want to go after the brokers globally with this new platform. We've spent this entire conversation talking about the retail platform you alluded to at the beginning. That's not where the end of the ambitions are, right? You have these increasing ambitions for the institutional platform, and I do want to give that a little bit of breath. But with this idea that a lot of what you seem to be doing is bringing institutional ideas downstream and accessible to retail, maybe you can talk a little bit about what innovations would this ultimately deliver to institutions who do have access to swaps already predominantly. What ultimately is the service or feature that you would be delivering, the benefit that you would deliver to institutions through this platform? Yes, we have to separate this between Omni and between Pro. So when you're asking about the benefits to institutions trading on Omni or potentially trading swaps, they are still huge with the main one realization that yes, institutions have access, the largest institutions I should say, that are primarily US based have access to swaps and other types of these Delta One OTC products, but the vast majority of still what I call institutional participants don't, right? I'm a family office in Malaysia and Indonesia, I'm a mid-sized hedge fund, frankly, even in the US that just is having trouble facing a tier one prime and getting it is done beyond. There is a lot of red tape here. And that's really one of the ethoses of crypto as a whole when it comes to like, bringing great ideas from institutional and traditional finance on chain is it benefits a lot of institutions on a global basis on a mid pack and so on basis and beyond, simply because yes, if you're the biggest of the big Millennium or Citadel and so on facing Morgan Stanley Prime, yeah, you have access to absolutely everything we're doing and beyond. If you're retail, absolutely not. And that's where we're structurally bringing the menus and the benefits to retail. If you're a global market participant, if you're a mid pack size hedge fund, if you're a prop shop, if you're a family office, you know, not necessarily I think is the right reading. That's interesting, even just on the Omni side and on swap side. And that's a huge market because we haven't even talked about corporates, by the way, a different type of institution, non financial institutions. People might need hedging flows for oil and this and that. Do you want to do that on a platform that's crypto native, has dubious execution depth and quality and this variable basis and funding rate that I can't predict? No. If I have a predictable cost of carry on something, it looks like a swap and I'm used to understanding that instrument in the first place. Now that's quite viable. I actually could be thinking about collateralizing with USDC on chain. I'm very comfortable with that. I'm a corporate on like Indonesia or Malaysia or Singapore or something. And now I'm thinking about hedging my cross currency exposure, my oil exposure. Like that's a fascinating use case of these instruments that I think is not well priced and they are absolutely not of a size, sophistication or geography where they can be trading swaps versus a tier one counterparty in the US. That's the democratization factor on that front. On pro, I'll just give you the short version of it, which is twofold. One, it's mostly going to be options, structured products and other types of OTC derivatives. There probably will be swap or swap like instruments, especially on the Delta exchange side of an options trade. You could call those a dated future at that point, though, because we have dates to work with. But regardless, it's mostly focused on institutional OTC trading, which I see as a nominee or products as the main focus. But then two, the corollary is then we're delivering a completely different type of value entirely. This is a multi-dealer platform to handle the lifecycle of trading, settling and clearing. It's not liquidity aggregation. Multi-dealer means the other side to your options or structured product trade would be another firm on Optiver or in crypto native like a QCP or an orbit markets or maybe even a Wintermeter or a Salini, quoting you say an option or structured product. But what we're providing there is bringing the whole flow on chain, not calling each other to get a quote, not doing price discovery on Telegram, not paying ridiculous fees to intermediate agents and not having ops teams in the global south, moving funds around with email confirmations. It's laughably antiquated. And I like comparing it as I did at the beginning of the chat to those reasons why stable coins are replacing Fedwire and Swift and current FX flows. I mean, they're just orders of magnitude better if you can really bring it on chain and systematize it. But that's a different type of value proposition for pro. Lukas, this has been a great conversation and I want to end it with the same question I've been asking all my guests for almost the last couple of years now, which is outside of work, what is something that you are just currently obsessed with? It could be an idea, it could be book, it could be music, it could be some form of media, it could be a concept, an activity. What is just something that's totally captured your imagination? Can I give you two? Absolutely. One, which is a half answer because it's a little bit work adjacent, but power and compute markets are so important right now with what's happening in AI and data centers and everything else. And I mean, there's a lot of overlaps with that to the day job for sure. In fact, I would see us probably making some moves in both markets pretty soon for Omni and even for pro longer term. But I just find these fascinating, like in a genuine personal education sense, I've been looking at various types of power deals and projects and I've been closely following that space. I think this is one of those moments in kind of capital formation and a new type of product stack and force flow under a business. And it's just so fascinating to watch, right? We're watching kind of a re-commodification of a new type of commodity. That's a fascinating thing to think about. It might be the most like common answer, right, is AI, power, compute, et cetera. That's one side. The other side is my research background is actually robotics for many, many years. I did everything from like computational physics to train robots and sim to sim to real and transfer learning to kind of everything in between, culminating some time at Google X working on some of their robotics efforts. I continue to keep very close to my research colleagues from back in the day, both in industry and academia. Back when I was in labs at Columbia, they've now become market leaders. These guys who I used to be working with when they were PhD students and so on are now running labs and running teams. And that is an area that I'm watching. I like to liken it to self-driving cars. I think people have gotten a little bit over their skis in terms of thinking about how far we are along to productionizing these things. I remember in the early 2010s when everyone was like, every car is going to be driverless by 2020, right? And then we had a huge shaking out. Most of these guys died, meaning like Cruz and a few others, Waymo, of course, is doing pretty well. So I think we're going to go through some booms and busts and some cycles, but I think within our lifetime and probably still in the very near term, a grand scheme of things, we're going to have general purpose robots just everywhere in all parts of the economy. And that's personally absolutely fascinating to me and certainly something I still spend a lot of time thinking about and looking at. So for you or for any of the readers, and I know there's some great overlap with crypto audiences as well in robotics, but I find that to be one of the most interesting and kind of defining new pickup, say, verticals of our time. Exciting times. Lucas, thank you so much for joining me. I really appreciate it. Thanks for having me. It was a great conversation.