Perspective on DEXs

Mark:

Well, thanks for joining. I appreciate it.

‍

Fernando:

My pleasure, Mark. Thanks for inviting me.

‍

Mark:

Today, 20% and growing trading volume is through decentralized exchanges instead of centralized exchanges like Coinbase, Finance and the ill-fated FTX. But as traders get more into crypto and learn self custody, they start to use these decentralized exchanges, but it's hard to understand them. So understanding them is really key to understanding Defi and crypto broadly. And so we've done intro episodes about Dexes and we'll do more, but today we have two Dex founders on the line, and so we're going to allow ourselves to get a little bit wonky and a little bit into the details and I think that's okay. So with that said, Fernando, thank you so much for joining us.

So Fernando, I want to start by make sure the audience understands what makes you such a good expert and guide. Obviously you're on the leadership team and a founder of Balancer. Balancer has about a $600 million market cap, about over a billion in TVL, and volumes are pretty substantial. It's about a quarter billion, which makes it one of the top 10 Dexes, and it's just very well regarded throughout the kind of Dex Defi ecosystem. But I'd love for you to add to that and let the audience know what else makes you an expert, what other color they need to know in order to understand and contextualize the comments you'll make throughout this episode.

‍

Fernando:

Sure, yeah. I wouldn't say I'm an expert. I'm a guide. Yeah. Thanks for all the kind words. But yeah, I am just a guy who was interested in AMMs and had an interesting idea in 2018 about creating multiple pools with multiple tokens, not only to, yeah, so that was the beginning of Balancer and yeah, I kind of worked with amazing people to get where we are now. I wouldn't take that much credit, just for starters.

I think I have a pretty good background in math, so it allows me to visualize and to kind of model things, an advanced program of mathematics and university, advanced calculus, so maybe those things, I'd say, help me be someone who understands a little bit more about the variance and auto mats behind AMMs. Yeah, I think one thing, talking about Balancer already, is one thing that helps us, us be kind of very broad is the fact that Balancer is a platform, is infrastructure for other AMMs to build upon.

We have amazing projects like Gyroscope, Gyroscope Finance, they're building a Stablecoin [inaudible 00:03:05] Balancer. There is a very interesting project that just launched as well. It's called Crown Finance. They're building TWAMs, who were described by the Paradigm team a long time ago, and they are live now on Balancer. So there's other protocols that use, like zero slippage in variance, [inaudible 00:03:26] finance. Yeah.

There's so many cool ideas that come to us and who want to leverage Balancer or speed up their launch time by leveraging Balancer, which is I think the only platform today to help founders who have amazing ideas, nice ideas. They don't want to build everything from scratch. And most importantly, they don't want to build the connection to aggregators because if you're not Uniswap, you don't have people using your front end or your smart conscious directly. They go through CoW Swap, through [inaudible 00:03:59] aggregators. So if you're not seen by aggregators, Mark, you know that very well, you're not going to have any meaningful volume. So we have through the interface, the common interface of the Balancer vault. We have that already kind of given to projects who build on top of us. So maybe I'm derailing a little bit, but I think the ...

‍

Mark:

No, that's great.

‍

Fernando:

The fact that we're generic, a generic AMM, helps us really go deep into every different crazy ideas that come through grants or people we're just building without us even knowing.

‍

Mark:

Makes sense. And you're totally right. I mean, the super majority of trading volume that goes through Clipper is from one inch. You know? And for those listening, they're decentralized exchanges and then they're aggregators. You can go directly to a decentralized exchange or you can go to an aggregator like One Inch or Matcha by 0X, put in your trade, and they'll show you 100 Dexes and which one will give you the best price. And they'll even route your trade for you, kind of like going to kayak.com for flights.

You know? There's this constant battle strategically that we think about, which is do we want to be the intel inside and provide best execution and rely on users to go to Dex aggregators, or does it make more sense to have a consumer and user brand like Uniswap where people come straight to the Dex? And that's something we struggle with a lot. It sounds like you have a third option, which is a platform for other Dexes, which will themselves be brands. I'm wondering which you think of as the most important future for Balancer, for Dexes in general, or is it just kind of a mix of all three?

‍

Fernando:

Yeah, that's a great question. I think already as a segue into the topic of L2 [inaudible 00:06:02], we just talked about, even for layer one, but thinking about ... I have a thesis that things will move to layer two. AMMs, they are very transaction intensive. Anytime prices move elsewhere on centralized exchanges, there has to be a transaction on every single pool or every single AMM to bring it back to the actual external price. And that gives an opportunity for [inaudible 00:06:26] to make some profit. And if it goes back and forth, then it generates, yeah, profit for LPs. But the point being that AMMs are very transactions intensive. I do believe that there's a future where most TBL four AMMs will be on layer twos.

That said, it's to me an obvious kind of color or obvious consequence that users will go through aggregators, even if Uniswap is still the biggest source of liquidity, the biggest, the lowest price impact, and it's not for all pairs. It's like Uniswap v.3 is very good at us, USDCE, but not on tail assets because you have to be always re-managing your positions which Uni v.2 two or Balancer don't require.

It's, to me, obvious that users will not go to Uniswap first because they know that if they go elsewhere, they will always see Uniswap and other things. It's like you want to buy a flight ticket. Great analogy, Mark. You're not going to look up Delta Airlines, American Airline. All the airlines. You're going to go to Kayak or other meta aggregators.

I think this is inevitable and especially because in L2s, what you pay for is call data. You don't pay. Complex execution is free. So if you have to go through One Inch or Match or CoW Swap and then you do lots of different trades on different exchanges to make sure you get all the best of the liquidity that's out there, you're going to pay the same as going directly through Uniswap because your call data ... I want to buy this for that, and that's the minimum I want to get back is the same. So on L2s, it's an obvious advantage that there's no, pretty much no downsides to going to aggregators. L1 is still a little bit different because if you know your best rated-

‍

Mark:

Can you just explain why call data or what call data is?

‍

Fernando:

Yeah.

‍

Mark:

... to the audience.

‍

Fernando:

Sure. So if you want to transact on Ethereum, you have a you call function and then you pass parameters. For AMMs, usually those parameters are, I want to trade token A for token B. The minimum I want to get back is X because otherwise you're going to be sandwiched and ripped off. And there might be other options. For example, Balancer, this is kind of bit advanced, but you can choose to keep your balance in the vault. So you say internal balance. True. So if you're trading a lot with Balancer, we have this idea of a single vault architecture. You don't need to be sending tokens back and forth to your contract or to your wallet. You just say, "I have some credit within the window, the vault. Since I'm trading back and forth the whole time, that saves me at the end of the day, a lot of the money, a lot of money. If I'm an [inaudible 00:09:00]."

‍

Mark:

It's kind of like in a bank. If you put your money in a bank ...

‍

Fernando:

Yeah.

‍

Mark:

Just to make sure I understand the vault, if you do transfers to another customer of the bank, it's instant and free.

‍

Fernando:

Exactly.

‍

Mark:

You want to do a transfer to someone outside of the bank, well, it's a $40 wire fee and it's going to settle in a day or so.

‍

Fernando:

Exactly.

‍

Mark:

And so you can keep your money in this vault. Still non-custodial, but enables you to do a lot of trades with less gas.

‍

Fernando:

[inaudible 00:09:23].

‍

Mark:

Less. Less.

‍

Fernando:

Exactly.

‍

Mark:

Less [inaudible 00:09:26].

‍

Fernando:

It's not custodial. That's key. But yeah, like I said, you're just keeping it there. It's a tab you have open. Sending money around costs a lot on layer one because it's token constraint transfer. You have to pay that for that gas. So it saves a lot of gas for people who trading a lot on Balancer. And there's very, yeah, there's a lot of people. Arbors and protocols that are doing that as of now. That was just an example of another piece of call data.

You have to say, "I want to keep this there," and roll ups or layer twos, they're mostly charging people on the call data because this is what they have to submit to the layer one. All the processing is done, ZK, zero knowledge proof. So you can prove that it's correct, but you have to know what people actually called and what the parameters were in order for the whole thing to work. Yeah, so that call data is what costs actually on money. I'm kind of the oversimplifying a little bit, but you can have the most complex operation. It won't add to your cost as long as your call data is kind of constant.

‍

Mark:

Would it be fair to say from the call data is what is anchored onto Ethereum, and so the expense of the transaction is gas. On the layer two, that is trivial, but it kind of requires an anchor down to level one and that happened to the call data. And so whatever happens in the call data is one of the biggest ...

‍

Fernando:

Costs.

‍

Mark:

... factors in the transaction costs.

‍

Fernando:

Perfect. Yeah. And then with all that call data, you can create a proof, a ZK proof, that using those inputs, very complex calculations have been done. It doesn't matter how complex they are. The output is correct. So anyone can challenge that for optimistic roll up. So that is. Yeah, that's very interesting. But the main point here is that I do think that there's going to be more and more people interacting directly with aggregators, not directly with AMMs, and at least us in the Balancer ecosystem. I speak for Balancer labs, but we have a whole ecosystem with lots of service providers and companies. We think that things will move over to layer twos.

There's going to be a multitude of liquidity sources like Clipper, like Balancer, like all the AMMs that we see around. And there's going to be specialized players that make sure that they see the bass trades given all those liquidity sources. That's my theory.

‍

Mark:

Makes sense. Okay, great. I think there's a lot to be said for that. I mean, transaction volume by number of transactions is still, I mean, huge on L2s. Dollar volume is still larger and L1s. Maybe that will persist, but maybe one day L2s will flip L1s.

‍

Fernando:

I believe in that. I think L1 has been around since 20, whatever. Ethereum was launched, I think 2015 or '16. There's a lot of leaning effect. People trust Ethereum. It's like WEF. You know? You trust WEF, but you don't trust new smart contracts that have been created. So I think L2s haven't been around for a long time. There hasn't been a case where there's been a massive attempt to steal funds and people got slashed in this thing proved itself. So far, yeah. That hasn't happened to my knowledge. So I think people are not, especially if you're aware, if you have a lot of money, transaction fees don't matter much. Right? Why would you be willing to take the risk of losing a lot of money if you can't stay on Ethereum, which is battle proof.

‍

Mark:

Well, that's the thing, right?

‍

Fernando:

Battle, tested, right? Yeah.

‍

Mark:

And so if that's the case, then there's always going to be more liquidity on L1 ...

‍

Fernando:

That can change.

‍

Mark:

... than L2. And so if you're doing a very large trade, then you'll probably still want to do it where the liquidity is nest instead of where the gas is cheapest, which means you'll probably want to do it on L1. And so what we may see is a continuation of the trend where the dollar volume is still on Ethereum, but it's really from a very small number of traits.

‍

Fernando:

Yeah. That's a good point. I think even that can change. If people realize that by LP on AMM, on an L2 where the chain, the network, doesn't suck out so much value, because value goes to LPs, to traders, and to arbors. There's no other places. It's a zero-sum game pretty much. So if LPs realize that on L2s, because the fees are so low, they're making more money, because they can harvest a bigger chunk of all the fees that are being generated, maybe that will attract a little bit more money from the L1 to L2.

‍

Mark:

I see. Yeah.

‍

Fernando:

Slowly. And people realize that, well now, it's not even dollar amount. It's like TVL is actually higher on L2, so people are going to trade bigger amounts on L2s and then that might flip. I don't know. I think for sure, smaller pools, smaller projects, like long tail tokens, I think that that is to me very, very 95% certain that it will be like on L2s. But maybe you're right.

‍

Mark:

[inaudible 00:14:45].

‍

Fernando:

USDCE, maybe that stays on [inaudible 00:14:47] and we'll see.

‍

Mark:

So that. Okay, let's talk about that, because I'd love to get your feedback on this. 70% of trading volume is ETH and USD Stablecoin pairs and that's 70% of Uniswap's volume. Right? It really goes through just a few pools. My guess is I think of there being three phases of this thesis playing out. One is that as long as Ethereum and Bitcoin are growth assets, then a lot of people are just going to want to hold large amounts of ETH and the cost of capital for doing anything else is quite high, because Ethereum, Bitcoin are ... people are bullish. Once those become stable assets and crypto penetration is reached steady state, which is not going to happen for a while, then ETH and BTC are no longer ... you know. I don't know, maybe this is a million dollar Bitcoin. Whatever it is. You know? Then the world changes and they're no longer growth assets.

And at the point where they're no longer growth assets, then people don't just want to hold them for the sake of holding them as much. Right? And so they're chasing. They would chase returns elsewhere and maybe them's a better store value, but I don't know. We're just spit balling here. And so, long as ETH is something people just want to hold for its own sake, especially a lot of institutions, seems hard to see that liquidity migrating, in big ways migrating to L2s and flipping L1. Whereas I could totally see that happening once the growth expectations for Ethereum of itself cool off. But that's so long term that maybe it's too speculative.

The shorter term thing I'd love to get your feedback on is rising rates. Right? So we've had the Cappella upgrade. Now there's essentially a risk-free rate for Ethereum, call it 5%. I don't know, I have to check, but we can check in any moment, for staking Ethereum. There's also treasuries are coming on chain. So yields for USD are 5%-ish. 70% of volume goes through ETH and USD pairs. I guess I'm struggling to understand. If that's the bulk of the Dex market, are people going to want to stake that into liquidity pools for Dexes when they can get a pretty big risk-free rate else-

‍

Fernando:

I have [inaudible 00:17:14].

‍

Mark:

[inaudible 00:17:14] on the chain? And what happens to Dexes and Defis over the next several months? Will there be massive pool flight? I guess I'm struggling with that a lot.

‍

Fernando:

That's a great idea. That's a great point. I think we have an answer to that and I fully agree with you. It's clear that holding ETH and USC as TVL on a pool, like as an LP, makes very little sense. You have to have a lot of good flow, which is users actually trading, which generates fees. And for you to get even close to those 5% you can get as a risk-free kind of interest rate just by staking or using all the other USD options.

The key here, mark, is that you can do both. Balancer allows you to have both the best of both worlds. How? Use LSD like you use liquidy staking derivative tokens for ETH. So you have staked ETH with Lido. You have RETH with Rocket Pool and there's other providers like stake.us or steak StakeWise. So instead of pure ETH, you can have the derivative token that kind of maps into that interest rate for stakers who are-

‍

Mark:

So basically you're saying that the pool instead of ETH USD Stablecoin will be yield generating ETH.

‍

Fernando:

And yield generating USDC.

‍

Mark:

Yield generating USD. But doesn't that imply that the 70% of volume between E and U S D today is going to want to switch to trading these two yield bearing assets?

‍

Fernando:

It does not.

‍

Mark:

I mean that's a big shift if most people want to ...

‍

Fernando:

It does not. No. It does not. I agree. People want to trade ETH and USDC. Not their derivatives. This is a very interesting innovation that we created at Balancer quite long ago. And when people hear about it and understand it, they're amazed at it. And I think it's a great idea. But not a lot of people know about it. It's boosted pools.

What boosted pools are is you actually have two tokens that are actually pools themselves. So instead of having USDC and ETH in a pool, you have this called linear pool boosted pool of USDC and this linear pool of ETH. That pool holds a little bit of buffer. Let's say 10% of wrapped ETH. The other 90% is wrapped staked ETH. Right? This other side of USDC, it's 90% Aave USDC, AUSDC, and 10% actual USDC.

What happens is, when people are trading, they're actually trading directly USDC for ETH and that's what they see. But most of the liquidity is in the form of you generating token. And the cool thing is that it's kind of a very interesting idea that everything on Balancer can use the same pool, which is a linear pool, that has the AUSDC and USDC. And the key here is that if a lot of people are buying USDC with ETH, this buffer will start to go down until it goes to zero.

We have a way to add a very small fee if the buffer's going in the wrong direction, like it's kind of depleting, or it has too much because then you're giving up on yield. Right? If you have everything in USDC or not AUSDC, you don't have any yield.

You want to be in this kind of sweet spot. Let's say 5% to 10% of USDC. If you go beyond the range, then you start to accumulate a little bit of fees so that any arbor or we have lots of ARBs looking at those pools.

‍

Mark:

The arbitrage.

‍

Fernando:

They arbitrage.

‍

Mark:

Yeah.

‍

Fernando:

They bring it back to the middle of the sweet spot and make $100, so actually, you have AUSDC and let's say AETH or whatever staked ETH. You're not trading every time AETH for ETH and then for USDC and then for AUSDC. You're not doing all those steps.

‍

Mark:

But in that sense, I mean, basically, it's kind of like analogous to a bank. I'll just draw an analogy to a bank again, which is, you have some cash in your reserves, then you have a duration mismatch with the rest of your assets. That would take a little bit of time to liquidate, except that you as people want to withdraw one for the other, you make it more expensive to do that. So you deter it and you attract arbitrageurs to come the other direction. It almost seemed like two Dexes in one.

‍

Fernando:

Yeah, I think that that's a good way to put it. Yeah. So when you invest in a boosted pool, you have all the risk kind of disclosures. One of them is like you're exposed to Aave. Aave risk. And if everyone wants to withdraw their liquidity in Die, Aave might not have enough Die for all the Die that's being converted, because most of the Balancer liquidity is kind of converted in Aave to generate yield. So everyone withdrawing the same time, then it might be like ...

‍

Mark:

Yeah.

‍

Fernando:

But then Aave is built for that. If a lot of people want to withdraw, then the borrow rates rise and then more people will ... it's a very temporary thing. Unless there's a contract like bug, it's a system that works pretty well. Yeah.

‍

Mark:

Okay. Great. Clever solution. I'm curious what you think happens to Uniswap, right? I mean ... how do you think they handle this? Or does 70% of their volume kind of go away because there's a drain from their pools?

‍

Fernando:

Yeah. I think-

‍

Mark:

Because that is the incumbent in our space ...

‍

Fernando:

Righty.

‍

Mark:

... that we benchmark ourselves against, right?

‍

Fernando:

Yeah. Two things. I absolutely love Uniswap. I think they deserve all the success they have. Yeah. Having directed with Hayden in the early days. And Dan Robinson, their team is amazing.

‍

Mark:

Good people,

‍

Fernando:

Very good people.

‍

Mark:

And they can innovate a lot.

‍

Fernando:

Yeah. They helped the space a lot.

‍

Mark:

Yeah.

‍

Fernando:

I'm a big fan. Nothing-

‍

Mark:

So with that caveat, what do you think the market dynamics are around ...

‍

Fernando:

I think things change-

‍

Mark:

... this rising rate environment?

‍

Fernando:

Yeah, I think things change quite quickly. So of course there is a lot of users who are there, they go directly to the Uniswap front end we just talked about. I think that will change over time and they will go more to Matcha in One Inch. Yeah. It just makes sense, especially in this L2 world. You generating thesis that we just discussed is very strong. Why would you be LPing on a pool that has the cost, the opportunity cost of 5%, right? You have a bunch of ETH. Why don't you just hold RETH? And those staking derivatives will get more [inaudible 00:24:05]. Now with Capella, it's been a major de-risking event, so I feel better not having ETH than having staked ETH or our RETH. Yeah. I think this will change slowly and I think Uniswap is in a great position to kind of adapt to the change. I think they're working on a V4.

‍

Mark:

Yeah. They have the users.

‍

Fernando:

They have the users. That's the most important thing. I'm pretty sure they're working on a V4 right now and they're always very innovative. But I think that things can change quickly. I've read some analyses that are pretty solid that LPs are losing money except for some very specific ones.

‍

Mark:

Likewise.

‍

Fernando:

Right? A lot of LPs are losing money.

‍

Mark:

[inaudible 00:24:47] has put out some great analysis on that.

‍

Fernando:

Yeah. And I saw. They haven't replied to that. They said first that this is not true, but actually, their analysis was missing a minus one flag or whatever and actually was.

‍

Mark:

Then they just haven't followed up.

‍

Fernando:

Yeah. They just went silent. Yeah. MIA.

‍

Mark:

Yeah.

‍

Fernando:

So I think it's clear that it's not for everyone. And yeah, they have mostly-

‍

Mark:

All right, So big changes coming to Defi.

‍

Fernando:

Absolutely. It's not going to stop changing in the next 10, 20 years. I think it will take a long time, until, as you said, Mark. Bitcoin's at 1 billion and things are kind of more stable like gold is. No investment advice here of course, but I'm on your side here.

‍

Mark:

Of course.

‍

Fernando:

Yeah.

‍

Mark:

Okay, great. So regulation is one of the hottest topics of the hour, right?

‍

Fernando:

Right.

‍

Mark:

There's just been testimony on Stablecoin and from the SEC in Congress and there's been a wave of enforcement actions. Meanwhile, we have Mica in Europe with a new regulatory framework. How do you think about these regulations and the regulatory environment as applied to Dexes in particular? How worried are you and how optimistic are you?

‍

Fernando:

Aw. Yeah. That's a great question. One of the important reasons why we want to be infrastructure is really to let every project deal with their own risk assessment. So if you think you have a front end that uses Balancer on the backend and you are in a jurisdiction that's very friendly, then you can have a different way of working, operating, than if you're working in the US.

We want to be kind of less and less exposing Balancer to end users and more to developers. I think that's kind of clearly the direction that we're going. And this is good regulatory wise. Also, we have made sure to decentralize. So we have now various service providers, various entities, that are working within the ecosystem of Balancer. I think this is also clearly kind of going away from this labs entity that does everything, so we just deal with smart contract. There's another entity that's doing the UI, another entity that's doing developer relationship and everything.

Regulators will realize that being on custodial is such an important thing that differentiates us from the next ftx. It will keep existing, Binance and all those centralized entities. Users are safe. Their funds are safe. We can't do anything with their funds unless there's a hack or backdoor, which is not our case, hasn't been so far regarding a hack.

It's a fundamentally different idea that they will have to adapt. And Mica, to my knowledge, is not clear about taxes and sexes. There's still a lot of gray areas about these assets. But regulation is going to evolve. It will have to evolve. It can't just say every Dex is ... trading securities, they need to geofence US, or KYC, all the users. This is something. It would really break crypto's reason of being, like if you had to KYC everyone.

I really love Erik Voorhees and he says he is that analogy between email and crypto. If you were to KYC everyone who's using emails because they're sending information, and information in a way is what we're sending through blockchain, it has monetary value. The world would've been a much worse place, would've become a much worse place if we also had outlawed encryption in the early days. I think regulation will have to evolve. I don't think we can do anything. We are in a regulated world. We want to do things that are correct and we want to help regulators find ways to regulate the industry in a way that makes sense because we don't want people to lose money like they did on FTX and all the hacks. We want them to use transparent blockchains that are auditable and completely open. So that's my thought. Yeah.

‍

Mark:

Interesting. Okay.

‍

Fernando:

Yeah.

‍

Mark:

Makes a lot of sense because the regulations really started around this idea of there's people custie-ing other people's money, so they must be regulated to protect those people. Right? And it's very hard to take that and apply it to a situation where actually no one's money is being custied, nor does it make sense, nor is it possible because you can't really stop the code. When regulation meets reality, reality does tend to win.

But it's really interesting that the path you've outlined is there's a separation between the smart contracts. Right? Where, while we don't know regulation where regulation is going, we know it will have to adapt around reality and the front end that's actually engaging with users and is probably going to end up being localized. And so that's a very interesting approach where you're becoming more and more infrastructure for other developers and their front ends as opposed to necessarily doing that yourselves. Very clever and is a good forward pass. Can you just conceptually explain how it is that a Dex can be a platform for other Dexes?

‍

Fernando:

Yeah. Balancer has this vault architecture where all the pools store their funds in a non non-custodial way. Balancer has hooks that allow pools to connect to the vault, and the pools themselves, they have the logic for the variant or how the traits happen. So it's completely future-proof. You can have stable swap pools like curves, logic. We have a lot of those. You can have weighted pools. You can have two token pools. You can have elliptic curve pools, which Gyroscope developed. So we really have this interface that's open, very simple, and then anyone can bring their own logic, their own protocols, to be built on top of Balancer. I'm quite bullish about this and this is what we're doubling down on.

‍

Mark:

I see. You outline a set of parameters, whether it's the type of token, the number of tokens, the weights, or the pricing formula. Within those constraints, people can experiment, and if they stay within those constraints, they can tap into liquidity you already have.

‍

Fernando:

That's correct. Yeah.

‍

Mark:

I see. And the liquidity providers basically understand that and accept that they might be exposed to a bunch, an unknown set of risks, within the constraints of the parameters that can be ...

‍

Fernando:

Right. Yeah.

‍

Mark:

... tweaked.

‍

Fernando:

They're not exposed to other pools, so the vault knows exactly what pool has what assets, so they don't commingle. But if you invest or deposit tokens to a pool, you know are exposed to that pool.

‍

Mark:

I know we're coming up on time. Is there anything you would like to share with the audience about yourself? How to follow you, how to learn more?

‍

Fernando:

Yeah, if you want to know more about Balancer, just follow @Balancer on Twitter. You can go to Discord about Balancer.fi to join our community. There's a lot of people there. If you have questions about our technology, [inaudible 00:31:56] Dab. So we're mostly geared to Dabs. We want more like developers to use Balancer to build on top of Balancer. So you'll be very welcome on our Discord.

‍

Mark:

Awesome. Well, thank you for your thoughts today. We appreciate it.

‍

Fernando:

Thank you so much.