IPOR: Interest Rate Swaps and Automated DeFi Vaults

0:00 Ladies and gentlemen, welcome back to the permissionless podcast. Of course, brought to you still as always by the team at Guardian. And of course, we're going deep with DeFi teams, building on the edge of what's possible with DeFi tech. And today we've got really a wonderful conversation with Lukasz from IPOR, which is a protocol that has has been around for some time at this point, offering fixed rate lending on top of protocols like A.

0:28 And in this discussion, we get to go in depth on the kind of underlying technicalities and complexities of the IPOR protocol, but also discuss a little bit about how did Lukasz even find his way in so early to a D5 protocol that was really ahead of its time in terms of complexity and imagination. And then finally, we discuss a bit about what's in the near future for IPOR and and what we think DeFi is headed towards in general.

0:57 But of course, we will go deeper on that in just a second. My name is Owen, and over three and a half years ago, I founded Guardian based on the premise initially that DeFi was going to house trillions of dollars of assets. And really, we needed a step-wise higher level of security to be able to ship new code and complex protocols in that environment.

1:19 And I'm happy to say that today we have achieved that standard with Guardian using not one but two distinct auditing teams on the same scope, worldclass security researchers and a dedicated fuzzing engineer to fuzz mindn numbing edge cases into oblivion. In fact, we've been able to prove out this model to the point where we're confident in putting our own funds up as a bounty on the contracts that we review.

1:45 And so at this point we have achieved the standard necessary and now it's just up for DeFi to catch up. And of course without further ado let's dive in to this interview with Lukasz from IPOR. Well, yeah, dude. I mean, first of all, appreciate you joining us and making the time to to hop on and and discuss with us a little bit about, you know, pour your journey into DeFi, what your experience is and and anything else in the web 3 space.

2:16 But yeah, I mean, really, I think it'd be great to just start and get a sense of how did you first find your way into DeFi, and then we can discuss what got you into from there. Well, thank you for having me. my name is Lukasz, by the way, for those who don't know me, which is probably like 100% of people watching this.

2:31 And well the story goes back you know several years. I was actually in in Shanghai. I was living there at the time and I met this guy who who I now work with and you know we had very long long discussions about you know blockchains and this is back in 2015 I guess and at the time I was you know on a skeptical side like a old school gold bug if you ask me and but you know the the idea of the blockchain was you know speaking to my heart especially you know the self-custody and you know a resilience of the blockchain is obviously you know it's not the same as gold but it has its own unique properties and you know he you know he talked me into buying some Bitcoin at the time

3:17 And you know I bought some that some some Bitcoin I was even day trading quite successfully at the time but you know I have no nerves you know it's it's just too high stress for me you know I'd rather build you know that's what I was doing at the time and so I'm on the engineering side of things and but you know I got some Bitcoin then when I was moving out of China about the same time, just maybe a couple months earlier.

3:40 OKX or OKCoin I think it was called this it was called like that back then they locked me out of my account with my bitcoins because money laundering rules in China and all the foreigners and even though I it was fully legit account and I have to prove the taxes that I've paid on on the money that I p put in my account they said you know we can't use it and two two years later I almost lost hope that I will ever get it I wrote it off to be honest But you know couples but anyway I finally miraculously got it back.

4:15 But the guy that I used to talk about you know crypto with he called me out of the blue one day and he says listen you know I have this idea and we're going to build derivatives on blockchain. It's like, "Oh, tell me more about it." and and he basically went into what DeFi is how we know it like you know how do you interact with Aave you know how the flash loans work and you know how everything is variable rate and you know and he has this idea how to make variable rate fixed rate you know through interest rate swaps and you know I'm kind of you know I have a background also in economics so you know say oh interest rate swaps you're saying so and we went into interest rate swaps and

5:00 How they can work on on blockchain. And he he basically said, "Listen, I'm looking for our team." So I told him like, "Listen, I I know some people, they're excellent, you know, they're absolutely amazing engineers, and you'll do fine with them." So I introduced them to him. but then they refused to go without me. and and my buddy refused to take them on unless I join.

5:21 So they bullied me in joining. even though I was quite happy where I was at the time, it was, you know, like a little corporate job. building apps for Fortune 100 and so I joined I was like I mean I've been to a couple startups before I mean it's kind it's kind of fun and so that actually at that time I started getting myself into blockchain like very very deep into technicalities of it but my journey is is maybe different than most people in the blockchain cuz I came from a you know as an engineer that was brought in to solve very specific problem.

5:59 Okay. So I was given a problem of building an interest rate derivatives and a constraints of the blockchain of you know Ethereum at the time say how can can we replicate this model on Ethereum you know given those constraints you know like you cannot iterate over lists and you know you need to keep portfolio of of assets managed and it was it was amazing amazing challenge and you know we we did it and we actually built a very sophisticated algorithm make derivative on Ethereum.

6:32 It's still running. It's still being used. Y so can allow leverage to I think theoretically to about a thousand. So it's pretty pretty cool. Yeah. Yeah. The we we run into some issues of like manipulation of the market but I think we we we we come around to find a solution to it. So so it's this this particular battle ain't over.

6:56 Very cool. So what what year is this around where you're coming in and you're getting dragged into this very specific problem? That would be four years ago. So this is 2001. Yeah. 2001. Yeah. 2001. Yeah. So but I've been working in tech for you know 10 years before that you know various different tech projects. Yeah. So coming in this is right around the Bitcoin crash from like 60 to 30 or whatever.

7:23 Oh yeah. Yeah. Yeah. Exactly. I've seen I've seen a couple of them already. So yeah, I I lost I really don't if I'm honest I I don't really pay attention to that that much. You know, I own crypto, but you have to ignore it or you're just too busy in this space. There's just so much going on 247.

7:43 Exactly. And you know, you want to you want to keep your sanity, you know. Yeah, of course. Yeah. I mean at a certain point you just you work with the tech so closely and you see like just how powerful DeFi is with you know infinite composability and anybody can interact with it that it's like you know the price really doesn't even matter.

8:02 It's like this is nothing else in the world has this like ability for a an open financial system and the compounding nature is only compounding. It's it's it's doing exactly what it needs to do with composability. But yes, I mean that's super interesting that you specifically were dragged in for this very particular problem and it actually ended up being like I think a cornerstone of of a poor and and what you guys are offering and I think it's the same contracts are used to to this date or you know probably some changes had to be made or some configurations or something like this but so that's actually pretty unique.

8:38 I actually came in around the same time, but obviously, you know, did the whole like losing a ton of money on DeFi projects and then learning about DeFi and all of this stuff before ultimately we we started doing security work. How did that individual know to like to work on interest swaps? Like what what was their background?

8:57 How did they see this this opportunity in the marketplace? Well, I mean, there is, we we look at Aave actually, you know, when we were talking about it, we look at Aave and Aave at the time had the fixed rate loans as a product. I think they discontinue it afterwards. Yeah. But it had a pretty interest like you know quite a few people actually taking those loans at the fixed rate and the rate was crazy high you know and it it just you know got us thinking that you know this actually there is a demand for fixed rate and everybody was trying to set solve the fixed rate and then Pendle at the time was you know addressing the fixed rate with with their own unique model and we thought like you know can we

9:38 Can we make you know this more efficient because you know the problem with Pendle was and this is before they rebranded and that rebranding was actually a massive success. Yeah. and they improved the model also. You know, you essentially were always limited by the amount of of liquidity that you know users would bring. And and when we look at the traditional markets, majority of interest rate swaps are actually synthetic.

10:08 So you don't really need you know underlying asset. And we were like, can can you actually trade the swap? Can you trade the yield without underlying asset? You know, like pretty pretty much you put some collateral and you trade interest rate of Ethereum. Mhm. And and we were looking at actually this is where the name comes from.

10:29 The IPOR stands for interprotocol offer rate or overblock rate. So you know in traditional markets you have this thing well you had it before it was killed LIBOR and or various other indexes like software or Poland there would be a V-bought or whatever and and when you build your financial instruments you normally do it of that rate.

10:53 So that rate becomes your benchmark and and then you you can trade your risk on that rate and get your fix or floating and switch between them using synthetics. And we thought well can we build something like this and you know we designed it say oh we can build it. and and we built a synthetic synthetic protocol.

11:15 It in the first couple of months we actually traded over $4.5 billion dollar worth of notional. So it's it was extremely scalable. but then we kind of run into some issues with like market manipulation. You know there people that started okay can we manipulate the market in order to take advantage of IPOR. So we had to it was like this u you know cat mouse game where we had to build automation to you know monitor the market and you know cut the essentially the depth of how many derivatives we could offer.

11:50 Yeah, depending on the market depth, you know, so it was susceptible to being manipulated because the liquidity was thin, you know, we had to lower the limits and it was very unfortunate, but actually I think we are now much closer to solving this problem than ever before with our new product working together with our old product. So that chapter you know we I'm I'm happy to speak about the new product as well but you know there was certain events that led us to working on a new one and I think they both you know going to work quite well together.

12:23 Well first of all there's I mean there's a lot that we could we could talk about from here. So I think maybe before we jump into each like individual product area and I definitely want to get more in depth on the the AMM that you guys are using underlying here and how you kind of Yeah. So the AMM the AMM that we built for the derivatives it's a bespoke it's a bespoke AMM that is for our derivatives specifically.

12:47 So it it essentially like it has algorithmic well it has algorithms to calculate the risk. and it's actually very unique solution because because of those gas problems we had to be very innovative. but then that led it to be a bit harder to use for traditional DeFi users because part of the risk parameters that we monitor and we we use in the in AMM they're very extremely hard computationally and we use quant models in the background and those cannot be run on Ethereum they're just way way too heavy and so what we end up doing in our AMM was to rely rely on off-chain calculations for the risk management to offer the the swaps.

13:39 So once the swap is open, it relies entirely on the onchain data. Okay. But before it's open, the offered price for the offered spread for that particular swap is it it takes advantage of the quant model that is running in the background. And then we we essentially compile the set of data like we call it risk parameters.

14:02 It's a it's a the whole array of of parameters and then we refresh it every couple seconds or I think every up to a minute I think and then we have algorithm running on chain to fill the gap for that minute. and and it's all signed by our oracle. So the oracle publishes the off-chain data with a signature and then the contract will accept accept those data with the signature will verify the signature and then it will offer a swap if it all matches.

14:34 So that was a very hacky way to solve the problem of of computation and it works quite well but the main problem is it's making it much harder for composability for the regular users. you know, so this is product this is definitely not a product that you can automate as a regular DeFi user. So this is you can you can totally automate it but you need quite a bit of skill and that kind of led us to to the second to diffusion.

15:01 This is the main current product that we have. But yeah, like the AMM was bespoke. We couldn't use we couldn't use any offshelf solution because the product was so much different. Yeah. And this was before even Uniswap v3. At the time it was just Uniswap v2 I believe. yeah, I think when we started there was Uniswap v2 and then in the meantime Uniswap v3 came out.

15:25 But you know we had we had attempts at solving the swaps with AMM from Dexus. I think there was a vaults. I don't know if you're familiar with vaults. they were using I forget now whether the the V2 or V3 type the uni V2 V3 type AMM. Yeah. but that was they they they got exploited at some point because it's just not a good model for this Pendle has its own bespoke you know AMM we using slightly different I mean very different model because we we have like we literally don't rely on asset we rely on the IPOR index that's compiled from the borrowing borrow rates on Ethereum so there is no IPOR index is a is a synthetic thing okay we grab all the index rate all the the rates from a and you

16:21 Know DSR and so on and compile that into an index and that becomes a base for on what which we trade there is no yeah so there is no there is no AMM that you can use unless you you have that u you know specific case yeah so the so the index is is an aggregate of all these different lending venues of what the floating interest rate is well it's not an exact average but you know Think about it like it's like a LIBOR.

16:48 Yeah. It's a library of of DeFi. Yeah. So, we actually use it until this day, you know, we we you know, we it's a synthetic, you know, or it's a it's a benchmark of like what the rates are in DeFi and we lose use blue chip markets and and then we see, you know, how's the markets doing and now they're doing crap, but it's a cyclical market.

17:09 They go up and down and we can see it and, you know, been collecting IPOR for like what four years now or more than that. Yeah. Well, I mean, yeah. I mean, talk about a product that has like lasting power. That that's pretty impressive, especially in DeFi to have something that's going on its its fourth, fifth year at this point.

17:24 So, yeah, and we can definitely get into more of what you guys are building on top, but I just want to for everybody listening, maybe we can walk through like what is the actual mechanics of an interest rate swap. So in my understanding it's like you have you you you go and borrow USDC on like a morpho or a and then you swap it for IPOR USDC which is this index and then so you've swapped at a certain tick or bucket to hear how you guys are delineating this and then you hold it and then that IPOR index trades wherever the interest rate goes.

17:58 Yeah. interested to hear like what what how does that come to life with if you if you don't mind I would like to kind of step up with the ladder of abstraction. Yeah. and and maybe just start with what the interest rate swaps are. Okay. So in in the traditional markets you know when you take have you ever taken a loan like maybe a mortgage mortgage is probably a good example.

18:19 So if you take a mortgage, you can do you can take either e either a variable or adjustable rate mortgage or a fixed rate mortgage right and you know if the interest rates are like normally the adjustable rate mortgage usually has a lower rate than the fixed rate okay because there is a risk the interest rate will go up and then you know you're on the hook on you know higher interest rate payments.

18:41 So, but let's say the interest rates were low and you know you were kind of reckless and you say, "Okay, I'll take a loan at you know 1%." adjustable and then the rates go up and then you want to you want to maybe switch to a fixed rate because I mean you expect the rates will keep on going up.

18:58 So what do you do? You go to the bank okay and you have to refinance meaning you need to get entirely new loan okay for another 20 years. But so this is what it takes to switch between adjustable and fixed. And normally you're going to get a rate height. Okay? So maybe you got 1% and now the situation change and you get maybe 6% fixed.

19:21 I'm I'm making those numbers up, but let's say 6%. Okay? Imagine at the same time I took a loan at the same, you know, same amount. but I took a fixed rate mortgage. Okay? I got it at 3%. And now I'm still at 3%. Okay? and you went from one to six theoretically, right? You went to maybe one to five but adjustable.

19:43 So I have three and you have five and you want to switch to fixed. So what you can do is you go to the bank or refinance or you can come to me and say, "Hey Lukasz, how about we do it this deal for one year?" Because I'm guessing one year is enough and everything is going to change for one year.

19:58 We signed a contract that you will pay my floating rate, okay? And I'll pay you 5%. Okay? So you pay me less than you would pay in the bank. So it's a win. I I pay you and I pay you 3% which is 2% you know delta. I'm taking a risk if the interest rate is going to go up.

20:17 Well I get Yeah. Yeah. I get Yeah. So I get sorry I get so yours is five maybe we we agree on five and a half. Okay. So I get five one and a half% delta. And my cost of capital is also lower because I'm, you know, I have my base at 3%. So you you win, I win.

20:37 Okay? I'm making money, you're making money. Bank doesn't make money. But for one year, I pay your interest, you pay my interest, you know, you save. I I I gets to earn some extra. And this is what actually banks do. If they want to switch or in big corporations, when they want to switch from fixed to floating or from floating to fix, they don't refinance.

20:58 Okay, they keep their own loan or they keep their own deposit. Okay, the the money doesn't move. All they do is they get a swap. And swap is essentially a contract for parties to pay each other's different rates. You know, I pay you adjustable, you pay me fixed or vice versa. So what we want to do with the the interest rate swap is essentially replicate that model on chain.

21:22 So we say the floating the rate that's going to change is the IPOR rate and the fixed rate is whatever you get on the contract. So you can come to the AMM on IPOR and you say I want to pay fixed okay of whatever the contract offers me and I'm going to receive whatever IPOR rate is and the AMM will calculate the cash flow difference between both rates okay over period of time.

21:46 So the you know if you keep it for 30 days, you keep it for 60 days, you keep it for 90 days, you know, you essentially get paid on the notional. So you say open a million dollar notional and you get paid interest on $1 million at the you know as a fixed rate and then you pay floating also on the million dollars.

22:07 So all you need to put in the contract as a collateral is you know part of the notional. So theoretically, the IPOR protocol allows for like a thousand times leverage. So if you want to trade a million dollars principle and then interest on that, you need like $1,000 collateral. but I think I I haven't checked it, you know, what the the machine is saying these days.

22:32 But it can lower that leverage depending on the market condition. So if you know the market is extremely volatile, you know, it will lower the available leverage. So, so interest rate swaps are extremely powerful. So, they're second biggest class of derivatives in the world. And I like to say this is this the biggest class of derivatives almost nobody understands or knows about cuz everybody knows forex, right?

22:57 Everybody knows options and you know futures and maybe even warrants you know CDOS's you know SDRs or whatever. But interest rate swaps are you know for the financial people pretty much only. Yeah. And so in and your approach is essentially kind of like a synthetic. Yeah. So so you can what you can use it for you can obviously directionally trade say oh I think the interest rates will go up.

23:25 So I I go long interest rates meaning I pay fixed and I receive floating and if I think interest rates will go up you know I I go short. So I want to you know receive fixed and pay floating because floating I I presume is going to go down. You can use it for that but you can also use it to hedge your loans or deposits.

23:47 So if you you know you borrow on Aave okay it you know right now the rates are like what 5%. you know maybe maybe less depending on the asset like USDC would have a slightly different profile than than Ethereum or Ether so say now the rates are 5% I think they're going to go up okay so my cost of borrow will go up with the floating so I want to hedge myself so I have a you know position on on IPOR where my my swap is you know longing interest rates so with the rates go up, I I'll win on the on the swap.

24:26 If the interest rates will go down, I lose on the swap, but I also win on the on my borrow position because I'm not paying less interest. And reverse would be true for the deposits. So, this is what normally is done on the in traditional markets. Yeah. Yeah. And and to do it on like a large deposit, you know, a large collateral, it would be insane.

24:47 So, that's why I mean it's it's a very efficient product. So you don't really need much collateral to secure that position. Yeah. And obviously this index that you're using and how you're calculating the fixed rate is based on onchain values is based on what the current rate is on all these venues. And so you talk about price manipulation.

25:07 What did those kind of vectors look like? Did they just go to like an a and like literally manipulate the rate and then make guaranteed profit? Yeah. So, so this is you know if you allow to take you know extremely high leverage and you don't restrict how much money is funneled through you know you know the derivatives you know you can actually allow for extremely high number like high notional you know even exceeding the amount that is in Aave or you know crossredit markets so you know at some point it becomes profitable try to corner the market to ex you know exploit the derivatives right because the derivatives are you know high leverage.

25:54 So what we end up doing is essentially we we monitor the depth of the market and theoretical profitability of such manipulation you know including things like you know borrowing assets from another market and running some arbitrage or cost of like costs of capital in general to essentially keep the market issuing swaps until it would start being profitable to exploit you know underlying even large credit markets to move the rate.

26:24 So I mean this problem in traditional markets is not really that serious because markets are extremely deep. Yeah. And and in crypto you know we had times where markets are very deep just to observe that within one year they would dry up so much that you know it now is you know you make 5 million deposit into a that moves the rate by 2%.

26:50 So even though you know IPOR derivatives are they're not susceptible to liquidation like if you use DEX AMM that's another reason why you shouldn't use Dex AMM for anything like this. The Dex AMM prices the tokens right. Mhm. And there is a the interest rate is it has to be baked into the price of the token and it may happen that with the you know if the rate moves like from 5% to 40%.

27:18 Yeah. You get liquidated. Yeah. Okay. Because your token Yeah. So, and the way it works with IPOR, we actually look at the time aspect. So, the rate can move from five to 40%. That's not a problem. The problem is where the problem can be is like when it stays there for too long. Mhm. So, if you have a spike of the rate, you know, to whatever number, and it's just a spike for a couple of minutes, I mean, it you don't care.

27:45 It's, it's more of a, you know, 30-day interest rate payments. overall rather than momentarily moves. Yeah. And of course that's much more expensive to pull off than Yeah. Yeah. Yeah. Because you need to maintain that manipulation and that the market but the participants come into the market right to you know take advantage of higher rates. so I mean this this is and this is again playing into other thing that we we just built.

28:14 That in itself has a stabilizing you know effect on the markets and and then the IPOR rate becomes more of a trend which is exactly what it was supposed to be from the start. Yeah. And so I think I mean that's a good leadin to this this new fusion product that you guys are working on. What what sort of problem did this come out of that maybe?

28:36 That's a very good question. So So you see where we built AMM for the derivatives when you have multiple positions open you don't really care like you don't need to pay each other each other party right all the time. So all the liquidity that you have to make the market work they're actually unused so you keep them in your balance.

28:58 So what we what we wanted to do from the start is is to allocate them somewhere so they can generate some yield for for the LPS even though they're not necessarily used all the time. they're just there to make the market and and so we built this kind of you know an asset management but then you know we realized that if you want to change something you know make a different allocation slightly different strategy for those assets to be you know making money you know what we did at the time was a bit more too rigid okay and it required a lot of work from us sorry I had sorry had the call so it was a bit too rigid So it required a lot of work from us to to make any adjustments.

29:46 Yeah. And we thought well like it's not a core of what we do but we have to fix it because it just takes too much it just too much effort and too much time for something that should be fairly simple. so you know we we run this because we in in IPOR we we run those sessions when we analyze different products we do brainstorm and go deep dive on you know whatever product there might be out there you know all the major ones we already did.

30:18 It was very hard for us to keep up with the changing market. you know make adjustments, run upgrades to contracts and we thought like this is unsustainable and you know we have those brainstorming sessions we run every other week when we analyze products and and this time we decided like okay let's let's brainstorm what would solve the asset management problem that we have.

30:41 Yeah. and at the time we have already analyzed you know quite a few protocols like Yearn and and couple others they're very similar like can we get some inspiration from what these guys do and but we wanted to be extremely flexible because flexibility at the time was not our strong part you know it's like we we are very good at squeezing you know the efficiency but is very bespoke what we do right and but let's let's make it less bespoke spoke see if we can make it very flexible so we don't need to make all those decisions up front how to make it so it's going to it's going to be easy to work with down the down the line.

31:23 So we designed this actually so we brainstormed it's almost like a internal hackathon. It was a design hackathon that we run internally and we came up with this concept and we thought well this is this is pretty cool. so we think we it can be built. So we scoped it we built it as a as a prototype and we tested it and it was it was working quite nicely actually.

31:49 So we decided to build it as a both standalone product but mainly for our asset management. So so then we run our asset management through it and it worked very nicely. and so all those assets the IPOR protocol was was keeping you know we could actually put them to work and don't do it in a rigid fashion but rather op like optimize allocation because we were doing allocation to bluechip credit markets like aa and you know morpho and so on but this time we built in the fusion fusion is built really with automation first approach so it's really built for automation And the way it works you have the vault that you deposit assets to.

32:37 So in our case another protocol which is IPOR protocol is deploying assets here. And then the management of this vault is divided into different roles. So what we call an alpha an alpha is like a manager. Okay. So alpha all it can do is to allocate assets according to the rules that are defined by by the what we call an atomist atomist is a is essentially an admin of the vault.

33:09 So the atomist says you know you can put assets to ava you can put and I'll you and tell you what assets you can put to a or you can maybe put liquidity into morpho or you can maybe borrow from morpho these assets maybe you can swap on dexis maybe you can harvest rewards from fluid so you decide you design what can be done you put the thresholds you put the assets methods you can you know even put the fees which was another issue that we wanted to solve and and then let you let alpha to automate it okay and alpha is more like a off-chain service that has you know permission to allocate but that's it cannot withdraw the assets it cannot arbitrarily put them here or there swap it or you know without

34:06 You know any constraint drains on the slippage. So this way we you know we we build this platform where it's extremely flexible. So you can we have right now run evolves running strategies where they provide liquidity to credit markets but when this is not profitable they do rate arbitrage and when this is not profitable they do looping.

34:33 Okay. So and one vault does all of those those things. Yeah. And one vault has one automation and it just constantly shifts the strategies and moves assets around and as NLP you don't care. You just put assets in and you kind of let you know let it do its thing. Yeah. So it's actually you know it's been working extremely well.

34:54 We've been outperforming some of the you know the household names on regular basis. I'm not going to talk about it. We'll have a public report. I don't want to step on anybody's u u toes but whoever wants to do the research they will see. Yeah. So that was the problem that we were solving the asset management problem and it was it was so good that we say okay anybody can put money in it.

35:17 So yeah and and is there like a vault token that comes with it as well like a 4626 token or something like that? Yeah it's a it's all 4626. so and from that perspective you can even have structures where you have like a top level vault that allocates to you know those algorithmic vaults underneath. So we actually have and this is going to be coming later this month in in integration with one of the asset managers where they use fusion to allocate between their own vaults.

35:51 So they have their own vaults. They're like a singlepurpose vaults. They're running on their own fixed strategies and as the market conditions change those vaults would become more or less profitable but they don't want to change the vault. The vault is already efficient. It's efficient doing its own thing but they use IPOR fusion to move assets between those vaults.

36:14 The vaults that they own they they manage. So it's you know so it's it's pretty cool it's pretty cool structure where we don't even compete with other asset managers. Yeah. Yeah. An excellent way to like compose different strategies. Yeah. Yeah. You can you can do that. Yeah. I guess that's like one use case that you're mentioning there.

36:34 But you know as you've rolled IPOR Fusion out like who is the ideal user or like what other use cases have you seen for the product? So the most the most users right now and I'm not talking about LPS because LPS are obviously you know they they searching for yield and sometimes they don't necessarily care you know how that yield is generated but you know we see a lot of interest you know crazy amount of interest from the asset managers.

37:02 So we have asset managers that either manage the their own assets or they manage assets for their clients or the third kind is they want to both manage their own assets but they want to charge fees by managing other people's assets like letting them participate in their own strategy. so I would say those are the three main kinds of users.

37:26 I mean I mean I would call that clients really you know cuz they come with their own strategy. say hey we'd like to curate that strategy. some of them they want to automate it also which we help them to do through the SDK. sometimes we help them to run the alpha. but you see we can run the alpha and they they take the ownership of the vault and you know we can only do what they let us do and everything is on chain right so you know where the assets are at every given point.

37:54 You know what the logic of the vault is what it can do. and it's very transparent. You see what fuses we call fuses there, what fuses are connected, what assets are whitelisted and you know in this way you can you can manage assets and you can have you know sophisticated strategies quite exotic actually. There's couple that we working on that, you know, they're really crazy.

38:18 Like if you go to our our front end, you will see there's not too many vaults right now because there's quite a few more that are waiting for the prime time. They're already built and they're just waiting to be launched. There is one for reservoir you see is making about 20ome%. we actually running an upgrade to that strategy with the customer.

38:41 I'm I'm hoping to get on a call with them later today because we need to well they need to change configuration to let alpha you know allocate assets in in in a new way to to get some extra yield points. So it's pretty we we stoked by flexible by how flexible this is and how many different customers or how many different parties can actually take advantage of it.

39:08 Like the use cases are you know very diverse. It's not just I mean it's not not just managing assets but you can do looping you can do liquidity provisional dexes in in ranges or you know you looping through the credit markets you're looping through the dex you know LP tokens rate arbitrage and those are the strategies that we're running and often we mix them on the same vault so one vault can do you know looping arbitrage or dex liquidity you know depending on what you one.

39:41 Yeah. I mean, you could even make like one of those like yield bearing stable coins with this like literally with one of the vault tokens where one of the positions is you go and you borrow and you loop up on a and then you go and you short on like a GMX or something like this. Yeah. So, if somebody's interested and wants to do that with IPOR, we open like we we see ourselves more as a platform rather than curator.

40:06 So we happy to work with anyone who you know has a you know idea for sustainable strategy you know so happy to talk. Yeah there's a lot you could do with this. So it sounds like under the hood really the way it works is you have these fuses which you can plug in and those are kind of the adapters between the specific integration.

40:25 Yeah. So from from an architectural perspective and this is probably interesting for you more than anything is this is this is oh I've now lost now lost the name this is diamond proxy right so we actually use diamond proxy on the previous product on it's it's like second version to increase the flexibility of it and and we realized that it's a very very nice way to run run the fuses.

40:59 Yeah. Because everything like the vault essentially becomes like your EOA. Okay. So it it retains ownership of everything and you don't need to hardcode functionality into the vault but rather have fuses to delegate call to other protocols. So like the fuse is stateless has no you know essentially the validation is performed on the vault with the params well with the logic from the fuse and it has no balance nothing you know it it's it's like a completely transparent way to give functionality to the vault and you can easily add it you can remove it in the hard so it's a it's a very nice architecture that you know gives you tremendous this flexibility and it's fairly straightforward with it being a diamond.

41:51 Like if a client wants to add their own custom fuse, their own custom facet to interact with like let's say their own protocol or you know some really off-the-wall protocol that isn't already supported then they could go and just write that themselves and Oh absolutely. So the catch is you know on the front end we would flag them as using unverified fuses.

42:17 And because we would like to you know because we like we are not we are platform but we would like to you know make sure that there is a review of of what can what's used so the the LPS they understand what you know what the risks are. so if you want to build your custom fuse and you want to use it and let others use it because that's another thing, you know, if you build fuse for yourself, every other vault can use the same fuse.

42:43 Yeah. So, and there is no cross-pollination, you know, like the fact that I'm using this fuse, it doesn't affect any other vault using the same fuse. Like everybody runs their own logic through that fuse and nobody sees each other's moves. Yeah. So, so that's actually a point, you know, like that's what we want. We want to see that there you know where you expand functionality like you've built fuse for your own protocol and then all of a sudden every single vault can use that from like like this from the same kind of like like that's kind of the beauty of DeFi in general as well as you know once a primitive is built then it can be used by the entire ecosystem on on the chain that it's deployed on and then also I mean

43:28 Now with interop you know literally across DeFi as a like for example like with an a like you guys are you know building off of and and adding on more functionality more more value proposition to a and morpho and all of these. So yeah I mean that the compounding effect that that we'll get more and more of as we get more and more people building cool stuff.

43:51 Yeah I mean you know what that that interoperability of of ava this is actually what convinced me to join you know all this blockchain thing. Yeah, cuz you know and actually I was stoked by flash loans. You know when when when somebody explained me how flash loans work, I was blown away. It's like this is the most amazing thing I've ever seen.

44:13 So yeah, when you say a you know it's u you know I I love a is beautiful. I mean there's there's also you know morphos and compounds. Yeah, of course we don't want to but I like a I like a I think it's a beautiful design. Yeah. And I also wanted to touch on so you mentioned a lot of the offchain stuff that you guys have like alpha to basically go and actually execute these strategies and carry stuff out.

44:38 What kind of logic exists there? Is there any kind of like or like that? Oh yeah, it's it's I mean I I don't want to spill maybe too much of the secret sauce, but Okay. But you know when you run those strategies there's a lot of things that you have to consider, right? You know when you do credit allocation you mainly concern about things like slippage, right?

45:02 Because let's say I'm I'm I have $5 million and if I dump it in some small pool on Morpho, I'm going to sway the interest rate so much that you know I I was better off dumping it in fluid or or or or Aave. So what what alpha does at this level is essentially looks at the slippage and it simulates the market, you know, because it it can allocate to multiple markets at the same time.

45:28 So it will will seek the highest overall profit when it break down you know will break down assets into multiple protocols. Yeah. You know to a point where you know sometimes it may deploy less then you know when you look at the numbers you say oh there one market is like 8% the other market is 7% and there is more allocated to seven than to eight.

45:53 And it's like why wouldn't you delegate more to eight rather than seven? And the simple answer is that's actually the more profitable approach because you keep market above the kink with lower amount of money. So when you do the math overall you're making more. You're going to have better rates normally but you will you will actually make less if you you know if you drop the rate below the king.

46:21 So alpha actually does this 4D chess you know where it actually maximizes return. Very cool. but when you do things like looping then it becomes much more complex because all of a sudden there is more to u more to watch. There is a you know you need to worry about your LTV and growth of your both debt and the collateral.

46:44 You know, you anticipate things like if you want to unwind, what's going to be the cost of unwinding because maybe you need to swap and if you swap, how deep is the liquidity? Same thing if you want to ramp up the the strategy. So for instance, if you take one of the vaults that we have right now, the reservoir vault that is allocating to srUSD loop, you know, we had a lot of like there was a tremendous interest in the beginning and people start putting money into that vault and suddenly APR dropped okay and the reason for that was the A alpha couldn't allocate any more assets.

47:21 Mhm. if it would allocate like it would allocate maybe 30% of assets and everything else was unallocated and and we were asked like oh why why don't you allocate more I mean you can maybe use lower leverage but no because the interest rate model is such that if we allocate more at lower leverage the vault would make less money so so we the the allocation had to be kept lower.

47:46 Yeah. So what we did we expanded the strategy to allocate remainder into liquidity on morpho which would drop the rates and then allow us to loop more at the higher leverage and then increase rate from like you know 12% to about 21 and now we expanding that strategy even further where it can loop into USDC rUSD and that will increase it by you know a little bit more But it will pull the assets from the liquidity which will drop the APR then will increase the APR through the other loop.

48:24 But as a consequence the rates on the loop would be higher which will have the consequences for other manual loopers on this pair. Yeah. But you know that's I guess their thing to worry about. So it's it's really like I'm so sto every time I'm thinking about it. It's like like this is such a nice thing. This is a finally this is a finally product that I can use.

48:49 Yeah. You know, and I'm I'm a risk averse person, you know, like I'm more of a traditional mind. Yeah. Like I said, I'm a like before we started the call, I told you, I'm a I'm a more of a gold bug type person. so yeah, now we can and you know, the liquidation protection, all of that, you don't all of a sudden don't need to worry about.

49:11 Yeah, I mean, well, that's that's fascinating. I think one of the hardest problems is in DeFi, like you said, we have limited market depth and it starts to become a really interesting problem when you start to try and scale a fund like this. Like I'm not sure how much TVL you guys are are working on allocating right now, but like I can imagine if you 10x if you 20x that amount, you're going to have whole different problems of like how do we allocate this?

49:36 Yeah. So, so that's the thing, right? So if you allocate $50 into credit markets, you don't you don't really worry about slippage, right? Yeah. but but the alpha that we built for at least for the credit optimizers and I mean for all the vaults that we automate personally because there's other vaults that will come into production that are managed by other parties and but those that we built at least we can probably scale them to a very large amounts and we can manage them from small to large because of the simulations the way we run simulations that's and this is another beautiful thing because you don't need to change the strategy you just adopt the avenues where you can allocate assets as you scale because you know you can allocate assets just the rates will

50:23 Drop. So what you do is you know you think about okay maybe I change the strategy you do this other thing as well and then your strategy evolves and evolves and evolves with the market with the size and you don't ever have to tell your ops to move your ass move their assets they all stay in the vault and your vault can be evergreen.

50:43 Yeah, and this is where it seems like you guys had incredible foresight on making it as flexible as possible because there's going to be liquidity pools and you know areas where you need to be able to get access to and with just being able to spin up like a new fuse you can pretty easily. Yeah, that's exactly right.

51:02 That's exactly right. And yeah, so we're pretty happy about you know how it turned out and even though markets a crap and everybody's thinking about it. I don't know when this is going to air but for the record you know today markets are crapping themselves maybe maybe a bit better than yesterday but but there is a demand for you know safety I would say you know and lowmaintenance allocation that is efficient and transparent.

51:28 Yeah, of course. And another thing that of course if you guys are you're you have this standardized infrastructure and you get this this distribution effect where if you expend all the resources to first of all, you know, make the smart contracts and and manage the the alpha offchain aspect and all of this stuff, but then you've also got this trusted vault that can be distributed to tons of clients and stuff like this where it's been security reviewed.

51:57 It's got all like you can secure this one piece of code and then distribute that to many different clients. It could be different kind of clients and they can also inherit the same sort of security principles that you guys have already invested everything into. And so I think that scales very nicely and it gives like like you said like people at the end of the day they're worried about safety and they want to have efficient allocation.

52:21 Yeah. And and you know when we talk about safety there is a lot of stuff built into the vaults that are you know for the safety reasons for instance you know every user that interacts with the vault can receive a time lock. So you can define a time lock on the user that can change settings of the vault.

52:39 Yeah. so you know I can add fuses but we can say you know this particular vault requires a week before the the change of fuses will go into effect. So everybody can re review you know whether they're happy or not and move their assets if if they feel that's the right way. So that's one thing you can you can actually delegate your management to DAO pro DAO tools like Aragon.

53:03 So by you know by from the from the get-go the vaults are compatible with the open zeppelin vaults. So if you want that you can have your u you know your vault track your balance over time and then that can be used by Aragon to manage your voting power and then because we use open zeppelin access manager that also integrates right into Aragon.

53:31 So there's a lot of things like that that you know we were trying to put into the core vault from the from the get-go and and there's I mean I can nerd out about details but I'll probably bore people to death. but I mean if somebody is interested yeah I'm happy to continue that conversation. No I mean with all of this like it seems like once we get this infrastructure up there's so many different things that you can plug into it and adapt it to use.

53:57 What what are the main things that you guys are working on right now? what's what's biggest on the pipeline for you guys at this moment? Yeah, so the the biggest thing right now for us is how we can how we can standardize our alpha approach because we have so much demand to run alpha that you know if if we only could you know we would have about 40 pools open right now or vaults it's just some of those strategies very complex.

54:26 So we we now thinking how we can change our our code base really to make it easier and faster to assemble those strategies on alpha side because I think right now it's fairly bespoke. I mean it's extremely detailed but it's it's it's almost a custom approach for every single vault and we thinking you know how to improve that so we can bring more vaults online faster.

54:54 I mean, as for the vault, as far as the vault goes, we consider it feature complete, at least for now. So, we we mainly focus on on alpha. Yeah. I mean, it makes sense. I think I mean, it sounds like that's your real secret sauce is that like offchain management. Well, we we'd like to have actually more parties to bring their alpha skills up to the the platform.

55:20 You know, if if you like if anybody's listening to this and they want to run automation, they know how to write Python or Java or whatever they want, you know, TypeScript and they want to run automations for, you know, managers with millions of dollars. I mean, you know, hit me up. we'll hook you up and you know, we very happy to, you know, for others to step in.

55:43 I mean, we we enough demand not to worry about, you know, being bored. Yeah. So yeah and I mean if in that sense you guys become the platform and you're connecting the the offchain models and the managers with the folks who want to allocate capital. So yeah I mean there's a lot of opportunity there. But is there anything else that you want to sort of shout out?

56:04 What are you guys working on? Any any kind of like offers that we can draw people towards like obviously debugging? No I mean yeah of course you know if you you know everybody welcome to visit app.ipor.io io to see I mean it will take you straight to the IPOR vaults the fusion vaults if you if you you can also participate in the in the liquidity provisioning for the for the derivatives so then you get yield from the derivatives and also from the vaults so it's like a double you know double benefit if you want and yeah we will have more and more vaults coming in in April this year and you know beyond that so just keep your eye and and see what sort of strategies come up and they're available.

56:49 Excellent. That's awesome to hear, man. Well, I'm looking forward to more of the vaults coming online and and seeing what kind of fuses you guys get up and and running and all the asset managers that come on. it sounds like you guys are talking with a few potentially big ones. So, excited to see what what comes out of that.

57:06 And of course, we'll have Twitter and website down below. So, if anybody wants to go and check out IPOR and check out some of their products that we talked about and just generally see what see what they're up to, you can go and see that in the description below. And yeah, dude, appreciate you hopping on and and taking the time to discuss with us.

57:24 It was awesome to hear about what you guys are working on. Oh, thank you for having me and you know, thank you for this wonderful conversation. Absolutely. Thanks, Lukasz. Okay, my pleasure.