Andre Cronje: The Biggest DeFi Boom Is Yet To Come (Entire Thesis)

[00:00:00]We are live with Andre Cronier, uh, true DeFi OG in the building with us here today. It's not his first time. You guys know him well. I'm sure you guys have seen his, uh, products on chain before.

[00:00:11]Probably deposited into one if not a few of them.

[00:00:16]Sonic, list goes on. Today, we're going to be talking about Flying Tulip. Uh, as I mentioned earlier, you know, very unique launch strategy here. Uh, we, uh, fundraising strategy as well. So, we're going to get into some of that. This is our weekly stablecoin show powered by FRA and so FRA is doing great work.

[00:00:35]Genius compatible stablecoin open stable borderless. Uh Andre, man, it's a pleasure to welcome you back to the rollup.

[00:00:42]>> Likewise. Nice to be back.

[00:00:45]>> Absolutely.

[00:00:46]>> Don't think it that long ago I was on here or at least it doesn't feel that long ago.

[00:00:49]>> I think it was just a few months actually. What uh what have you been up to?

[00:00:53]um roll out mostly, you know, trying to get products online as as quickly as security allows and hopefully as fast as investors want, which is always a a interesting tug and pull game, I think. Um >> do you uh do you find yourself wanting to ship faster than investors want you to? No, no, no. I mean, investors always want everything live yesterday. Um, security is always the biggest constraint, you know, because it's audits is one thing. Um, I I think while audits are very important, they're a little bit overplayed in the space and not not overplayed. too much emphasis is put on only audits and then a lack of you know infra security, key security, um secondary order securities, things like circuit breakers or outflow monitoring.

[00:01:59]Um and those things take time you know like um capped launches, slow approaches and it it also creates user friction which in itself isn't nice >> because you end up limiting what your users can do at any given point in time and that also you know prevents scaling.

[00:02:16]>> Yeah. Yeah. I just saw there was a recent uh I think it was a vulnerability on OIUM the perex and $18 million left the protocol like that and people are calling Yeah. Yeah. That's why circuit breakers exist. they're a feature, not a bug. Um, and so, uh, >> so historically, circuit breakers didn't really make sense because there were only two interacting parties.

[00:02:41]It was the depositor and the smart contract. There there was nothing else.

[00:02:45]There was no multi-IG. There was no, you know, off-chain counterparty.

[00:02:52]Um, so old we we've inherited a lot of the design principles of old DeFi into this new category where we are now, which which is closer to onchain companies than it is decentralized finance. Um, so you know, back then if you had a smart contract and someone could not have instantaneous liquidity, you you'd be dead in the water right off the bat.

[00:03:27]>> Nowadays that you have, you know, a line of communication, a support channel, people you can reach out to. I I I think the assumptions are very different. And circuit breakers now make sense as well, right? Because circuit breakers are there And and our our system is also multiple layers of circuit breakers. And I've seen the friction it causes for users, you know, cuz a user might be used to or or or the fear it can cause, right? Cuz cuz using our circuit breaker as an example, we when you request the funds out, it goes into a queue and then you can claim it in 6 hours. But now myself sometimes while using the system I'll I'll claim it and I'll go check on epher scan and I'll see oh it didn't go to my wallet it went to zero xcb which is our server I've seen this >> but you know in my mind I'm like holy crap what went wrong meanwhile you know it's just sitting there waiting to be released so that I can claim it in a few hours but but something as simplistic as that you know causes fear and panic from a user's perspective.

[00:04:31]>> Yeah. Yeah. And hopefully we can solve some of that with UX developments, right? And a lot of this is you know going to be obfuscated underneath the hood. And you know the the I think a lot of one of the building principles that I I've heard from developers as of recently has been look we're building for the next cohort of users not necessarily the cryptonative ones that have been here and are the most sophisticated. you know, we're okay with sort of showing something on the UX side, right? And and on underneath, right? we know that, you know, it's in the circuit breaker. It's going to get claimed, but we're going to show the number as if it's, you know, still with the user because ultimately it will be and sort of >> maybe take some shortcuts that, you know, back in, you know, the the more the the DeFi purist days wouldn't wouldn't have really been uh wi within the realm of what people would have been accepting.

[00:05:23]>> Yeah. I mean, back then, you know, if again, if your counterparty is just a smart contract, I think it's not it's not too much for the user to expect that they have instant liquidity and instant access. Um, that's a fair ask if the only thing I can talk to is the smart contract, you know, cuz I can't cue a transaction in the smart contract and then ask someone to expedite the queue for me because in some cases, you do need to do that. So I look I I think the the old age fight between security and user experience is as old as time.

[00:06:03]You know I mean it comes down to you can have a password of one two three which from a user experience side is more convenient but much less secure. So what did they start doing? They started enforcing more and more complex passwords. Now you need at least eight characters and you know alpha numeric capitalization and some symbols. That's a lot harder to remember but it enhances the security.

[00:06:35]So so security and UX will always fight with each other.

[00:06:40]>> Yeah. I think this is just a part of the space maturing where we're starting to realize, look, we have to start including these small UX frustrations that ultimately provide a better order of magnitude in security.

[00:06:59]>> Mhm. Especially as and and and this is again a huge difference between you know sort of 2020 2021 or even earlier as early as 2018 decentralized finance which was the goal was pure immutability and pure decentralization.

[00:07:18]>> Right nowadays I I'm I don't even preach that anymore.

[00:07:23]If if if if I work with a new team and they ask me if they should have their contracts upgradable or immutable, I always tell them upgradeable. I I think it's non-negotiable nowadays. You have to be able to upgrade your code. But that immediately means you have the biggest threat vector known to your entire protocol. Cuz if if if >> if if you have one developer key that's sitting on some guy's PC and he's able to unilaterally do a upgrade call on that contract, that can steal all of the money. So now you already have to make sure there are and and this is why I say, you know, smart contract audits aren't what they used to be. they're they're still majorly required but the best audits in the world won't capture that vector >> you know so you need >> traditional security in front now as well to make sure hey the the the key is properly offsite and you know probably some HSM signing or ideally the deployer key should be immediately handed over to you know a time lock and a multisc um but even therein is now again a different a different trade-off of consideration that you need to make because if if the contracts are fresh and you just upgraded it, you might you might not want the time lock because you might want you might need to be able to respond within a fairly short time window. Now, if something goes wrong, you can't wait 72 hours to do an upgrade or a freezer or something. So, >> so your your your contracts can't just be a single admin role and that admin is a time lock and that time lock is in a multisc. So using ours as an example, anything that moves money sits behind time lock and sits behind a multic.

[00:09:07]Anything that can that can momentarily pause or delay. So something that can either drag out the circuit breaker a little slower or pause funds outflow that can't sit behind the time lock. you know that we we we need to do in collaboration with and I mean there's some great teams out there that were also trying to still on board um Hypernative Seal 911 obviously um so you you know now all of a sudden from the base code level you need to start thinking about this role separation and how how these separate roles can sit in their separate areas.

[00:09:46]>> Yeah. Yeah. I I think you know this is one fantastic observation given that you've been you know in the space you've worked on multiple projects you've you've got the benefit of hindsight and uh and learning through the iterations of some of these things and you know there there's been several iterations that this space has gone through you know philosophically which calls for different design choices both on the on the security front as well as on the UX front and as you said these things are are constantly at odds with one another as we you We'll we'll get into Flying Tulip and and kind of the nuts and bolts there and and you know all all the updates, but I'm just curious before we dig in deeper. You know, some general observations that you've had. You mentioned auditability. You mentioned sort of this dichconomy, security, and UX. You know, some of some of your projects, you know, you you've you've worked on they're they're continuing to run. Um what have been the lessons like you know what are what are some of the the observations you know the general shifts and evolutions that the space has gone through that you're now putting into pro in in into practice with flying tulip you know in addition to auditability and security what else have you noticed about the space that's changed since you know the the the good good old days if you will >> um I I I think the most important Second one is decentralization is dead. Um we're no longer striving towards decentralization.

[00:11:14]If if if nowadays and and and and truth be told, I I I think this shift even started happening around probably towards the end of 2023, maybe even 2024 cuz I some year's fair launch which you know the the the so-called immaculate conception like like from the get-go I designed it as a mechanism of decentralization which is why you know I didn't have tokens everything was given away to the protocol users there was no team there was no foundation there was none of those allocations the goal there was decentralization right it's it's the it it it was trying to as closely copy the Bitcoin playbook where the original founder leaves and a community has to take it over. Um that today that today people are won't be interested in. People won't want that.

[00:12:24]They they need someone behind it that keeps driving value because the the tokens have become almost too valuable to the point where the the person holding the token. the and and this is also just an evolution of the of the user base interacting with blockchain and crypto as a whole cuz back then the majority of users were you know technocrats or at least technically educated people and they understood the token's value beyond it as an investment vehicle and I think nowadays everyone sees it first as the investment vehicle And second as a utility based something or anything like that. Um another example of immutability being dead you know is is something like um the decentralized um keeper network that we built originally for urine and [snorts] maker and a bunch of the stuff that was using it back then. that was again moving towards a decentralization and b immutability in that it was designed as a marketplace for unknown actors to be able to work jobs and provide facilities on chain. Uh again nowadays you know everyone just does their own infra offchain. They run their own they run their own bots their own keepers their own liquidation systems. It's it's not about trying to do it onchain anymore. Um it shifted more and more offchain.

[00:14:04]>> And then lastly, you know, immutability used to be a feature. Nowadays, I think it's a bug um in that it doesn't because cuz now if if you combine all of this, so you combine a new base that are looking at these things as an investment vehicle and no longer as a utility based thing.

[00:14:22]you are looking at teams that have to run more and more of their own infra offchain which means you know there's there's there's AWS or Google or Herzner or wherever you're hosting but but there's bills to pay right I mean apps again back in the original DeFi you know you'd throw up a UI on IPFS people would access it remotely you you wouldn't be hosting servers and infront stuff I mean the original year ran entirely on chain there. There was no off-chain server there. There was nothing. The only thing I paid for was the domain name. So that won't work nowadays because people need that extra usability. They need that access. They need that support channel. And and the second you need to provide all of these secondord functions that's not just smart contracts, you the the team needs to be appropriately incentivized. you know, you need to be able to pay salaries.

[00:15:22]People need to be able to stay on board, understand the vision, and want to see the vision beyond, you know, a two-year vesting term.

[00:15:32]>> Um, so I think it's drastically changed.

[00:15:34]I I I I don't think it's comparable at all. And that's why technically I'm actually a little bit against even still calling it DeFi because I don't think we're anywhere close to DeFi. I I generally when I talk to people I call it onchain finance if I do call it anything >> which is closer you know cuz because it's it's normal finance it's just we we we do things onchain that's the big difference.

[00:15:58]>> Yeah we call we call it neo finance which is effectively the same thing because it's not decentralized. You're absolutely right, but it takes there are certain functions that take place onchain and then there's a whole amalgamation of functions that take place offchain and users are just okay with it now because a of a combination of things. People have gotten burned, right, as a result of just code is law and not knowing any better. And also the cohort of people coming in, you know, they rely on some of the functionality that they've become accustomed to and the consumer behavior calls for, you know, customer support channels and forgot password buttons and and and these sorts of things. And so these are the design principles that they've come to expect. And I think you know maybe there has been this you know kind of this acceptance rather than you know trying to beat the drum and explain decentralization to people and expecting them to come to us and learn and adopt our our principles. We've decided no we're going to meet people where they are for the benefit of the adoption of this technology. We're going to build with their consumer behavior in mind and we're going to, you know, make the design choices that, you know, maybe sacrifice a little bit our on our own principles, but um, you know, as a whole, everyone is somewhat better off because something is better than nothing.

[00:17:20]>> In principle, agree. Yeah. I I I think the the gap comes in that we still have a lot of actors trying to sell >> it as if it's decentralized finance, you know, as if your counterparty risk is only the smart contract. And I think that disconnect is where the risk comes in.

[00:17:44]It's because I I I mean a perfect example is is is this new generation of curated vaults, right? Like like they would have you believe that the trust assumptions is the same as the old 2020 year vaults as an example >> where where there your counterparty was only the smart contract and the downstream you know whether it was a compound etc. Um, nowadays it's still happening onchain, but your counterparty is a curator. It's offchain. It's potentially a credit facility. It's potentially some non-liquidatable onchain RWA or some IOU. But if if you go and deposit in that vault, it's still kind of being presented as your counterpart is just a smart contract. Mhm.

[00:18:39]>> Um, >> yeah.

[00:18:40]>> So, and and and and the same is true for for tokens, right? So, I mean, if if you were to have a publicly listed token somewhere, there are very specific requirements in terms of disclosures, compliance, public audits, these kinds of things that are required. And we're trying to play in that space, but at the same time, we're still kind of avoiding that level of reporting and transparency etc. So now and and don't get me wrong I I don't think generally speaking it comes from a place of malice. I think it's a transitionary period that we're currently in where it is the old DeFi engineers that are now building fullyfledged businesses and it's just going to take time.

[00:19:33]>> Yeah. But I I do think there is a little bit of a disconnect in terms of what is being presented versus what it actually is.

[00:19:41]>> Yeah, I I I think that's right. And so I I mean you mentioned, you know, some of these curated vaults as an example, and you know, I'm sure we can dig into more of those, but I I'd love to dig into Flying Tulip as well. in terms of, you know, how you've been able to, you know, build Flying Tulip in such a way that you've attempted to, you know, make the most of these design choices because, you know, you're right, we're not in this DeFi purist, you know, uh, era anymore and we're also in this transitionary period where we don't have full, you know, clarity, you know, regulation in the US or just full market structure, regulatory clarity. And so there's sort of this transitionary phase and it's still trust me it's not it's not as bad as you know the Gary Gendler era in the US but you know we we've still got a long way to go as well. And so, you know, when it when it comes to some of these design choices, um, and let's, you know, just, uh, for the sake of honing in and and and, you know, understanding stick to the lending and the vault space, maybe you could just first give us a general understanding of Flying Tulip and then we'll get into the design choices that you've made and maybe how you see the the vault space and how you've been able to, you know, observe and implement some of these learnings in into the product.

[00:20:58]the the product itself, the the the suite of protocols that Flying Tulip really is is most simplistically it is the it's it's the composability of a lending market that can directly feed into an order book so that it offers you know trading and leverage that can directly feed into derivatives insurance.

[00:21:32]Um, and then we have our settlement layer, which is our native stable coin, the flying tulip USD. Um, it it it almost makes more sense why it exists and why it is the way it is. If you if you go back to again the premise of original DeFi, the original contracts one immutable two because there was no direct teams involved there weren't any fee extractions you know again using the original urine vaults it took zero fees so it's it was easy to build on those protocols because you knew the code was going to stay the way it was and you knew that there weren't fees. Nowadays, you know, you you've got unis swap v2, v3, v4, you've got a v3, v4. There's there's almost always and and they have to because, you know, they've become proper companies.

[00:22:37]So they need to a be able to continuously upgrade and deploy new systems and b be able to charge fees because they have to pay the people who are supporting these things. But that essentially killed composability cuz that meant cuz I could build a margin trading system on top of a cuz all you're actually doing is le let's say I want to I want to 10x leverage long BTC.

[00:23:08]What I'm what what what you would do is you would take a flash loan out of USDC.

[00:23:14]You would then on whatever trading venue you could, let's say unis swap, you would trade the USDC to wrapped BTC. I would deposit that wrapped BTC into a I would then borrow that amount of USDC and repay the flash loan. Now I have a 10x margin position in a but there's no incentive for anyone else to build that because the the fees are all going to a and if a then later on upgrades their system then same problem right I have to now redeploy and rebuild the whole system. So a lot of these features that we've combined in terms of composability is really just because the base layer which is in in the permission world the base layer is the is the lending market. In the in the in the non-permission world it's actually the AMM which is an interesting distinction. Um it needs to be aware of all of the things that's going to be built on top of it. not not necessarily in a it is already built methodology but at least in a terms of I am building in such a open way that these things could in future integrate into me um and and and that's that's also a design element that's you really don't see nowadays in onchain finance anymore because everyone is kind of building to protect their moat which also makes sense because it's businesses now right So you're not really building with other people are going to build on top of me in mind anymore. I I I think those those those days are largely done.

[00:25:01]>> But that left a massive gap, right? Cuz cuz one thing and and this is actually something I wrote about many many half a decade ago at this point which was back then called dairy swap, derivative swap.

[00:25:11]It was just one medium article and it was just a concept but then Gendler came in and kind [snorts] of ruined all chances of launching new stuff. So as you say the space is definitely better off. Um but the concept was simple. The concept was why can't idle liquidity sitting inside of a lending market at at the user's choice be deployed into a into a um single-sided liquidity position or if it is already in a AMM why can't the liquidity not being used in the AMM be deployed to a lending market you know and and we've sort of seen people try to do that right cuz you'll have Again using AVA and unis swap as the example nothing stops someone from going to unis swap and launching a a USDT a wtc which is ava USDT and ava wrapped bc pair and depositing that into the pool.

[00:26:09]The general problem there is that if I as a user rock land on unis swap's front end and I want to trade BTC to USDT, the engine doesn't know that a WB2C and a USDT actually represent the underlying assets because it only sees the tokenized version. Now aggregators have done a lot of work to be able to integrate those custom roots. But now I'm already talking about so many hops, right? like I first have to go to AA deposit stuff then to unis swap then get this stuff then an aggregator then set up this wrap and and it it if we were on that original sort of immutable and fearless and composable world you wouldn't need to do it that way right it it should be as easy as a tickbox that says yeah cool I want to opt in my liquidity for this and and that's the opportunity that's always existed that I've always been like yearning to exploit Um, and and that's where we're at now, right? Is is we've we've we've started with the margin accounts. That's that's and and now I'm probably getting a little bit too much into the weeds of the >> No, no, this is this is great. This is great.

[00:27:23]>> But but another thing, for example, is is LTV does its job well in old decentralized markets. So, our our lending market, and I'm even hesitant to call it a lending market. I I generally refer to it as margin accounts. Our margin accounts use an equitybased system similar to a margin account on an exchange would. So, so it compares the equity of your account. you know, that takes into account your P&L, that takes into account if you have netted assets like staked E versus E, >> um, staked Salana versus Salana, that that takes all of these extra things into account, which gives you a much more robust position than just LTV cuz just off of LTV and and and this originally actually started because USDE, Athena, um, which is the first time I even started discovering about, you know, Delta neutral yield farming essentially. But but I I was always frustrated that you can't do what Afina does offchain onchain.

[00:28:34]And the more I tried to kind of discover why, it's because you can't due to the nature of loan to value LTV, you can't hedge your delta away. Because if if my staked E is 90% LTV, there's always that 10% margin I can't get rid of. So it has to be 100 to 100. And you can't do that with LTV, but you can do that with equity based accounts. So you know, there's there are also a plethora of these more granular innovations that I think as a whole are also a step forward in the industry. Um, another small thing, you know, is when we do a trade on our own system, it doesn't just hit our order book. It automatically creates a RFQ order, and that RFQ order goes out and hunts the entire blockchain for liquidity. So, if it can give that user that wants to facilitate that trade a better quote somewhere else, we want to give them that quote.

[00:29:38]>> Mhm. because it's too often that you know if you go onto the specific AMM's website and you do a trade you're just going to be trading against their pools and and that that's that's not necessarily what the user wants. Now that does it gets fixed again because the responsibility is kind of shift onto the aggregator but but at the same time it shouldn't need to. Um there are lots of these little micro innovations we can go through. Another one I'm really proud of is actually that our liquidation system itself is RFQ based. So, it's not just, you know, you repay the debt, you get to get a chunk of the of the collateral. It's, hey, here's the collateral. You're an RFQ goes out. Give me a bid. And whatever the best bid is is what ends up getting sold. So, I mean, we've had it's unfortunately bad market. So there's been a lot of liquidations on the system and almost always our bad debt are well the liquidatable debt not bad debt the liquidatable debt is repaid dollar for dollar without a haircut to the user.

[00:30:39]You know it it's small but these little things add up. Mhm.

[00:30:42]>> Umh but look, all Flying Chulip mission statement really is is that there is no reason that every blockchain, whether it's a layer 1 or a layer 2, shouldn't be a fully featured centralized exchange. Right? If you go log into Binance, Coinbase, Kraken, whatever else right now, you can do lending, you can do yield vaults, you can do spot trading, you can do margin trading, you can do hedging. And and and that's the mission statement, right? It's having the central entry point where you can deposit funds and then offer you all of these features while giving you the best yield, capital efficiency, and composability. Because again, something simple, right? If you can place a limit order on our spot system, that limit order of yours is earning yield because it's in the margin account being lent out on the other side. So there there's never a point where your capital isn't put better to work. But anyway, I'm feeling a little bit too pitchy now and >> this was not the goal.

[00:31:48]>> I I understand. And I think well I think you're you're also talking about design choices in general that these chains are making and I think you know you've gone and decided and build in in this direction with Flying Tulip but you know the most the most recent example of this is Mega ETH right they they sunset their mega mafia cohort right of third party applications their incubator that they were doing and now you know they're they're going and building firstparty applications which is you know uh what what you say every chain is going to end up doing whether it's an L1 and L2 uh with some of these features in mind because ultimately we've learned you know through so many of these chains these are the primitives that people want and it just makes sense that we have a more composable world of each of these functions on each of these chains I and we uh you know I I want to get into the stable coin element of Flying Tulip. It's something that you didn't talk very much about but it's a core component and this is our stable coin segment on the show. So I want to talk about that. Uh be before we do I just want to close the loop on one thing that you mentioned earlier which was an interesting point. You mentioned how the core base of these composable systems in the traditional world was typically the lending market and I think that was mostly because of the margining profile and you were able to run and now you've actually implemented this equity margin like account on flying tulip. Um, and that's sort of, you know, the base account that gives way to most of these features. But on chain in the past, it had been these AMMs, right? And so I'm just curious about the distinction. Why was it the case that it was AMMs? And why did you choose to include this equity provisioning account on Flying Tulip more akin to the traditional world than the onchain native world?

[00:33:30]>> So well, the answer there is that it's still both. So, so within flying tulip itself, we have two distinct verticals. We have our permissioned set, which is the one we launched first because it's easier to control. So, the benefit of the permission set is that it's cross collateral. So, I can onboard BTC, E, USDC all in the same margin account. The problem with that is every time you add an asset, you are implicitly exposing the entire existing book to that same asset. Um, and that's true for, you know, any crossc collateral system. Every time a adds a new collateral asset, it's automatically exposed to the entire book. and and we've unfortunately seen the downsides of that um in the in the but but that means it can't work in the permissionless market, right? Cuz I want to >> sorry >> you're referencing like Kelp Dow and like how you know just the RS state ETH expose like potentially the entire ETH book on a and these sorts of things. Oh, unfortunately. Um, >> now now don't get me wrong, I I I don't think that was a a bad onboarding choice or curation choice or anything of that.

[00:34:52]I'm I'm just saying that the second you onboard these assets, you are implicitly exposing your entire credit facility to them and and that hurts your depositors.

[00:35:00]>> Yeah. Um but now let's say someone wants to you know come launch a new token and they of course can't automatically be part of that cross collateral margin. So our permission so so the permission set the first entry point is the lending market on the permissionless set the first entry point is actually still the AMM >> because >> Mhm.

[00:35:25]>> if it's a new token you have a few problems. Problem number one it doesn't have an oracle. So I can't just pull, you know, something from chain link or or redstone or or even just coin gecko, right? And sign it and post it on chain.

[00:35:38]So So I don't have that to work with.

[00:35:41]Thing number two is I don't know things like liquidatable depth or what kind of trade facility or I should offer or or how I should quote this asset even. Um so for all of the but you can solve each one of those pain points individually through the AMM itself.

[00:36:02]>> Yeah.

[00:36:03]>> So our our permissionless markets start with you launch your normal AMM which is two pairs right that's just your asset and ETH or your asset and USDC or your FSET and whatever else.

[00:36:17]this as that starts trading it starts building up your oracle data which is your TWWAP your time weighted average price >> and then it also starts building up something that I always found lacking in this space and it's something I actually included in the original um V33 designs all of those all of those V33 protocols actually all have this but I haven't seen any of them use them yet um and it's something I call Tiwir, which is time weighted average reserves. Because if if something is priced at $3, but I can only sell >> $1 of it because there's only that much liquidity, then its price being $3 isn't actually its price. Its actual price is how much can I sell this for.

[00:37:09]>> So more important to me in the DeFi space was time, weighted, average reserve. Now all AMMs have a have a A and a B reserve. So that's the number I actually want to want because that number means that at any snapshot in time I can ask the AMM, hey, how much can I liquidate this position for? Now normally you'd assume, hey, I can just, you know, ask it in real time. Why do I need to keep this historic metric? But that's unfortunately the world we live in with flash loans and LP manipulation means that you can never ask a question in real time. You always have to have historical snapshot of a question. And even if you do ask in real time, I need to be able to compare to some historical point that I can see is this a good reading, is this a bad reading, is someone manipulating. But the question you need to answer is how much can I sell? And the reason if you work back from that question, the the implicit question to that one is how much can I liquidate? So we actually end up using the the time weighted average reserve output to see what kind of LTV I can offer the user cuz now the same AMM pair let's say it's it's X and USDT.

[00:38:23]Now that same pair facilitates lending.

[00:38:25]So the user can provide their X as collateral and borrow their USDT and their LTV is based on their exposition relative to the pool. So it's not a fixed value. It's not you have a 50%.

[00:38:37]It's if you're if you're a very small bit of the pool, you get a very high LTV because your impact is [clears throat] low. If you're a big part, you get a low one. And inversely, >> um the next thing you need is interest rate models because you don't have that.

[00:38:50]So the other thing that these AM pools track is the the the moving averages and the volatility. Okay, >> the moving averages and the volatility also influence the the curve at which we quote the price. So assets that are are completely flat in terms of volatility are priced on the X3Y curve which means it's basically a stable coin.

[00:39:16]um the more higher that volatility is, the more it veers towards the traditional unis swap um XY and then anything in between it allows for and and this curve in itself I also need it because it facilitates forex trades because stable swaps or XY don't really work for forex trades because it's low volatility but it's not no volatility.

[00:39:38]So that also sits on this curve. So this curve the the pricing between the reserves is also influenced by the volatility and the moving average. The interest rate is also because in in the lending world you really have two interest rate models. You have volatile, you have three. You have volatile, you have major and then you have stable. Um and that's again answered by volatility and your moving day average. Um so now I can offer lending and I can already offer trading. I have all the data points I need to be able to set up these positions. And then the last thing you can do from that is you use the work from from GMX and their GLP to to allow that same LP to be the counterparty for derivatives as well. So that you can take a long or a short against that position. And because you already have the the lending market built in, now you can do leveraged positions as well. Um so so when when it comes to pure to pure >> permissionless >> single asset pair permissionless so anyone can launch this on their own you know they don't need to interact with us at any given point in time then your your starting base is still the AM and I don't think that's ever going to change because that's that's all of your data.

[00:40:51]Yeah, I mean this is fascinating stuff like h how you essentially you're given you know a single you know primitive and and you know single is maybe a little bit too too far you know you've got a couple of these different interest rate models you've got a couple of different AMM models right but from those base equations you're able to build this entire suite of functionality from the AMM you're able to build the lending market you're able to build the LP market and the per market all from this single AMM and you're able to do it in such a permissionless way.

[00:41:27]>> Only qualification I I'll make there, you know, is is it it's not a pure per.

[00:41:31]It's closer to a pre-market per cuz the pure per funding is obviously influenced between your external oracle price and your mark price. Here you don't have it.

[00:41:40]So your funding is actually closer to open interest. You know how much of how much are long versus short and then long space short. So it functions like these pre-market IPO per that's that's one small qualification I'll need to make but but other than that it it functions you know >> and I mean that's also been the benchmark for you know these low liquidity early stage perpetualist markets like that is the same way that all these other markets tend to work it's only just after you know they do go IPO they have some sort of liquid market price that's external to the to the book you know then you bring in that oracle price as a means of sort of smoothing out what would otherwise be a rather, you know, uh, uh, funky, you know, Oracle price because it's internal. Um, yeah, Andre, that I mean, that's fascinating stuff. I I I really enjoyed just kind of walking through these things, uh, with you and how you were able to illustrate the difference in building the per the their the permissioned stack versus the permissionless stack. So, makes a ton of sense. Um, I want to be conscious of your time, but I also want to get into the stable coin side of this. Um, you know, we talked a little bit around stable swaps and these sorts of things for low volatility pairs. Um, but maybe you could just talk about the flying tulip USD, you know, why it's such an important part of the overall system.

[00:42:59]Um, and and you know, generally, you know, how you foresee this growing as a core component to the book.

[00:43:07]Um I I I I think everyone will agree with the statement that we have seen stable coins are the primary settlement layer and liquidity of almost every single big system we have. Um the the original premise as I said came from Athenos USD. It was looking for a how can I >> do this leveraged delta neutral play onchain with LSTs because I also wanted to broaden it from just if you know I want to I I we obviously do SD E on Ethereum and SDSS on Sonic but I want to do Lista BNB BNB on BSC you know I want to do um Hydrax AX AX on Avalanche.

[00:43:57]Each one obviously has its own and and very often even even even if you do these things in a centralized fashion very often you can't because that staking asset doesn't really count as as collateral on most of these exchanges margin accounts. So there there isn't really a way to do that proper hedge. So that was from the outset kind of where we wanted to start with the stable coin is how can we do this leveraged deltaneutral tokenization of the staking yield. Um and and and we couldn't that's ultimately what it came down to we we needed something that you could properly hedge the position. So LTV was out right from the bat and that started influencing you know sort of we first looked at LTV but with standard netting so that within the config you could say this asset is netted against that asset. So if they are which which it kind of looks similar to to to aves emode um but you're still not going to hit 100%.

[00:45:06]You're going to hit 99 or something similar and you need to be able to hit 100% to properly hedge your position. Um so we built the equity market >> which is one of the reasons why we use the equity margin account. And so our margin accounts always allow that that play.

[00:45:21]>> Then the the other rationale of the stable coin was as a mechanism for scaling. So it's normally pretty easy to get any given blockchain's native asset in a large amount because there's a lot of liquidity for the native asset. So it's not hard to get a lot of people wanting to deposit because there's also not a lot of yield opportunities for native assets. So it's generally fairly easy to attract especially when you start talking to some of these digital asset treasuries and foundations and things. You know they have a lot to deploy.

[00:45:52]>> So we can get that easily enough. Now the other thing you need is in to your lending market is stable coin liquidity.

[00:46:01]>> Oh so maybe one thing one one this will make a lot more sense. I should have actually started with that. So with something like Athena's USDE, you know, they're they're what they are tokenizing is staked ETH as margin collateral and that's earning you 2.4%.

[00:46:18]>> And then they are taking out the the perpetual so the derivative short and then they collect that funding rate which is for argument sake four to 5%.

[00:46:28]It it's currently I don't think it's currently >> oh it is probably still positive. um and and then minus fees and settlement and all of those things. So you end up with a delta around like four to six%.

[00:46:41]>> Um our lending market is not a per market. It the pers are getting activated later on and then we will still do that standard trade but for now how we do it is we we use the USDC and the USDT that is deposited into flying tulip USD and that becomes collateral.

[00:46:59]So that becomes what we borrow the ETH against. And then we swap the ETH to staked ETH. And then the staked ETH becomes collateral as well. So the staked E and the E are still hedged, but it's grossly overcolateralized due to the USDT and the USDC. And that allows us to to leverage that loop as much as you know there is available deposits.

[00:47:20]But that means we're not we're earning the 2.4% 24% from the the the state E and that can safely be looped up to about eight times with this design. So we don't we're currently at about 1.5 because we don't want to push the system too hard too quick and it also depends on available liquidity but you can um so that already will give you so 2.4 but you're still minus about 2.1ish for every E that you borrow. So you net about.3 which that then at a eight times loop you know you end up it it's not 20 30% but it's still decent numbers and then you know the the 3.2 to 4% that you're earning on the stable coins and that gives you of your net of around like 11 to 12%. Um so that's how that delta neutral loop looks and that's actually different than you know the standard derivativebased one. Um, so the the flying tulip USD is almost a little bit of a Trojan horse in terms of liquidity as well because people deposit because they want to get the yield. By depositing, they're offering more stable coin liquidity. Now, our margin account has the two most important things it needs. It has stable coin liquidity and it has the native asset. Almost all trades on any blockchain >> happen in either stablecoin or its native asset. So by having a lot of those two, we can start offering very competitive trade rates in the spot or the margin or the RFQ. Um so we start on boarding those assets. More people do the margin trades which means they borrow out more of the money. Interest rates go up that influences the interest rate on flying tulip USD. So its interest rate goes up. It attracts more deposits and then the kind of flywheel starts kicking in together with that. Um and that's even before PERS are activated. Once per are activated, then you know we'll do the standard carry as well as a secondary trade while it's positive.

[00:49:18]Um but further you know something like our total return swaps or any of our derivatives is also settled in flying tulip USD and that's because we can do instant settlement with it if we can mint and redeem at any given point to close the trade and then once the underlying is settled you know we can burn again or mint again depending on which way we have to go. So having that sort of mint and redeemability also gives that instantaneous settlement on the derivatives and things like the total return swaps which is important.

[00:49:51]But then generally speaking, you know, we also just want to offer the flying tulip USD as a as a ARB layer for once for once it is the base pair of more of our assets cuz if we can if we can increase supply or contract supply based on mint and burn at any given point in time, we can always take that ARB as well and provide that profit to the user since we won't have fees on it. And then lastly, your standard CDP model, you know, allowing people to mint against it. But that's just another interest rate source of people paying so that we can just diversify all of the yield opportunities on the flying jud.

[00:50:30]Um, but as you can see, you know, it's a touch point everywhere. It's it's a touch point for lending. It's a touch point for trading. It's a touch point for derivative. It's a touch point for all of your trade settlements. So building building that as an attractive offering is is your liquidity layer for your entire system, you know, and and and that's something that's something you see with with [snorts] um all all of these per markets, right? Like like they always have these big whether it's HLP or LLP or or or or any of these other derivatives. Um they always have this one big stable coin vault and >> you can do so much more with that. you know, there's so many more opportunities that you can place against that and that also hedges you against, you know, potentially when you've taken too many bad trades or there's a bad liquidation or a bad oracle feed or something like that.

[00:51:21]>> So, so it's it's I mean I'm I'm I'm hesitant to use statements like this, but you know, it ends up being sort of the lifeblood of the entire system because you need it running through all of these parts.

[00:51:35]>> Yeah. And it it's also the, you know, the jet fuel, the initial gasoline to get the thing off the ground initially because you walked through mathematically how it it is possible to develop these components and stack them again permission stack, permissionless stack. It is possible to create the the functionality from these core components. But of course, if there's no liquidity running through these systems, then what's the good of the system? And and so you walk through how important the liquidity in the system is to the system as a whole and how the stable coin helps to stimulate not only the initial liquidity but it helps to perpetuate the liquidity as well as the system continues to grow. So it it is you know obviously crucial to the system as a whole. Um and it's also I correct me if I'm wrong here but it also seems to be the the value acrruel the destination for a lot of the things a lot of the yield sources these things tend to go towards so because it's a touch point for all these things it also tends to be like sort of this rewards or yield accumulation layer where a lot of that value tends to occur >> 100%. I mean, if if there's if there's settlement fees in the derivatives, it goes there. If there's liquidation fees, it goes there. If there's trade fees, it goes there. If there's lending interest, it goes there. Um, so it and and again, it's kind of like I called it at the start, it's kind of the liquidity Trojan horse for the entire system because it ends up having all of these different touch points. Uh, and and I think we've seen that everywhere, right? like like the lending markets that survives are the ones that attract a lot of stable coin liquidity. The trade venues that survives are that have a healthy stable coin order book. The biggest derivative platforms have the biggest um stable coin LPS. So it's it's directly correlated to how much depth can I offer, how much volume can I offer, how much how many options can I offer to my user. Um, and at the same time, you know, I think also having it have all of these different touch point also massively diversifies its risk because now you don't just have the single, you know, my my only counterparty isn't just per trades, right? Because I know even if something goes wrong there, there there's a certain amount of just USDC USDT sitting in lending that's, you know, has a much safer risk profile. Um, stable coins are so important, man. I mean, I' I'd argue that's probably one of the best innovations of crypto as a whole.

[00:54:17]>> Yeah. Um, I mean, there's this is the last question I got for you and it's a wild card. The We've got the chat here because we are live and so people are they're they're loving your appearance and so we've got this this uh this question here in the chat that I'm going to ask you. Um, and you're right. I mean, stable coins are extremely important. We've seen that, you know, as a result of their growth uh now with, you know, Robin Hood chain and these sorts of things. People are saying, "Oh, you know what? Tkenized equities are an even bigger opportunity than stable coins." Um, so we'll, you know, we'll we'll see how that plays out. Similar question here. And I, this is a wild card question. I just I but I want to I want to, you know, give you a give you a chance to sort of address this. Travel Wings in the chat here says, "Will institution institutional capital institutions seriously look at flying tulip? I think retail is done with Andre." Look, I I just I want to give you a chance to just address, you know, the the perception that people may have of sort of like your journey because look, I understand the the depth and breadth that you've had in this space and and obviously, you know, this is an a feat of um incredible development work that you've done building this system and walking walking through it. I really really do appreciate you walking through it here today. And just to address you know the what some people might be thinking out there. So given this given this sentiment like what how how do you respond to you know this kind of this question right around retail sentiment and institutional sentiment when you know the they're looking at something like flying tulip.

[00:55:54]>> The only thing that matters is does it offer better riskadjusted yield? Does it offer more features than I get somewhere else or is it cheaper? Capital doesn't care.

[00:56:06]>> So, I don't think there's much to address there. I'd say the only thing I would address there is that the the the little circle jerk of CT narrative is very very different than what's actually happening out there in the real world.

[00:56:18]Um, interest is big. There's lots of conversations, but what it comes down to is is it cheaper? Is it safer? Does it reward better? And that's all that matters.

[00:56:31]>> Travel Wings, I hope you got the answer to your question there. Capital doesn't care. Uh Andre, man, really really appreciate you taking the time today. I know we went a little bit over time. I really do appreciate um you know, just kind of very methodically walking through some of the questions here and explaining the way that you've you've built this system, you know, the lessons that you've learned from previous projects and how you've implemented this. And uh very excited for the growth story of Flying Tulip um to see how this plays out over time. And uh man, I'd love to have you back on in a in a couple of months again and uh get a get another, you know, temperature check on how things are going.

[00:57:04]>> Yeah, always happy to and thanks for having me. Always like talking to me about this stuff.

[00:57:09]>> Absolutely. Appreciate learning about it. Uh we'll see you again soon. Thanks, Andre.

[00:57:13]>> Thanks, boss.