From Physics PhD to Curve: Michael Egorov on Real Yield for Bitcoin

0:00Kinsa Durst Welcome to Panoramic Content Suite Singapore Edition alongside Token2049. We have the godfather of StableSwap, Michael, the founder of Curve as well as co-founder of Another project he'll talk about very soon. But yeah, excited to have you on. Thank you for Joining us. Yeah, thank you. Yeah, we'll get into YieldBasis and Curve wanted to kind of hear Your personal story in terms of how you ended up where you are today. So maybe you can kick us off With a self-intro and a brief history. I know it's a long DeFi cycle, but a brief history would be

0:42Michael Egorov Great for the audience. Right. Well, I mean, if you start really long ago, I would say, I don't know how long you want the history to be. Maybe like a condensed, condensed one. Yeah, Okay. So yeah, basically, I studied as a physicist and a PhD in physics. And then At some point, I went to, I don't know, went to the US to do some sort of startup, which I didn't Know. I didn't know which at the time, but I thought I would need to sort of get out of my Comfort zone essentially first. And over there, I well, I mean, at the time, I was discovering Crypto for myself. And then, well, probably, probably being around the crypto circles of The time and founded a co-founded NuCypher. It was before Curve. And well, it was great. And later, Well, one of the things that NuCypher was doing was essentially a stake token, right? So you had To stake into to run nodes, essentially something like that. And I realized that what will need to Be to happen is liquid staking. So I was thinking, oh, how do you do liquid staking before it was Called that, of course. And I figured out essentially, an algorithm for swapping stable Like pairs. And I thought, okay, well, at the same time, I was early user of early DeFi With MakerDAO. And, well, I didn't want to sell my ETH, so I borrowed DAI and sold it for USDC while having the ETH collateral so that I don't sell the precious ETH. And I had this Problem of swapping DAI because like, you know, your liquidity was pretty bad everywhere. So, And I thought, well, it looks like I actually with this same algorithm I was thinking of For liquid staking, I can do it for USD pairs. And it actually will work for me. So that's What I launched Curve with. And when it got launched, of course, well, that's what probably Some people remember. It's at some point, it came to liquid staking with staked Ether.

3:05Michael Egorov And, yeah, so it actually worked for this purpose as well. So that's how Curve got started. And, you Know, multiple things were created at Curve. Apart from stablecoin swaps, there were like volatile Swaps with automatically managed concentrated degrees he appeared and Curve stablecoin And lending. And we have all of that running today over there. And later, I thought, well, Actually, looks like we can do, I could do single asset exposure using based on CryptoSwap, If you like, essentially, get rid of this nonlinearity of the payoff function of CryptoSwap.

3:57Michael Egorov And I thought how to do it. And actually, when I was, the idea came to me when I was flying to Token2049, two years ago. So when I was boarding a plane, it occurred to me that actually, This is, that might work. While I was at the conference, I did some quick simulations, And it looked like it should work. And then, some time later, it was implemented as YieldBasis. And yeah, YieldBasis is running for, yeah, running on chain for about a year.

4:38Kinsa Durst Amazing, amazing. Congratulations. I think sometimes ideas can come out of nowhere. And just moving

4:45Michael Egorov Around sometimes can inspire you. Yeah, usually that happens to me when I fly air, I don't know When you're when you're like 10 kilometers above the ground, maybe ideas come from space.

5:00Kinsa Durst I mean, thank you Starlink. I don't know. Yeah. But for me as well, flights is like this zen Space where you're kind of disconnected from phone for a minute. Yeah, meditative almost in A way. Cool. Yeah, happy to dive into YieldBasis, but very impressive background. I didn't know You're a physicist, from dealing with molecules to ones and zeros, and financial infrastructure, Which is super cool. Then Curve really unlocked this concentrated, hyper efficient swaps almost With likewise assets expanding into other types of DeFi primitives, which is super helpful. I just mostly remember you guys for, not mostly, but I really appreciated the UI that I think you were the first ones to really launch with this like Windows

5:47Michael Egorov Text. Oh, it was essentially, I would say, MS-DOS style. Yeah, right, right. And yeah, Yeah, I can tell a story about that. Yeah, yeah, actually, that I'd be curious, Because you guys are the person to do it. And yeah, yeah, yeah. So, basically, I wanted to Create some UI before I found anyone really quickly, because I want to get it up and running Quickly after creating smart contracts. And I thought, well, how would I do that to do it fast? I found some template, which looked interesting to me, which is like a Turbo Vision template, Right? So Turbo Vision is like a thing, a TUI thing, which was used for like Turbo Pascal, Turbo C, and whatever, like the old stuff, right? But it was made like for web, so you could See this same look and feel in the web. So I thought, oh, okay, that's kind of cool.

6:43Michael Egorov What if I do that? So I coded it up in like pure HTML and JavaScript without like modern frameworks First. And yeah, I mean, deployed it, but it looks like some people really loved the UI, Some other people really hated the UI. And this polarized reception of it, it actually, Well, it actually kind of made it iconic, I would say.

7:11Kinsa Durst Yeah, I think it almost helped with the adoption of...

7:14Michael Egorov I think so. Yeah, because it was, yeah, essentially, this was doing marketing.

7:18Kinsa Durst Yeah, yeah. So you're not only just a protocol designer, but also UI, UX designer.

7:22Michael Egorov Well, I mean, I don't consider myself a UX designer. It was like, I don't know, A little bit, because I liked it. But actually, a logo was also kind of fun, Because, yeah, a logo was made in five minutes, actually. Oh, really? Yeah, it was like a standard Python library for plotting. Matplotlib had examples, And one of the examples was a Klein bottle, right? So I thought, okay, well, Maybe what we do in Curve, I was essentially doing any number of stablecoins, right? So it's Something multi-dimensional, and Klein bottle is like four-dimensional. So I thought, okay, Maybe I show multi-dimensionality here, and I ran this, I don't know, plotting library and rotated The Klein bottle in a way which looks kind of nice to me. And then, yeah, it was a logo.

8:22Kinsa Durst Incredible. Awesome, awesome. Yeah, I didn't realize you used not just the whole project, It seemed like it was really your baby.

8:31Michael Egorov Yeah, yeah, yeah. So I was creating this, I mean, I've got people after that, of course, To work on this, but yeah.

8:39Kinsa Durst Incredible. Yeah, those are really like the golden days of DeFi.

8:43Michael Egorov I would say so, yeah.

8:45Kinsa Durst Now the industry has evolved, and you've come up with new ideas as well. As you mentioned, YieldBasis. I don't think YieldBasis takes on similar UI, UX, it's a new project.

8:58Michael Egorov Yeah, it still has some sort of resemblance offers, you know, sort of old school UIs, But yeah, it's not quite that, but also you wanted to show, okay, on mobile and that old Style UI, it doesn't play well with mobile with all these window elements.

9:18Kinsa Durst Yeah. But with YieldBasis, what are you, I guess, what is the goal that you're trying to achieve? What sort of vision that you did you have with the project, maybe if you don't mind sharing?

9:29Michael Egorov Yeah, well, there are two things to that. One is, you know, they're kind of solving Impermanent loss, one is considered to be a holy grail, but that slide from technical perspective. From user perspective, it makes your crypto earn something, so it makes your crypto yield bearing. And what crypto people want yield on is, I think, basically BTC. Coincidentally, The best performance you can get, I mean, the best compromise between risk and reward, Which you can get there, is probably with Bitcoin. And also Bitcoin is the asset which Kind of lacks yield. For Ethereum, at least you have staking, so some sort of yield.

10:19Michael Egorov But for Bitcoin, you have no yield. And well, there are many Bitcoins where people Love to get some yield on those. And I thought, well, maybe doing that focusing on Bitcoin first Would be a really, really good one. Interesting. And I think, well, how many Bitcoins are there? Like not earning yield. I don't know, like a lot. And really, the maximum capacity of the Protocol is, well, it's probably a little bit less than the amount of Bitcoin not earning yield. So their vision for it would be, I guess, to first get to maximum capacity with Bitcoin and Ethereum.

11:13Michael Egorov At that time, then all the kind of background liquidity for these assets will be YieldBasis. When you say capacity, you mean? Capacity, I mean, look, if you make liquidity, Then this liquidity starts affecting their volatility. If liquidity is small, it doesn't Affect anything. And even if you have like 100 million liquidity, it doesn't affect volatility Yet. Maybe a billion liquidity also doesn't yet. But maybe 10 billion, 50 billion already would Suppress volatility and will start reducing yields. So that's where the real capacity is. And of course, we would love to reach this capacity, because at that point, it means that You are dominating the world liquidity for these assets and basically capturing the market.

12:05Michael Egorov But in principle, it can work for other assets which are not overly too volatile. Well, some people ask if it's possible to do with stocks. And well, maybe, but it's not that easy. So with stocks, you need to kind of to do different ideas, maybe, maybe, you know, Running multiple markets for different stocks at the same time, somehow interconnected, Because stocks, they have, they are sometimes more volatile. Honestly, Bitcoin is less volatile Than stocks. Yeah, yeah. Yeah, really, you think so. I mean, it depends on this output. Yes. I mean, volatility of Bitcoin is kind of better in some sense, because it's like, it's kind of Smooth, right? So Bitcoin doesn't, you know, that doesn't jump up or down too much instantly. And Stocks actually can, like, especially after weekends, overnight, they can just go up and down by a lot.

13:07Michael Egorov So they have discontinuities and Bitcoin price is kind of more continuous because price discovery Is always happening 24 seven. Maybe when stocks start having 24 seven price discovery, they will Be more like Bitcoin in that sense. It's on the way, I hope. Yeah, but honestly, I think Bitcoin is

13:28Kinsa Durst Priced a little bit more fair than stocks. Yeah, amazing. And yeah, YieldBasis. It seems like a Product that people can, I didn't realize that you guys were focusing on Bitcoin first. I thought

13:42Michael Egorov You guys would focus on like Ethereum ecosystem, given your own material. Well, it's only Ethereum, It's existing on Ethereum, right? But the asset which is early in yield is Bitcoin first, right? And That's, yeah, it's just because of the price properties of this asset. Interesting for the Price, right? Yeah, yeah. Ethereum is actually also not bad. Yeah. But Bitcoin price seems to Be a little better suited. So it's better to start with something which works better, Make it work better and expand to all others. I mean, also demand wise as well, as you said,

14:21Kinsa Durst There's so many idle Bitcoin out there. Exactly. A lot of Bitcoin L2s were launched A past couple of years to fix this problem, but I don't think any of them really have.

14:31Michael Egorov Yeah, probably not. In theory, it's good. In theory, it's good. But also like, They were not fully trustless, right? And even when they are fully trustless, I think they can be made fully trustless. Then the question would be, why should people use those Rather than Ethereum L2s? And even Ethereum L2s have, I would say, challenges, right? So when L2 is branded with someone, it probably makes sense, like Base or Robinhood chain or Whatever, right? If there was a good Bitcoin L2, and if one of these players decided to make their L2 as Bitcoin L2, of course, that would work, I think. Right. Yeah, be curious. Also, you know,

15:21Kinsa Durst There's a lot of different yield providers these days. Also, speaking of stocks, people are bringing Yield bearing stocks on chain as well, like Strategy's STRC. It's quite crazy. But it seems Like you guys have a pretty clear initial starting audience with me, Bitcoin holders, A lot of blue chip crypto holders. Curious how you envision the future for YieldBasis? Would Currently you guys have these pools and products where people can deposit and earn yield on it? Where how do you envision the product growing into giving your future coming a few years? What sort of market do you hope to capitalize on?

16:01Michael Egorov Right, right. So yeah, I think, as I said, after more simultaneously with expending fully two Blue chips, we would probably want to look at more traditional assets and see what happens. I mean, whether it's, if it's possible with stocks, it probably makes sense to try it. If it's possible with gold, it also makes sense to try it. Like who knows, maybe, Maybe people like yield bearing gold. Yeah, yeah. Yeah. But it's, it's, it's almost like,

16:36Kinsa Durst I think problem with yield is that people always ask where it's coming from.

16:41Michael Egorov Here it's clear where it's coming from. Yeah, exactly. It's coming from volatility from people Trading these assets. And you're essentially market making there. But you're sort of Delta neutral, but delta neutral denominated in this asset rather than in dollar or whatever.

16:57Kinsa Durst Yeah, so it's like a real yield that is generated.

16:59Michael Egorov It's a real yield. But also, of course, that that comes with limitations like when, you know, When people are very interested in this asset, volatility is high. So it's, I don't know, they Could be interested to enter or to exit out of this asset or better if they have that at the Same time. But, you know, it can happen to be the opposite, like everyone who wanted to sell Already sold, and everyone who wants to buy are still scared. And at that time, volatility is low. So the yield is very low. So you, you have this seasonal, seasonal component to it. But yeah, At the same time, there are a few other strategies, right? So and this strategy seems to be earning When some other strategies don't. Right. So it's pretty, pretty good, I think.

17:53Kinsa Durst Yeah, yeah. No, it's, we love real real yield. Yeah, I think too many governance token yields,

17:59Michael Egorov Unsustainable yields have been. Yeah, yeah. Well, you know, like, when, When I think one of the first governance tokens YFI was launched, it was called A valueless governance token by Andre Cronje. Right. So I guess some of the governance tokens Really implement this valueless vision today. But I don't know, I guess it is the narrative

18:30Kinsa Durst Game sometimes Andre Cronje made it made it work and drew liquidity from all across Different protocols, provided yield and active, just active governance participants as well. It was fun times. Fun time. Fun time. Absolutely. That's a, you know, given that you are so in Depth in the trenches on, I feel like you have a deep understanding of things on a protocol level. It's curious, you know, how you're seeing the current state of DeFi market ecosystem. Do you think there's anything that needs to be fixed that you're looking for? What is the next Holy grail? Do you think that if someone solves it, there'll be kind of like a new big player in

19:16Michael Egorov The industry? Yeah, that's a very good question. I usually try to hunt holy grails. Yeah. Right, Right. You don't want to share the... Yeah, yeah. What's the... I mean, I guess Hyperliquid has been Like the... Yeah, yeah. You can say that because I think one of the things people really wanted is High leverage. Yeah. But I wouldn't say that Hyperliquid was the first one to do that because there was GMX, for example, before Hyperliquid. So it's kind of an evolution in this sense. Yeah. Right. So, But maybe a very well executed, but... Hype is more of also like a centralized, Semi-centralized. Yeah, yeah. And really, I don't know if it's fair to call something semi-centralized DeFi, right? Exactly. I think GMX was more of a decentralized perpetual. Absolutely. Yeah, Very innovative. Right, right. But really, like, if you think why DeFi at all, like why don't do Everything centralized? And I think the answer would be that, well, you have really many Cracks these days, right? And even when the regulatory climate permits to do something Semi-centralized. Decentralized has an edge with security. You don't have OPSEC issues in many places. If it's not possible for someone to take user funds, Like, not possible at all, then you don't, like, you don't have these issues of protocol Or losing money because someone, some key was not custodied well or whatever. Right. Some

21:00Kinsa Durst Multisig got compromised or something like that. Yeah. Security is very scary. Maybe I'll Tap on a quick question in that vertical. I was curious if you have a take on the recent AI math craze that's going on. People are scared that private keys can potentially be Compromised, especially the one that have made transactions. Some people are, Justin Drake from,

21:22Michael Egorov Yeah. Yeah. So yeah, I actually noticed this, it was Matthew Green yesterday, Tweeted about this, like, you know, public key cryptography is doomed. Well, I mean, That's what he tweeted. And well, I think that he later clarified that he didn't mean exactly that. Maybe, maybe with AI or like we are, we would be able to find ways to, I don't know, to Essentially reduce effective security of the keys. Right. So maybe 128 bits ECC would be not safe, But 256 bit still safe, something like that. So that's, I think we are safe, honestly.

22:10Michael Egorov So practically speaking, well, there was a better question that quantum computers, of course. That's, that's another, which is, which is not too far from what I worked on in laboratory. Oh, yeah. Yeah. Although it was not quantum computers, what I worked on, but the way you Get to this, like, physical system where you do quantum computing these days is very similar. So I was working on cooling atoms down to like a very, very, very cold state. Like when they're quantum states, essentially, you, when you want to make a quantum computer, Their current approach is, I think, the most favorite approach today is to do Neutral atoms in optical lattice. And you cool them down in exactly the same way. And then you Essentially catch them in a standing wave of like made by lasers. So that's, I think, how you do it.

23:11Michael Egorov Well, at least that's what that's the most promising approach, I think. But at the same time, I'm not sure we will get crypto hacked by quantum computers anytime soon. Right. Suppose that these quantum resistant algorithms, but actually, this is what needs to be really Poked with, with AI, right, as much as possible, like these quantum resistant algorithms. What If there is something which we don't didn't find yet hiding in them? ECC is kind of very, Very well studied. But, you know, this lattice cryptography, for example, is it? Well, We probably really need to use all the AI power to find stuff in that. Right. But at the same Time, I don't think like ECC will be hacked too soon like quantum computers.

24:09Kinsa Durst Hopefully not. I think Google is expecting timeline of like 2030 for practical or commercial

24:16Michael Egorov Quantum computers. Yeah, I'm not sure about that, you know. Well, I mean, there is, Of course, there is number of qubits growing. Okay, fine. But you don't need just number of Qubits. You also need some, I don't know, some sort of precision of these qubits. Right. So some small level of noise, like the smaller, the better. And then you should have a quantum Error correction working. So there is some threshold below which you can apply that. And But when you apply it, then you need exponentially more like qubits. So I don't know. But anyway, We also might discover that, like, I mean, we don't really understand quantum mechanics that Well, I think. Right. So we don't really understand where this randomness is coming from. And Quantum computers probably would be some of the best devices, although quite expensive to Really find limits of quantum mechanics. And yeah, I mean, if, if someone Creates a quantum computer, they probably will, if it's like really good one, which can supposedly Hack Bitcoin, right? Well, hacks are just bitcoins, right? Then they probably will end up in one of Two states. One is yes, they have cryptography. So what everyone is scared of and everyone who Is investing is hoping for. But it's also very well possible that cryptography will not be hacked.

25:56Michael Egorov But instead, some groundbreaking discoveries in quantum mechanics will be made. So I guess Investors would be a little bit disappointed that their money would be used to advance physics to The next level. But it's a possible outcome. Amazing. Yeah, super exciting. I think in a

26:17Kinsa Durst Few years, I know right now YieldBasis, we're growing as a company, but we might need your help With quantum cryptography in a few years. Either yes or not. So we'll see. Yeah. Yeah. Super fun Chat. I think there's some deep rabbit holes we can dive into, but maybe that's for the next Podcast. Absolutely. Meanwhile, thank you for hopping on. For the audience, if they are Interested in YieldBasis or want to get involved, get in contact with the team,

26:49Michael Egorov What's the best way to do it? Yeah, I think the best way is to go to YieldBasis Twitter or X. There are links to YieldBasis telegram. People can just go there and ask whatever questions they

27:09Kinsa Durst Like. Cool. We'll add the links below. But thank you so much for your time. Thank you For hopping on. Great chatting. Yeah, thank you.