On The Brink with Castle Island - Michael Bentley (Euler) on Euler's Comeback (EP.650)

Episode Date: July 28, 2025

Wyatt sits down with Michael Bentley, the founder of Euler. In this episode:  Euler's founding story, infamous security incident, and comeback story How the team and protocol stayed together in the face of immense challenges The value of personal relationships in crypto despite a pseudonymous culture

Matt Walsh and Nick Carter are partners at Castle Island Ventures. All of these expressed by them or the guests on this podcast are solely their opinions and do not reflect the opinions of Castle Island Ventures. Guests and host may maintain positions in the assets discussed in this podcast. You should not treat any opinion expressed by anyone on this podcast as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of their personal opinion. This podcast is for informational purposes only. Brought down by bad mortgage investments, Lehman, which has 25,000 employees will be liquidated.

The federal government loans American International Group, AI, $85 billion. This is a different kind of market, and the Fed is asleep. The federal government is stepping it to stabilize Fannie Mae and Freddie Mac, the two mortgage giants that have been threatened by the housing crisis. The Bank of England has pumped 75 billion pounds more into Britain's ailing economy with a new round of quantitative easing. And print a couple trillion dollars, and all of a sudden, people start to worry.

So out of this worry, we have something called a Bitcoin. Bitcoin. The story behind Euler Finance is one of the most gripping in crypto's recent history, and it's the one that we're going to explore and tell today. Michael Bentley founded Euler in December of 2020, alongside Doug Hoyt and Jack Pryor, with the intent to build a decentralized crypto lending and borrowing platform out of his hometown of Leeds in the UK.

Unlike many founders who entered crypto at the time from finance or software, Benley actually came with seven years of postgraduate research in computational and mathematical biology. That background became foundational to Euler's design, which we'll discuss further. Benley drew inspiration from control theory. A branch of engineering and applied mathematics focused on influencing dynamic systems, like biological processes or machines, to build Oilers' interest rate model.

Where other crypto lending platforms at the time relied on manually set rates, Bentley and the Euler team implemented feedback loops that allowed interest rates to adjust automatically in response to market conditions. As you later put it, what got me excited was the potential to apply my background in system dynamics and control theory to a new kind of open, programmable financial system, that being crypto and smart contract enabled finance. Oiler raised multiple rounds of venture capital and launched on Ethereum in late 2021. Despite immediate macro headwinds, Bitcoin plummeting from nearly $70,000 to below $20,000, oiler gained traction, attracting over $500 million of deposits and establishing

itself as a key player in Defi. Then, on March 13th of 2020, Three, disaster struck abruptly. In just 15 minutes, Oiler was exploited for nearly $200 million, one of the largest hacks in Defi history. The Euler team launched a three-week hunt to identify the attacker while managing an understandably shaken community. Against the odds, they succeeded, not just in tracing the individual, but in opening lines of direct communication. Bentley ended up developing a personal relationship with the attacker and through weeks of negotiation and coordination, oiler pulled off a stunning full recovery of the stolen funds. At that point, most assumed Euler would fully wind down.

Return remaining capital, shutter operations, and walk away. But one thing became apparent. The team widely had ambitions to continue building in this space, even if it would not be with Euler. Conversations among the core contributors revealed this shared ambition to keep building. So they asked themselves, if they were to keep going, why not keep going with Euler?

ultimately, backed by the full support of their investors, the team regrouped and rebuilt. Eleven months ago, I spoke to Michael just before the relaunch of Euler v2. At the time, the protocol had only a few stray dollars in deposits and faced considerable skepticism. Fast forward to today, Oiler has grown to over $2.5 billion in deposits, nearly five-x its peak before the hack, and has earned a reputation as a leading lending and borrowing platform across multiple chains and asset types. It's a story of conviction, resilience, and redemption. Today, I had the privilege of sitting down again with Michael to talk through that journey and the road ahead. For disclosure, Castle Island Ventures is an investor and Euler and remains a strong

supporter of the team moving forward. Without further ado, here's my conversation with Michael Bentley. Michael, first off, I want to open by saying, I can't speak to the amount of respect that I and the team have for you and the Euler team's resilience and commitment to everything you've done in the wake of the experience that you've had over four years. And we speak about it often as something that we deeply admire. And it's incredible to have you back after we spoke about a year, a year and a half ago before V2 launched. And it's incredible what's happened in that time span. So I wanted to say that first. And I find an incredibly moving story because in the backdrop of crypto, which so often functions with anonymity being a norm,

You can't tell who wallets are. You can't tell who's governing protocols. There's a deeply personal aspect to the Euler's story in that you knew who the person who exploited the protocol was. You've spoken about the personal struggle that you had during the time. The team opted to stay together when that was something that was very much in flux. So it's very unique, really. And I think maybe it speaks to pieces of a narrative that can be essential in keeping a protocol

like that together in the midst of real challenges. Yeah, crypto throws up these wild stories, and you never ever want to be the main character in one of them. And I think coming into Building Oiler, I was well aware of the high-risk nature of the industry. But what happened was just completely crazy, you know, it was a really, really awful experience, obviously, but also just a really crazy, the twists and turns I now saw it just beyond anything I can possibly imagine coming into it, honestly. And there were lots of human elements. There's always the people behind things. I think if you look on crypto, Twitter, there's like a gallows humor that people show on.

Because bad things happen fairly often in the industry or have them historically. And I think the gallows humor is kind of healthy. It's part of the culture, and I think that's good. It sometimes does overshadow the fact that there are real people in the background that are kind of suffering. Part of the reason we're able to even tell this story is that you've deliberately made the decision to be a public-facing figure within the crypto industry, which I would say especially early on, many did not. And I'd say it's thoroughly appreciated by people who are users, people who might have had funds lost,

by people like us who are supporters of the protocol today. How did you make that decision and how do you look at that decision versus taking the road of being potentially more conveniently anonymous? Yeah, I think it's important to build trust. It's not something I'd ever considered. I don't come from a kind of cyphabunk background, I wouldn't say, and I was an academic before there.

No one's unonymous. in academia, although they might effectively be academic. Names are out there, though. As far as the population is concerned, but yeah, the names are all on the internet, but no one needs cares about them. So it wasn't part of the culture coming into it, so I wouldn't have considered it necessarily.

It was kind of new to me coming into the industry to see that so many people had that, and I completely respect that, especially given all sorts of reasons to want to be anonymous or to reduce the extent to which it names out there, and I respect all those reasons. But yeah, for me personally, it felt like when you're building something,

that does involve money and people are putting their trust in you, it's important for people to have somebody that they can kind of go and say, hey to and interact with. They can see that it's somebody that's thoughtful and trying to do things to the way behind the scenes. It's a little bit scary. You can probably be in public as well.

The next point happened, I'm sure people must know the amount of kind of crap that you get during that period is quite intense. Yeah, how did you deal with that? Well, I've always just had this philosophy that obviously people are going to do that. they're upset, they're frustrated, people are also wanting the merchants online, and people like to do it for fun.

You just can't let it get to you ultimately. I've got friends and family at home who know what kind of person I am. I've got my colleagues who know what kind of person in there, the kind of relationships that matter to me. So I kind of tried to brush it off a little bit. Some of the things were particularly awfully. My son was born like four days before the exploit, right? And some people decided to use him in some of the messages.

It's terrible. I thought that was particularly overstepping the mark. But they weren't too many like that. Yeah, you just have to be pragmatic, realistic, and accept that that's just part of internet culture. So trying not to let it get too, too much. Was that a lot of the community that responded at time, or how did the community at large respond to that period?

Oh, it was a real mix. I mean, I would say most of the feedback and the responses were really good. I mean, we got lots of really sympathetic messages from all the developers and builders in the industry on Telegram. Obviously, we'd been trying to build things the right way, So we were very friendly with a lot of the security community as well already, and they knew that we tried to do things the right way. And this wasn't through carelessness or anything on our side

where we deliberately cut corners or anything like that. So we had a lot of sympathy, I think, a lot of respect from people in the industry already. So we've got a lot of supportive messages, honestly. And on social media, you got a lot of interesting theories, right? There's people saying that maybe it was an inside job. Maybe it was me. Yeah, that kind of thing comes up and you see it.

And I remember my wife at home was reading some of this deterrents who was saying, I don't know how you don't respond with pure just anger and mage these. I said it's just not helpful. It's just the last thing. When an incident like that happens, you need to focus entirely on trying to get the best outcome possible and letting silly people and their messages online get to you is not helpful. And it's not a good use for your time to respond to all that. And I also said to our team, were some people like writing messages back?

And I said it also sends the wrong signal. it sends the signal to people out there that you think that that's more important and that you've got the time to do that rather than trying to solve this case, right? So I said, yeah, we just like it, ban on communication when it comes to responding to people's messages online. And that's how we approached it.

How did the team stay together at that point? Well, I think we all like each other and all decent people. So there's a really incredible story within OILO actually. It involves a person that those kind of shirreelians during that time. We hired my co-founder does lecturing universities and colleges and things from time to time. And during one of his lectures, one of his students came up and started talking to him afterwards,

showing a lot of interest in cryptography and blockchain and so on. So he invited him at all point to become an intern Euler. And, yeah, he went on from being an intern, to being a smart contract developer and a senior developer and a sort of leader of the smart contract team and so on. Anyway, during that period, when the exploit happened, I remember me talking to the team, frankly, about their roles and the likelihood that they stay on and everything. And this individual Casper, he said to us, I'm not going to stop until we catch this guy. We've got a chance to catch this guy.

We have to catch him. Don't care about the future. Let's focus on just catching him. And yeah, he's now actually CTO of our company. So we decided to promote him to CTO. My co-founder is more like a presidential place more like a presidential wall. So he's one of the people that was there in the trenches with us during that experience. And most of the team that we've got now are all there as well.

So the team really have stuck together through thick and thin. I think we all like each other, all have a lot of respect for one another. It's a very smart team. It's very strong engineering culture. So, yeah, I think that really helps. I've been in other war rooms and things trying to help out other projects. And sometimes I realize how lucky we are that we all feel that way. There was no finger pointing.

There was no like, oh, you own this to life of code. We all just wanted to get the absolute best result. And we were hyper-focused on achieving that collectively. And that's actually not the normal when you go into some situations, you realize there's quite a lot of finger pointing and disagreements. So we're very lucky that we have that kind of team, I think, and that kind of culture of OILA. Yeah, especially. CTO come head forensics and FBI and the whole thing, right? Yeah, yeah.

He's done fantastic at Oiler. But that's the kind of culture we have. It's very egalitarian. We don't have, like, big hierarchy. I wouldn't say within the organization, it's fairly flat. So everyone can challenge and be challenged by the people. And I think that kind of fostered mutual respect and open dialogue and things. So, yeah, people tend to enjoy working here, I think, and thrive when they do well.

We like to promote them. And I love in the story that I thought was interesting was you guys decided to keep the Euler name and brand as opposed to potentially doing a rebrand. What went into that? It was a big discussion. I'm not going to lie, we spoke to a lot of our stakeholders and investors to get their thoughts on this. There were a few different schools of thought. The reason we kept it was because actually we did have a strong brand name before that.

We had a brand that's associated with engineering and building quality products and security, you know, the rest of it. So we had a good brand name coming into it. And then that was part of the reason why the response from the border community was so strong. I think, as to what happened. So that was one thing. It's a good brand. And we didn't want to then kind of water that down and make it look like we were hiding because we didn't feel like we have anything to hide, right? This is a really tough industry to build in, and you can do everything possible to secure a protocol.

But unfortunately, there's no guarantees in life, and it's a particularly challenging industry to build and like software. And we've seen it's happened to many, many good projects down the years that suffered on security issues. So there's nothing to be proud of with the security incident. Obviously, but there's also not necessarily something to be ashamed about as well. If you've done your best and tried to build in good faith and everything and something's happened, I don't think you should necessarily try to hide from that. And that was one thing that we didn't like about doing rebrand was it might look like we're trying to sort of hide this and pretend that like, oh, there was some different thing.

And we're now like a completely new team. And that didn't really fit with our ethos about being really transparent about things. We tried about the whole process to be as transparent as possible. Overall, we just felt like that was a better way to continue to build the brand and build trust is just through being really open as open as we could be about everything that happened and focus on that and fully own it as part of. who we are and our story. But yeah, I mean, it's a strange brand because on paper, it's not a great one, actually,

in terms of just from a pure marketing standpoint, the name OILA is derived from a mathematician from the 18th century. It's pronounced differently to how it looks like it should we pronounce and all the rest of it. So there are many good reasons for a branding and marketing standpoint, I think, to ditch it and to put some shiny new name on. But just in terms of the history and who we are and what we've done, I think we felt like it was the might thing to just keep it and be forward with our story. I think you make an important point, too, if you're innovating, and I think this goes

in crypto or beyond, you expose yourself to some level of risk when it comes to what can happen. And you guys early on, I'd built a truly innovative dynamic interest rate model. And I've loved reading and learning about how that was informed by your background, but it seems to shy away from innovation if you're going to do so in fear of what can happen. Yeah, I mean, everybody that's doing anything innovative takes risk in our industry. The risks are quite high, I would say. They're not necessarily as risky as they would be if you were building an airplane or a new type of car, right?

So there's definitely higher risk industries out there. Innovation is a matter of life and death. So it's not quite true here, but the innovation is very challenging and opens up a lot of potential for risk. That's absolutely the case. That doesn't mean that you shouldn't try to innovate, though. It's a part of human history is trying to push boundaries. is trying to move things forward. Nothing worth doing is ever easy.

So that's where we're at. I mean, they come back from a background in science, right, where a lot of the issue of science is just littered with mistakes and things that went wrong. But science moves the world forward

and ultimately is made the world a better place for quality of life and things. And I think in crypto, not necessarily, doesn't have that direct impact on quality of life necessarily, but it is a really powerful thing in the background about making finance more efficient and transparent and open and all the rest of it.

I really do believe in. There's lots of kind of garbage going on in crypto at the same time. It's just a technology at the end of the day, and it can be misused and abused and all the less of it. But there really is something, it sounds a bit trite, but there really is something there to it. And yeah, that makes it kind of an interesting and challenging environment to work in both definitely one that's worth doing. Yeah, identify that attitude as a strength that you and the team have clearly had since

early on. And in that vein, I wanted to ask, if you look at the full lifespan, including the exploit of the protocol, is it similar? or the same strengths that you think Euler v1 had and that you guys had from the outset that carried the success that you've had in the past year? Or was there something different? Was there something that you learned?

How would you compare how that evolved? I think it's mostly the same strengths and the same attitudes that we've had since we started, honestly. I think that nothing has changed too much. I mean, we still build in the same way we used to build. Obviously, you adapt and you learn things over time, but I think it's the fundamentally the team

has that just kind of engineering, scientific, sort of, yeah, approach to things. And I think that's served as well. Oh, we're very meticulous, very thorough. We're very collaborative. And that's really how we've built Euler from scratch. People ask us a lot about why we've had success. And part of it just comes down to the technology. I think people sometimes forget belting crypto because there's so much that's dependent

on hype and tokenomics and all these other things that are nothing to do with technology. But in Euler's case, that's not really the case. We didn't come in with M.B. hype, we came in with like anti-hike, we came in with intense skepticism about whether we could survive and thrive again in the industry. So, yeah, that anti-hike means that we had to have something that was really offered something unique and new and that was innovative and help people achieve things that they couldn't achieve with other tech and so on. So it's really been a tech-driven period of growth for us. I mean, we've also pushed out on other areas, of course, you know,

in terms of business development and other innovations. But yeah, tech is the main component that ultimately comes back to who we are as. people, what we like to do, which is build things and engineer things. I like the point that you make about the just technical and building first attitude that you guys have. And I think you alluded to the fact, we certainly go through periods in the industry where I don't think that's the case, where I think a lot of the people that are drawn to the industry come from backgrounds that aren't necessarily aligned with pushing forward the best technical outcomes. Would you like to see more people from, say, a mathematics, a computational biology

background who are coming in with maybe a problem solving or an efficiency perspective and a more robust technical background into the industry? Do you think that's something that we would benefit from? Yes, I know. I think you need a mixture of people. On the technology front, I think we already do have a lot people like that. I mean, they just tend to not be as visible, I would say. Like, even at Euler, There's not many people that are visible. People sometimes talk about Euler like I'm Euler, and that's just because I'm on Twitter and everyone else is. And if I could not be on Twitter, I wouldn't be on Twitter, frankly.

It's not the environment I feel most comfortable in. But it's important for me to be visible, I think, for the project. But yeah, the rest of the team here are not that visible. I think that's true more broadly in the industry. You know, the incredible tech around a lot of the podcast course people use, but most people will never see all the little micro decisions that went into structuring things the way they are. There's a lot of deep thinkers and sort of academic types.

and computer scientists, mathematician and things, building all the stuff behind the scenes. So, yeah, I think we do have enough of them. It's how you bring that stuff to market, though, and how you do the sort of growth phase and how you distribute the technology. There's lots of interesting things about our industry that make it easy for the wrong types of people to benefit

and to kind of miscommunicate the way that things work. I mean, just one thing is it's, it used to always be the case of, like, the forks of both calls, where somebody else does all the hard work to build it. someone just forks it, slaps the token on, and then they go out and sell it. And the average retail person, the average person that's not paying attention day and day out can't tell the difference between the original and the thing that was forked, right? All they see is it's a protocol.

It allows me to swap token A to be, right? And that's all they see. But I think it's good for the industry when the people that built the technology are the ones that benefit from it financially. The forks are the ones where tend to be the ones which are most heavily marketed because A, the team that built them don't have. have that engineering background. They're more specialized in marketing. And B, they don't necessarily have to spend all the money on the building the tech, so they have a bigger budget, tend to spend on the marketing. And they're maybe not there for the long term as well. I sort

believe that builders tend to build things partly for the upside, but also partly just for the sheer love with building stuff. That's certainly true about it. And I think that's true more broadly in the industry as well. People, engineers just like engineering stuff. We like tinkering and building stuff. And we would probably be doing it whether we got paid or not. there's a desire to solve problems, right? Yeah, and it's fun, right? It's part and parcel of it. So I think, yeah, builders are one of the best parts of the industry, for sure,

but you can't just build things and expect people to come. There's a really painful lesson. A lot of people that come into the industry learners, you also have to have distribution down and figure out how to communicate these ideas and the technology to the end users. And so, yeah, there's other skill sets, which I think are definitely needed.

You make a great point that there are a lot of brilliant people in the industry who have that luxury of not being. scene and they take that view and hear you and I are talking on the podcast. Yeah. Well, this is distribution. You have to communicate, right, and talk to people, let them know who you are, what you're about, what your motivations are and so on. But yeah, a lot of our team, they're not anonymous, but they might as well be in some sense

because they don't have a public presence. And so they're the ones behind the scenes doing a lot of the engineering work. Do you think Euler today would look different at all had it not been for the exploit? Yes, yeah, I think it would be far more successful. I think prior to the exploit, Ola was taking market share out from competition and growing very, very quickly, and I can't emphasize how much of a setback it was. I mean, it's so much harder to come back than to do it the first time around. Is that because partners were reluctant to work with you, or what were the major parts of that setback? There's a whole bunch of factors.

Firstly, liquidity is a powerful asset, and so when you have liquidity, it makes your approach call more useful than, when you don't have liquidity. And that grows over time. It's one of these things that just takes time to build. So having liquidity, resetting that back to zero, having hit almost a billion dollars of liquidity before, I think we sing from zero and then having to start from scratch, that's the challenge.

I think Lindy and having the Lindy effects around for longer, there's another one slightly related, but it's more about just like the security of the protocol being battle tested with funds at risk for a longer period, shows that it's less likely to be exploitable than something that's just brand new and shiny. There's also a lot about distribution in terms of tokens. Some of our competition have benefited massively from launching a token

with a high valuation that they could then use in deals and negotiations for partnerships. If you have a high-valued token that the people are optimistic about putting a million here, a million there for the deals or partnerships, it's absolutely nothing, right? It's just, you know, if you've got a billion-dollar valued token, you can use that very, very strategically to grow very, very quickly. But when you come back from the grave, basically, there's no value in the token. It's a liquid. So it's not really an asset to the project anymore.

It's almost like a liability, honestly. It doesn't necessarily help the projects grow. And so tokens, whether people like it or not, are a very important distribution mechanism, like shared ownership that you can get in the early growth phase and project when people are optimistic. And it's a really, really important mechanism. And some of the best projects in crypto have used that massively to their advantage. And we haven't had that luxury, unfortunately. So yeah, I think almost certainly Euler would be a much bigger, stronger protocol today if the exploit hadn't happened.

What resonates with users from Euler as a protocol? When you take your average user, what do you think attracts them? And you mentioned the landscape of other players in DFI. What stands out? I think Euler has always been a place where we've not tried away from offering Fibera worse. And it's no necessarily right way or wrong way, but we've not shared away from offering more advanced features on the protocol and allowing people to do more advanced things. So we've always had like a user interface that allows people to batch transactions, for

instance, and build like more complex positions, then we balance those positions more easily. And so, yeah, some of our competitors were really just adding this today in 2025, right back from 2021, we had a lot of these cool features. So that engineering, I think mindset that we had benefited us because we attracted people that really appreciated those kind of more advanced features rather than just a deposit or just borrow. There's more you could do. The vertical is more efficient, I think, in many ways,

more capital efficient. You can see that in the charts. There's higher of oil utilization. People tend to prefer borrowing on Euler than other platforms. In some scenarios, not all of them. Certainly, if you're borrowing from something like half a year, they have a lot more liquidity. Sometimes it's easier to do that because of that.

But on the purely technical front, I think if we make borrowing easy on Euler, We show people lots of information about their borrowing positions and statistics and things that are in other protocols which help people manage risk a little latter. We don't penalise people when they're liquidated. We try to minimize the costs from liquidations as much as possible, whereas if you get liquidated on other protocols, it can be really devastating. You pretty much usually get wiped out, I would say.

Whereas on all of that's not the case, you usually are able to keep almost the net asset value of the account is retained after the liquidation rather than going to liquidation fees or liquidation penalties. So, yeah, I think borrowing overall has been a big thing that people tend to like, oh, or not another than other protocols. And when you build out product functionality or build out markets on the protocol, do you do it with a bit of a sandbox mentality

of this is functionally something very good? Or do you do it more with a view of, okay, people will do this because they'll want to go then do this on pendle or etherfi, or you're thinking about how those use cases are ported elsewhere? Good question. I said this is something that we learned from V-W. Firstly, I'd say that just our engineering style is to always try to build things in this modular way.

So we try to abstract things and then package them up into little modules that you can then recombine. What's useful about that is that things that come up that people end up asking you for that you didn't necessarily anticipate, they can usually be done out of the box by then recombining the modules and allowing them to interact in different ways. So you can sort of use as more of like a kit, a developer's kit for building products. So we don't start with the products as such. We start normally with the underlying engineering in mind. Some people would say that's the wrong way to do things. Many people advocate for building specific products for specific users

and just tailoring to a particular use case. But I would say we do things the other way. We sort of build from the ground up and tend to try not to anticipate all the possible use cases, but assume that we can hopefully build any kind of product on top, any kind of credit-related product on top of the kit that we got all over. So we hope to not have to keep releasing the V3, V4, V5 and so on. We want Oeolid to be as flexible enough to add new features, some of which I'm sure we'll talk about in the pod today.

Yeah, so that's our general philosophy. What have you learned about user behavior over that time? Oh, he's changed massively. I mean, the use cases that people used to do in V1, mainly doing V1, are still there today, but many more use cases today. So in the early days of DFI, people were mainly using lending protocols to borrow against volatile crypto assets. And they do that for a few different reasons.

Sometimes they want to borrow against the ethel of Bitcoin because they want to go long. So they let them redeposit back into ethel bitcoin and amplify their exposure. Sometimes they wanted to use their assets in the real world. So they just borrow against it and then go buy a car or pay for something in the real world. Or take a bad idea of another opportunity out there in DFI.

It was mostly loans against volatile assets in the early days. And I think that changed with state teeth. When state teeth emerged, we had yield-bearing forms of ETH, and that was like the first time that we really started to see growth of a different type of trade, which was more of like a carriage trade. We deposit something that's high yield, then borrow something that's lower yielding, and then you can scale up or down your exposure to the yield, basically by a rebossessing back into yield-bearing asset. And that's just completely exploded as the main use case, I'd say,

for lending borne protocols in the past 12 months, which is why protocols like Euler and also morpho have done ever so well during that time is just that these kind of carriage costs. rates are now the main use case for lending protocols. I would say people are arbitrageing interest rates and capturing yield from stables rather than just borrowing against volatile assets. Which I look at as positive to go back to the root of our discussion, it's market efficiency rate discovery. Yeah, I mean, it's interesting. I think there's this tendency a lot in FAP finance where you have assets that are volatile. And what people want to do is they try to compress that volatility as best they can. Now, you can always compress volatility, but you can't, in my opinion,

fully remove it from the system. So, you know, if you've got something that's volatile over 12 months, you can make it less volatile over 12 months, but there might be one really big spike in that time. I mean, seeing that's kind of what's happening with the change in crypto today, things like Athena and so on, where still benefiting from the yields on some of these volatile assets and things, but we're finding ways to effectively compress the volatility into products that are more predictable, I would say, at least on short timeframes, and then people focus on trading rates

with those rather than focusing on directional trading where the thing's going to go up or down. They try and turn that into a Yudelburn product instead. So I think that's like a common trend and overall probably a good thing, but one downside to it maybe is that it allows for more leverage to build in the system and less predictability in the long term.

If you wanted to go leverage long on Bitcoin, there's probably a ceiling on the amount of leverage you could take, let's say it's 5x or something, where there's like a sane amount of leverage that you can potentially take if you earn some Bitcoin. I want it to go long. But if you have a stable coin, it's yield-bearing, maybe you get 20, 30, 40x leverage out of that stable coin. And yeah, over a short time frame, that might be perfectly fine. When it does eventually unwind, it can do so maybe even more badly than the people who are trading on leverage on Bitcoin.

So, yeah, I think there's tradeoffs with the way the industry is moving. But, yeah, certainly stable coins in your bearing assets. work quite well with lending protocols, I'd say. The two have very strong market fit with one another. I want to just highlight that point because I think for people who aren't active in defy or crypto, one of the most underappreciated aspects of the market is how speculative people are on stable coins through looping to the point you make. I think people think of speculation as being on meme coins and alt coins,

but I fully agree on the point that you're making that people are speculating on stable coins. Yeah, it's a huge trade, but it's the same in traditional finances too, right? It's nothing unique to our industry. Yeah, trading rates basically and leveraging stable coin yield, I think is probably the biggest trade in the market today. Maybe it's already the biggest. Certainly if it's not today, I expect it will be in the nearest future. To touch on another piece of this related market structure, you've probably seen the same. We've seen a lot of these new CDP or collateralized debt position protocol. is coming about, which to me makes it clear that people have a desire to hold exposure to their alt coins, but to essentially park them somewhere and instead do all their financial things with a stable coin, which gets to your point of they want to take whatever stable coin they've now received and maybe go do rate arbitrage on it with a different stable coin. My question seeing this is, do you think eventually we get to a point where the alt coins almost become more idle and most of actual

defy finance happens using stable coins. And I say this seeing you go on oil or amorpho writing these platforms now, the top assets are stable coins. They didn't used to be that way. So I'm curious if you think that is the vehicle that defy is happening with, at least from a transaction perspective, long term. Yeah, I think so.

I think it is. It's just much easier to move these stable coins around and also to handle them and to price them. So like I say, there's no free lunch. I do think that over a longer time horizon, we will see some of them de-peg quite violently or wind. So there are definitely challenges with these things. There's no free lunch in finance ever.

But they simplify the day-to-day experience rather than holding strictly volatile assets. And I think that's why a lot of traders prefer these things. And where we're seeing so many, it's tough growth in that particular sector at the moment. Thinking over the longer term, it feels like non-USD stable coins or FX markets could be something that could be interesting for a platform like Euler. Do you see that as being a potential defy reality? I do over some time horizon, but it's not today. Why is that?

I mean, I've talked to people a lot about this particular topic and whether or not people will start creating FX like markets on oil and no other swap. But yeah, there's been some challenge. I mean, like the dollar is just so powerful. It's the currency that everybody likes to use, use for international trade. No one wants to hold euros, it seems, on chain at the moment. I think Circle have made a really big push with your C,

and it hasn't really taken off today. I do think it will be like that forever. I think at some point it will start to move and we will see more FX like trading. I mean, there's more, more assets are coming on-chain all the time. We're seeing stocks, equities and tokenized funds and everything. So it would be strange to me if FX doesn't grow, but there are some disadvantages with on-chain trading.

Off-train FX, the amount of leverage that people can take is absolutely huge, and the order books and everything very, very efficient. And things are a little less efficient on chain, I would say, at least today. That makes it a little harder. Perhaps to trade these things as profitably as he can do with off chain. So maybe that's why they're yet to take off. But some things just take time. We look at Pendle, right?

Another protocol like ourselves that has been around for a while and it didn't really do an awful lot until one day it did. And it's one of the most important protocols in Defi today. But for at least a year, maybe longer, fix-rate protocols were everyone to say. there's just no appetite for it in crypto. And I think you see that other product clients as well, you know, options, not really come to crypto yet, not really come to defy in it in a major way. But, yeah, some things just take time and you reach a critical mass,

and then they just really take off and bloom after that. So, yeah, I wouldn't rule it out for FX, but it's not there today. That's a good segue talking about some of these functionalities that maybe haven't played out and some of those areas of the market that you can explore. What's your vision for Euler as a platform over the longer term? I think it should be a liquidity hub, a defy cGy per app effectively, where it holds significant liquidity for people and provides a lot of functionality for doing different types of trades

or different types of financial operations on top of that liquidity. So the underlying technology is very flexible. It allows you to build all sorts of different types of credit products on top, whether that's variable rate, fixed rate is something that we've got coming in the pipeline. And I think fixed rates are going to come back in a very big way in defy. in the fairly near future. I think all the major lending protocols are taking a look at this

and rethinking how it should be done. So it will be interesting to see how that changes in the near future. But yeah, certainly you can build a very good fixed rate post-called to Boililer. I think that will be coming. Oil swap is something we added fairly recently and it's still in beta,

but it's been looking extremely promising as a new technology on top of oiler. Can you talk about how oil swap works? I'm very excited about it. I mean, there's a backdrop. Just tell you about some of the pain points of existing decentralized exchanges. one of which is that when you put liquidity in a decks and it's not traded on, it's not

utilized within the given block. It's just tidal liquidity and it's just wildly inefficient to do that. Even on blocks where you have trades, most of the liquidity isn't utilized, just a small percentage of it is. There's just so inefficient to do that. The rest of that liquidity should be earning yield in some form or another. And that's how it works in Trabri, right? If you're trading on some off-chain order book, the liquidity isn't literally sat there in units being like moved around than the order book. It's actually sat in the background somewhere, in some account, earning yield for people, and then they settle up the order books at the end of the day or whatever, right? And on a swap solves that problem.

It effectively pulls the liquidity inside lending or credit markets by default, optional for LP, but by default, that's the issue in position. And then liquidity is then used just in time to supply liquidity to the market for swaps. And so when you LP on oil a swap, you were going yield by default from lending and borrowing first. And then if you want, you can also then switch on the option to earn swap fees and then extra sources of income from the swapping. So it's just much more capital efficient from that standpoint. But another major pain point over FedEx markets is borrowing against your liquidity.

People want to do this for all sorts of reasons, but it's just not really been a solved problem. In my entire time in D5, people have always wanted to see if they could borrow against their uniswap LP tokens. I remember people trying to do it in V2 and it didn't work there because it was very hard to price the tokens. and there are all sorts of exploits associated with the unit swap of E3, it's arguably even harder because now the tokens were NFTs and they were even harder to price or to be able to use as collateral

and a fungible type of collateral that lending protocol. So oil swap to decks that when the liquidity by default sits in the inside credit vaults, and if those credit vaults are exacted as collateral in inside other vaults, then that gives the native borrowing functionality. So when you LP into oil swap, no only do you earn an interest on the assets from lending, if you want, but you can also natively borrow against your LP positions. And this doesn't require any fancy logic in terms of moving the LP tokens around into new markets. I think it's just native

functionality that you get out of the box. So ultimately what happens is, let's say you went to LP, ETH and ETH and USDC, you put ETH into a ETH credit vault and then some interest there. You put USDC into another credit vault and interest there. And then you as the account holder would install like an AMM curve on top of your account that would define, it's kind of a bonding curve that defines the rules that swapers have to obey. And the swappers, what they're doing, when they're in they're effectively just rebalancing the amount of Ethan USDC in your credit accounts now. Someone sells ETH, you get more ETH, and you take some USDC out, and vice versa. And if you want to borrow now against that position, you can because ETH and ETH and ETHC are collaterals in Euler

and lots of our markets. And so you can just go away and borrow against that position. Now, what might you do by borrowing? You could maybe borrow and supply even more liquidity, so you could or leverage the LP token and multiply the yields that you're earning. You might just borrow against it as a hedge. You might want to hedge out the ETH side of the position, and I think you don't want to be as exposed to that side of things. You might want to borrow just to access funds and liquidity to go use elsewhere and EFI or some of the strategy.

But, yeah, lots of reasons to borrow against things, and it all comes out of the box, we'll all the swap for free. Yeah, it's a very, very powerful rebalancing primitive people's accounts on top of Euler. And yeah, I think is going to completely change the way that the Dex's work. We launched some beta testing vaults using the Dow's liquidity. And in under two weeks, I think we've done over a billion dollars in trading volume on the USDC, USDT pair. One of the reasons for that was because when people swap,

the swaps sometimes can be used as collateral to borrow the output amount. So an LP position of a million dollars can be used to simulate the same effective liquidity as a $50 million LP position in uniswap. And then you were leveraged swap fees and everything on top. But you just provide overall much deeper liquidity by using the borrowing functionality of the LPD tokens. So yeah, it's a very, very, very powerful, new primitive, and very optimistic baron. And I see one immediate use case given everything we talked about.

It'll probably be a great stable swap, I imagine, right, because of all the stable coin liquidity that you have. Yeah, hugely beneficial for new projects that are starting stable coins as well. I mean, the current system for most projects starting a stable coin today, is the very first thing they need to do is somehow we'll get Dex liquidity. How do they get that Dex liquidity? Well, they often have to pay for it. People often ask me, why is L-Ping on oil swap immune to permanent loss

and all these to toxic flow and all this of it? And the answer is no, that's not where the animation is. People aren't L-Ping all the time because they are profitable L-Ping just on the swap fees. People are L-Ping stable coins because they need Dex liquidity, because without the Dex liquidity, it's hard for them then to go on and build credit markets for their stable coins. And if they don't have a credit market, then the stable coin isn't that useful, really.

So the ex liquidity comes first and foremost. And what's unique about oil swap, I think, is you can build a credit market in parallel to a Dex market. And you can actually do the whole thing in one go in a much more capital-efficient way than people have done this in the past. You don't have to spend weeks building liquidity on a Dex and pay millions of dollars to incentivize that liquidity that they're from private LPs or from public LPs.

you can effectively deposit the assets, collateralize them, go and borrow from a USC market, for instance, the deposit so that you get a leveraged LP token, you now grow the dex liquidity. Once you have let dex liquidity, you can now connect that to other vaults and other things in the system,

where you can start to grow your credit market. It can all be done in the matter of days or a few weeks, rather than taking months and months and costing millions of dollars to bootstrap things. It's just much, much cheaper and faster. Yeah, I think it's brilliant, and inevitably we'll see a lot more. stable coins having to bootstrap. And so it's to your point really attractive for them.

I want to touch on a couple more questions, maybe pivoting a little bit. Do you still have a desire to build out these primitives? Is this what you want to do sort of for your infinite game? Or do you ever want to go back to academia or how do you assess where you'd like to spend your time? I really enjoy the engineering side of the industry we work in. Yeah, I really, really enjoy it. I mean, most of all, this, what was my engineering ideas and things. So, you just, you really enjoy it. using stuff from my background, doing a lot of maths and things. So I really enjoy that side of things. Yeah, running in companies is definitely also challenging and quite stressful, I would say.

It's certainly a very different experience from being an academic where you just work very much, where you work as part of the team, but you're kind of in charge of your own things, and you don't necessarily have loads of people to answer to and people's livelihoods and things dependent on you and so on. So, yeah, it's quite different. I wouldn't go back to academia personally. A division about what academia being an academic will be like that didn't really play out when I was there. I enjoy some aspects. Doing the core scientific research I really enjoyed,

much like I enjoyed in the engineering stuff at OILA, but the day-to-day stuff, the amount of bureaucracy involved in being an academic is enormous. It's more about writing research grants and writing papers and it is about doing core research, I would say, these days. I think that whole industry probably has some evolving to do in order to keep up with the way the world's working. I think probably AI is changing a lot of things, but we'll completely shake up academia as well. I wouldn't go back, but I still sort of romanticize some elements of it, I suppose. Just being able to sit down and just chew through you, like challenging, problem solving stuff, day in day out. I think it's really fun.

The point you make on having to lead a protocol, in some ways, you personally become the victim of your own success, right? Because the large you get, the more attention you have, the more commitments you have, the more inbound you get. How do you balance the rest of your life with that reality and your family and alongside your commitments to the protocol? It's a good one, yeah. I mean, I'm not going to lie, it's quite challenging, but families are a natural balancer, I think. Before you have a family, you always think, how would I possibly have time for a family? But once they arrive, the time kind of sorts itself out for the most part. Like, you realize all the areas of life where you can sort of trim stuff away, stuff that you just use to take granted. They, like really prioritise a lot of things for you, I'd say, naturally. And they're a really great leveler. When you have a job like I do that's quite stressful and quite full on and 24-7, going hanging out with family on a weekend, just doing stuff with the kids. In those moments, you don't have time to worry about the nitty-gritty of the crypto markets to what's happening, whether Bitcoin's up or done, people are sad or not. For me, I think it's been really helpful.

A big distraction from work, because if I didn't have that, I just would work non-stop. I find it very kind of addictive, I suppose, like a bit of a little bit of a workaholic. I've often found that I don't need that much sleep compared to the average person. I heard recently there's a genetic variant where maybe, Some people need a couple of hours less. Yeah, I don't know whether I have that very, but yeah, I'd be quite happy just spending every hour, every day, doing work of some form or another.

I think that's how I got into crypto in the first place. I just used to do, even after I'd finish work with my day job as a scientist, so I would get home, maybe watch some TV with my wife, and she'd go off to bed, and I'd be like, well, I've got a good few hours now to do something by myself, and I would just do hackathon competitions. I'd like... Plus crypto markets at 24-7, so it's great for anyone. That's it.

The new programming language, yeah, try and build a market-making bar. I just used to do all sorts of, like, hacky things. And that's sort of how we got me here, honestly. Yeah, I'll be doing that for the rest of my life. I'm quite sure. Michael, I greatly appreciate you coming back on to talk to us. And like I said, before, I have a deep respect for everything you guys have done with the protocol and everything ahead. Oh, thank you.

Yeah, I appreciate it. I can't believe it's been so long since the last talks. It sort of feels like yesterday's. Yeah. So, yeah, it really shows how. faster years gone from that perspective. Just so much was happened. But yeah, it's great to be back, especially with lots of positive things to say, lots of growth to talk about and showcase this time round. Yeah, it's only the start. Yeah, definitely. Yeah, there's so much more to cover. I think both from

the industry, it looks very promising right now, but also for ourselves. We've got lots to come. So, yeah. Well, thank you again and have a great rest of the day. You see. Thanks all. Thanks for listening to another episode of On the Brink with Castle Island. To find out more about Castle Island, visit castle island.V.C. To listen to all of our podcast episodes, please go to On the Brink-podcast.com or just click on the tab in our website. Thanks for listening.