Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
defi/acc: AMMs vs CLOBs (w/ LFJ, doppler, kuru, perpl & clober)
Recorded: May 7, 2025 Duration: 1:00:28Space Recording
Short Summary
In a dynamic discussion on DeFi Act, industry leaders from Trader Joe, Doppler, Kuru, Clobber, and Purple explored the evolving landscape of decentralized exchanges, focusing on the competition between Automated Market Makers (AMMs) and Central Limit Order Books (CLOBs). Key insights included the launch of innovative protocols, the importance of liquidity provisioning, and strategies for attracting retail users to on-chain trading.
Full Transcription
Thank you. GM GM.
How's it going?
Austin, how's it going, man? Good to see you.
Hey, Austin, can you hear me?
I think he can hear you, but he's not a speaker, so he might not be able to reply.
I just invited him as a speaker.
Oh, you're right.
I can't see him as a speaker.
New York.
There you go. A wonderfully warm and nice New York day.
Nice, nice.
Very cool. I think we're waiting on two more people to be on and then we should be good to go okay awesome uh looks like everybody's here um so we can get started. Welcome guys to the third episode of DeFi Act. And today's topic is going to be AMMs versus Clobs. We have some very special guests with us from the Trader Joe team, the Doppler team, Kuru team, Clobber team, and Purple team. And we have folks representing both sides of DEXs,
the AMMs, as well as the Clobs.
I'll just pass it over to each one of you guys in order,
and then you guys can just introduce yourselves.
And then maybe we'll start off with the opening arguments
and see what's everyone thinking.
So maybe I'll pass it over to Fish first from LFJ.
Hey, guys. I'm Fish. I'm one of the co-founders of lfj previously known as trader joe um we originally started on
avalanche and we will be launching on mona when it goes live awesome awesome can we have you next? And then, Frivashi and Clobber and PBJ,
you guys can just popcorn.
Yeah, so I'm Austin.
I'm the founder of Whetstone Research.
I also sort of run, I guess, an AMM protocol called Doppler.
And I guess I used to work at Uniswap doing AMM design.
So I guess I'll be supporting AMM side here too.
Travashi, you want to take it away? Yeah, right.
Hi, I'm Travashi, one of the co-founders of Kuru.
Kuru is a fully on-chain audiobook text
for spot assets built on Monad.
And yeah, happy to be here.
Thank you for having me on.
Clover.
Hey, guys.
This is Kevin from Clover.
Clover is a fully on-chain orderbook Dex protocol
with a very efficient on-chain matching engine.
Currently live on Monad testnet and nice to meet you all.
PBJ, you're up.
Hey guys, PB from Purple.
We're building a fully on-chain perpdex on Monad.
Excited to be here.
Awesome. I appreciate all of you guys coming on.
And yeah, let's have a constructive conversation today.
Just to set the setting here and then I'll pass it over to each one of you guys to kind of make your opening arguments um we initially had you know dexes on ethereum all
the way back in like 2017 18 and um the first one of dexes that we actually saw in ethereum
were not amms they were order book dexes but uh back in the day nobody was coding tight liquidity
on them and eventually what started off as kind of like a Reddit post by Vitalik ended up being
adopted by a bunch of DEXs.
And then most famously, Uniswap kind of popularized the AMM concept with Univ1 and Univ2.
And then eventually, we saw a lot of adaptations from the original balancer or
Uniswap models and being extended to ecosystems outside of the EVM as well.
So even if you look at some of the biggest taxes on Solana today, they're all automated
market makers based on like the original kind of uni V2 curve.
And you know, given that a lot has changed over the past seven or eight years and, you know, the reasons for which order book taxes did not succeed on Ethereum mainnet back in the day, a lot of those problems at this point have been solved by a lot of the new chains that have come to market.
when Solana came out, Solana had a lot of TPS, extremely cheap gas fees. And we had this order
book DAX called Serum, which did quite well back in the day. Eventually, I think Serum folded as
a project, but then now we have Phoenix and a bunch of other products on Solana, Sui, and a
bunch of other high throughput chains. And the reason why I think it's interesting to talk about
AMMs versus clubs is because
when you look at a high performance chain like Monad, the question really is what kind
of ducks can we see performing well in this kind of environment? And, you know, the obvious
answer if you look at, you know, centralized exchanges might be that, you know, order books
are the dominant form of trading in all of crypto and beyond crypto
as well. If you look at traditional financial markets, but it's only in the on chain environment
where we see being as sticky as they are. And I think there's quite a few reasons for
that, but I'll reserve those things for later. And maybe we'll have each of the speakers
make some opening arguments for their side.
So I'll pass it over to Phish first and encourage you guys to bring on your spiciest take.
And if you feel like the other side is not going to make it, let's definitely have that out and argue it out on why we think one side is going to be more dominant in the future versus the other.
I'll pass it over to Phish first.
Yeah, I think we're quite qualified to talk on this topic.
Just to give you guys a bit of background. So LFJ, aka Trader Joe, we invented the Liquidity Book,
which is also known as the DLMM.
And you can see that on different chains like Meteora on Solana and Metropolis on Sonic.
So DLMMs are kind of like a hybrid between Unisop v3 and an order book. It's kind of like
contrary to liquidity, except that you have these bins and they basically act like an order book in a sense. So what we're seeing actually is like people like,
I feel like the future will steer towards order books
just because it is much more flexible for a market maker
to do smart coding strategies.
And we're starting to see a lot of liquidity
in the space become centralized, right?
You have a lot of on-chain market makers now and RFQs, and I feel like they need to be able to quote however they want and change the
spread however they want. And that's just not possible on AMMs right now. However, I do feel
like AMMs still have a space and that will be reserved for specifically new tokens just because it's just very easy um
even after so many years uniswap v2 is still very very popular and that's mainly with um you know
a lot of these new new tokens meme coins um basically pump funds like all the new soft
20 curves
yeah that that definitely makes sense.
And I've definitely seen the evolution of LFJ over the past few years.
It's been exciting to follow along and awesome.
Yeah, I appreciate you making that opening argument.
And now I'll pass it over to Austin from Doppler and we'll love to hear his take more from
the AMM side of things.
Yeah, I guess like I find these debates funny. I've been doing them for like four, three,
four years now. It's like the same thing every single time. It's like, you know, AMMs are
going to die soon. Quarter books are going to win. And it's always like oh it'll do it next year and then next
year comes and uh hey i'm still winning so i do find it a little bit like it's a very circular
argument i don't know i guess like long term i think that order books are a good model i think
they're very in a local maximum of like an assumption that like uh liquidity gets provided by uh like hyper
sophisticated or providers um i think like the really big powerful unlock of amms is the ability
for people who are like retail liquidity providers to provide liquidity and it makes more sense to me
long term that amms will get more more sophisticated and will continue to eat.
Order books and the other way around, I think like, you know,
you said it yourself, like new asset issuance happens on AMMs because of the
programmability and like the ease of market use.
And it makes more sense if we believe in a world in which technology gets more
efficient, that the AMM will continue to get more efficient,
not the other way around where order books get more efficient that the amm will continue to get more efficient not the other way around where order books get more efficient i actually think that the future is like both of them
together and i think like what we've seen with uniswap b3 is like it looks like a
a dex and it looks like an order book in some sense like that's like the original design was
like a smooth order book and so to me it's of like, who really cares about what it looks like?
Let's just like give people assets. They don't want to trade. Let's give people what they want.
And it sort of seems like AMMs are giving people what they want. It'll continue to go in that
direction. Yeah, I definitely agree. I think if you, if you look at, you know, how AMMs evolved,
you if you look at you know how amm's evolved um eventually um we will see amm's being relevant
beyond the fact that you know they enable training in a non-chain environment because also because
of the fact that you know you have a lot of retail um participants that are able to provide liquidity
who might not be able to do it in um a concentrated liquidity ducks or an order book ducks um so those
are definitely definitely definitely valid points.
With that, I'll pass it over to Favashi from Kuru to talk more about the CLOB side of things
and what they're seeing with initial feedback on Kuru.
Right. Yeah, I'm all for CLOBs, but I think it's important to appreciate the nuance with on-chain.
What AMMs do is something irreplaceable, and I don't think Klops can fill that gap at all.
But at the same time, which is a better market structure? It's just that
Klops just works way better for better assets. So it's really a question of what kind of asset
do you want to be trading, and what's the best market to organize this asset turns out amm's work really well for assets which have low liquidity not enough demand not enough
liquidity providers and that's where they shine the most where retail can participate anyone
across the globe can be a part of this market but where order books shine the most is when there's
a lot of liquidity providers who are interested in market making or interested in trading this asset.
And that's where the global shine the most.
So at Kuru, we realized this early on and we didn't want to pick a side.
So we decided to put the Vault product on our order books,
giving flexibility for users to use the order book both as an AMM as well as an order book.
So picking the best of both worlds, we didn't want to pick a side. And I think the future is going to look like that. Like Austin said, most AMMs
are trying to become closer to Globs and most Globs are realizing that retail liquidity is
important. So they're trying to incorporate these products like walls and other things.
Yeah, I think that's what the future looks like. A hybrid of both and not really either of them in isolation
that makes sense just as a quick follow-up question to that do you do you think the the
stickiness of uh an emm in addition to the club is like in parts has to do with the with the ux
factor or do you think it's it's just because it's a sticky product that's been around and people are just used to it now?
I think it's more than people being used to it.
It's, I think, a brilliant way to get people to participate in the market and provide liquidity.
We can't expect them to run these bots or ascertain the fair value of the 20th derivative meme coin.
So it just turns out that providing liquidity and using a v2 style
will make more sense for that set of assets. But on the other side, if you want to trade
highly liquid pairs, then it really doesn't make sense to be doing that with a v2 style.
It makes more sense to be closer to a product. That makes sense. Cool. I'll pass it over to Kevin now from Clover.
Yeah, I think one of the biggest contributions that AMMs made to the DeFi ecosystem is that it actually invented a way to provide liquidity when high-frequency liquidity provisioning is not an option.
liquidity provisioning is not an option.
And it also opened up a lot of new opportunity that,
like, previously, in TreadFi,
retails could not provide liquidity in the order books.
And, like, I mean, they could provide liquidity
through limit orders,
but they could not, like, actively participate
in the liquidity provisioning side
as of, like, market maker.
So I think that is kind of like a YouTube moment for liquidity provisioning side as of market maker. So I think that is kind of like a YouTube moment for
liquidity provisioning. So I think there is a lot of contribution that AMS made and
there will be, you know, order books will not be able to have that like passive easy to easy liquidity provisioning feature without any like without
centralization point so there there could be something like HLP like faults but it is also
there is some centralization so I think there's like there's a part that order books cannot replace simply replace AMMs. But in terms of on the opposite
side, I think there's also only things that only order books can do because if you think about it,
like whether it is AMM or order book, it is all about how to bring the liquidity, right? How to
provide liquidity and what liquidity providers need and if you see
how the liquidity comes from within that like thread fight it is not just from
like small set of like liquidity like market makers that provide like both
side of liquidity actually a lot of liquidities are coming from the
directional bets on as of as of limit orders. So like pension funds don't do just to do the market
order with like billions of dollars with their month with their capital. So those limit orders
contributed contribute a lot to the liquidity book. And also for even retails provide a lot of liquidity through directional limit order bets.
And since AMMs cannot offer that feature with scalability,
I think there should be rooms for order books as well.
And also for sophisticated market makers, they are not able to provide liquidity with their more sophisticated algorithm using AMMs, which
means that it prohibits them from participating in the liquidity provisioning in on-chain
ecosystem unless they're using something like off-chain RFQs. And if you see more sophisticated
markets like options, bonds,
a lot of these markets are not really suitable for AMM-based liquidity provisioning. It is not about how sophisticated it can evolve.
The method that the fact that they cannot offer limit order itself
prohibits them from offering that feature.
So I think there's also
room for on-chain order books in the future as well. Yeah, I think that makes a lot of sense.
And we definitely see more market maker participation in the order book format, just
because it's easier for them to place orders, cancel orders, and make sure that they're
providing tight liquidity without getting picked off by toxic flow. But then at the same time, I do see a lot of market
maker wallets interacting with Univ2 pools. And I always get a chuckle out of it because the way
that they're always interacting with it is they're just arving the pool. And I remember this tweet
from back in the day from somebody that worked at Alameda. And he said that there was a saying inside of Alameda
that impermanent loss with a uni V2 pool only exists
if you're the person that's not arbing the pool,
which is kind of funny if you think about it.
But yeah, with that, I'll pass it on to PBJ from Purple.
And I think it'll be good to get your perspective
on how you see, um,
the, the AMM format versus the cloud format for perps, uh, given that, you know,
we've seen DEXs like GMX in the past and, um, yeah, like a perp DEX just has way
more volume and way more activity than, you know, regular spot DEXs.
So curious how you see these things.
Yeah.
Um, I think like the way I was thinking about this is, um, you see these things? Yeah.
I think the way I was thinking about this
is with vaults on a club, I think you can pretty much
replicate an AMM strategy.
It's similar to thinking about uni v4 with hooks.
You can pretty much write an AMM-style strategy market-making on a club.
But you really can't do what clubs do on an AMM.
And what I mean by that is clubs enforce basically time of when the order comes in.
So the earlier resting orders are the ones that get filled. And Uni v3,
if you think about it, there's no way for a sophisticated market maker to kind of be early
into an order other than just parking a ton of liquidity in a certain tick. So when you just
think about it from that perspective,
you can pretty much replicate any AMM curve
with a vault on a club,
but you can't replicate the most important thing
that clubs offer market makers,
which is Q position on an AMM,
other than just being like extremely inefficient capital wise
and just parking a ton of liquidity in a certain tick.
As far as like bobs and perps i think it just became very obvious that like gmx was like a good solution for perps when you can just do an amm but as hyperliquid has shown you really need a order book style DEX for efficient market making, deeper liquidity.
If you go to GMX or Jupyter or any AMM style perp DEX out there, you're going to get killed on slippage.
You're just not able to provide a ton of liquidity.
Market makers aren't able to cancel their orders or provide liquidity in the
ranges that they want to so they just rely on passive lps and they're just not able to offer
the depth that hyperliquid can but to add some merit to amms yeah you can list tokens faster but
i think now with the advent of more um vaults coming on board i think that's going to
change as well there's going to be more sophisticated vaults market making on clubs
where passive liquidity providers can still earn a yield and smart vault creators can also
create this new ecosystem of like interesting vaults and
strategies that weren't available before. Yep. I think that makes a lot of sense.
And yeah, capital efficiency is definitely one of the biggest things, um, that, that comes to mind
when you think about clubs and you know, what they enable in the, in the defined environment.
Um, cool. So, um, going forward, we can just make it an
open mic format. And I'll just bring up some of the questions that the audience is asking and,
you know, questions that I have and anybody's feel free to just unmute yourself and speak to
those questions. So the first question that I have up here is, if you look at most of the
leverage products that are built on chain within
within a DeFi ecosystem. So whether it's a lending protocol, it's perps, it's, you know,
leverage yield farming, any of those things, a lot of those primitives are usually built on the
premise of having predictable liquidity within the ecosystem. And what that basically means is there
is a there's an LP with a significant amount of liquidity where at any point, if you need to liquidate the assets that are being used to accommodate that leverage, you're able to do it atomically without the protocol accruing any bad debt from that, like being on the other side of that leverage position.
But when you look at order books, one of the issues with order books is that it's all just-in-time liquidity and bids can typically be pulled from the book at any point.
I'm curious if you guys have thoughts on the impact of order books on predictable liquidity and, you know, what the next stage of DeFi could possibly look like if a majority of liquidity is not sitting in AMMs.
a majority of liquidity is not sitting in AMMs.
Yeah, I think this is like one thing that I think a lot about,
which is like the benefit for AMMs is that like retail liquidity providers
provide liquidity and they want, when they provide liquidity,
they're doing it to not be Delta neutral.
Like it's like, it's like assets they're going to hold anyways.
And that's like really valuable because like during down periods,
they're going to hold these assets. They don really valuable because during down periods, they're going to hold these assets.
They don't really have to hedge the risk as much,
which is valuable because that's actually a benefit for them,
not a bad thing.
And so long-term, they are willing to provide liquidity
during high volatility periods. and that liquidity is much
more predictable which is like a really valuable thing for the ecosystem yeah um i i agree with
that uh for amms right uh this liquidity is not sophisticated so they don't really care about the
volatility or any of these factors so it's always available for you to go swap with them.
The same is not true for order books.
However, with a kind of a hybrid approach,
you can arrive at a similar situation
where there's still retail liquidity,
but this time living in an order book.
I'll have more to share on this in a few weeks,
but essentially, bids that don don't disappear that stay on the
audiobook because the retail part has been set a range.
I think it's good for the trader but not really good for the LP. Like you're right
that in AMMs, LPs the liquidity liquidity is more reliable, it's always there.
Like, if the asset is going down tremendously, you're kind of forced to, like, buy as it's going down, right?
Keep the bids up.
And that's good if you're on the other side of your trader and you are, you know, getting a good price for a token that's going down. But you don't want to be buying a very down-going asset in that scenario.
So it's a bit of a trade-off, I think.
Yes, it's good for the trade-off,
because it's a reliable liquidity,
but for the LP, that's kind of how you get wrecked.
That's kind of why in order books,
you see market makers,
they'll probably widen the spread, or if they are providing liquidity, they're hedging it somewhere else.
Right.
Yeah.
My general thoughts is like, it also depends on what you're trading.
AMMs will still be prevalent just because most of the tokens have tens of thousands of tokens,
and most of them are low liquidity.
They're pretty much all altcoins, right?
And you can't mark and make all of them.
So there'll always be demand for AMMs.
But for some of the majors like Bitcoin, Solana, ETH, I think it makes more sense to put them on an order book instead.
I think it makes more sense to put them on an order book instead.
Yeah.
And regarding composability with other protocols on-chain, right?
So there's a lending borrowing protocol that wants to liquidate assets
during a market event.
Market makers can pull out their bids,
but LPs and AMMs don't pull out bids because they just have LPs.
But I think what's going to happen
is we're going to achieve this equilibrium
where at a point it makes sense to be not pulling out all your
bids and eat some of those liquidations
because you're getting a better price
here than you would be getting anywhere off chain.
So it kind of doesn't mean that you have to pull out your bids.
It's just that you can't rely on this liquidity.
That's only the downside.
I mean, in general, the argument that the question assumes applies also to threat-wise order books.
But I don't see Binance having any issue with liquidation.
So I don't think and also like technically you can AMLPs can also pull out if they are in market volatility.
It is just more, I think it is more like whether the liquidity are coming from more sophisticated participants or rather like more passive participants.
And also, as I said, not all the liquidity are coming from a handful number of like delta neutral market makers.
A lot of those limit orders within the order books are actually coming from directional bets as well.
So I don't think that even during the market volatility,
it won't be like everyone just pulling out from the order book
and the order book becomes suddenly empty after a block.
So I don't think that will be a huge, huge problem, assuming the efficient market.
So just a quick interjection there, but I think one of the things that we've actually seen and
the reason why markets become really volatile as soon as you have a move in a major asset like
Bitcoin by like two or 3% is because the liquidity that is coded on the books actually becomes a lot thinner because there's a
lot of inventory balancing that is happening across exchanges um and then the market maker
also ends up taking on one side of the position where you know they're um overall trying to be
delta neutral um so i think that might actually be a counter to your point, Kevin, but I definitely do see what you're saying.
My take is that on a lot of chains, the liquidity that you see on these AMM pools are actually
part of the governing foundation.
And I'm guessing they must be eating a lot of IL because of that but they also have a duty to
have a deal with theirs so people can can trade stuff that's generally why I think
I don't know for sure but that's just my kind of gut feeling I mean yeah I think that's like
that's actually Austin that that's for sure like if you look at maker, that's literally what they do for adding liquidity to the maker token. So why wouldn't these foundations do the same thing?
Yeah, I mean, I think like people say that a lot about Uniswap and on Ethereum, and it's just like, provably not true.
I mean, like I've met a lot of these people who are liquidity providers and they're like hedge funds and their family offices and things like that.
So perhaps on some like not as large chains, but I do think that like the narrative that leaks like the only LPs are sort of like foundations is just like provably incorrect.
incorrect. And so I do think like maybe on like some chains, that's true, but I,
And so I do think like maybe on like some chains, that's true.
especially like on Ethereum and specifically if like Uniswap, it's just like, no,
like, it's like, I've met most of them. They're like real people.
I guess just shifting gears a little bit. If we were to talk about, you know,
what these platforms look like in a handful of years,
we're getting to a point now where there's a question being posed quite a lot,
which is for assets that trade on both a DEX and a SEX,
I'm curious to hear your guys' take if you believe that there will be a world
where price discovery is actually happening on DEXs.
on taxes.
Yeah.
Yeah.
My take on this is I'm optimistic
that at some point there will be price discovery on chain
with a few conditions attached to it though.
The infrastructure
supporting this has to
grow at that speed as well.
As of today, it's impossible for
chains like Monero, Solana to
compete with something like Binance,
or even Hyperliquid for that matter.
But there is one important advantage
that on-chain market making carries.
If most of the logic lives on-chain,
it's essentially like having a co-located market making
bot next to the Binance server for all practical purposes. And I think Toli speaks a
lot about this, but if there was a world where infrastructure caught up, I don't see why
prices curry for global assets might not happen fully on-chain, just given the fact that it's
not a latency game anymore, it's more of how much logic lives on-chain versus off-chain.
more it's more of how much logic lives on channels of chain yeah I agree with
for Vashi there I think the reason why we haven't seen that yet is just because
the infra still is not there in order to have a very performant order book you
know you need to be on the level of speed of Binance right so like we're
talking about you know maximum, maximum 10 milliseconds.
Last I checked, I think like even the fastest blockchains like Solana is probably like 400
milliseconds. So, you know, we're definitely getting closer than years before, but it still
needs work. And the main thing is like, first of all, having infrastructure. So being super quick,
fast block time, but also super cheap because
market makers they need to post a lot of orders but then cancel them and sometimes they need to
post like hundreds and cancel them a lot over just a few seconds the second thing is just getting
deeper liquidity and that's something that will take time um i personally think that the way that we're progressing we will see
a future where you know slice discovery does happen on dex
but it will take time i think the liquidity part is the hardest part to solve
just to dissect that further for for people in the audience like there's sort of two answers
within that there's one where you, infrastructure still needs to improve and latency and, you
know, shorter block times, allowing people to quote more cheaply, all needs to happen.
But, you know, for Vashi, you touched on this, like it seems clear that the answer is not
that there's a race for latency there.
Like an on-chain order book will never be more performant than a centralized system.
And the closer you inch to that, the more you might lose the properties and the underlying benefits of running this in a decentralized way.
if you have to pick a specific way, you know, a single way in which this happens,
is it that the majority of participants and volume and liquidity exists for an asset on chain,
you know, such that it's higher than that of centralized exchanges or like what actually gets us there?
of centralized exchanges or what actually gets us there?
Personally, I think I'm even more optimistic than other guys here because I think, of course,
the infrastructure matters a lot. But I don't think we, in order for that to happen, I don't
think we have to, on-chain has to be as performant or even more performant than centralized setups and which is impossible.
But the reason why I think it can still happen is that because like being on-chain gives you a lot of benefit, like its own benefit, right?
So you cannot have like, for example, KYC list trading on a centralized startup.
So I think as long as we have enough trading volume happening on-chain, the liquidity will come.
Because market makers will eventually make money because if there's a lot of trading volume, even though the latency
is not really ideal. So I think it matters more in terms of how big the ecosystem can grow,
rather than can we just beat the centralized setup in performance.
Yeah, I think that definitely makes sense.
And yeah, one of the one of the really interesting questions that we got from the audience is
from Larry from Drake Exchange, but he's asking more about hearing thoughts on the different
types of costs that protocols have to bear for acquiring liquidity for each of these
models, whether it's AMMs having to add token incentives to get for acquiring liquidity for each of these models, whether
it's AMMs having to add token incentives to get people to LP on top of the swap fees or
paying market makers or giving them rebates.
So curious how you guys think about, if you were to look at the same tightness of liquidity
on an order book versus an AMM, how do you guys see those costs differing and what's more efficient?
I think just again going back to the capital efficiency of AMMs, you just need to incentivize
a lot more liquidity to provide the same spreads and depth as a club. I think I'm going to keep coming back to this.
Vaults are the future.
And if you can kind of have a mix of both vaults on a club,
you get the best of both worlds where you can have the club efficiency
and also some passive liquidity.
But yeah, I think even when you look at liquidity provisioning,
amount of, for example, interest in the perp space, the amount of fees and value that has
to go to LPs on GMX is pretty significant. And they need to continue doing that to have
liquidity on the books for people to trade. I don't think it's that expensive to have market
makers on your book as long as you have good order flow and interest of people wanting to trade and
what i mean by that is like not just like uninformed order flow so if you have that uh i
think hyper liquid is kind of showing that vaults plus a club is definitely superior in terms of like
the amount of like liquidity you can attract and also the preciseness of that liquidity
um since you mentioned hyper sorry let me just ask a question since you mentioned hyperliquid and
um now they're venturing into spot do you see a future where because they're essentially a club
right you see a future where they will become the dominant decks over say unisop so one of the things
i was going to mention in the earlier question with price discovery i think one thing that people
are kind of calling out is each of these spot assets are being you know
launched on different blockchains and the advantage that finance has or any central exchange has is
they can host all those assets whereas if we were to host like solana assets on monad or
ethereum assets on solana there would be like wrapped versions and they would just cause a huge amount of like
infrastructure headaches.
So it's just hard to see how like spot trading
fully moves on chain unless like one blockchain
becomes like the complete dominant force for that.
Same thing with Hyperlequid, right?
Like I think, again, they have a spot X
and they're having this like third party
or whatever you want to say with unit
having to bridge those assets over there.
One, you're trusting unit as a good actor
when they're like, I don't know, three validators
or whatever they have.
It's just really hard to imagine a future
where people are bridging
ethereum assets over to hyperliquid and that's winning over binance that's where like spot
trading uh is a little bit more i don't know if i'm that sold on that but perps obviously it's a
lot easier to do that because it's just usd stable and Oracle price, and you can pretty much list any new market.
So one quick point on the price discovery thing.
When Trumpcoin launched, it was listed first on Hyperliquid for like six hours, and they had a ton of volume there.
And price discovery was happening on Hyperliquid for those tokens.
And even like these newer token launches, they're usually listed on
Xs first.
I think that is a totally valid
way of eating away at the
off-chain market,
centralized exchange marketplace, by
listing tokens faster
and having reasonable liquidity
for them. And that's
how I think we can win the
moving trading on chain.
Yeah, I think these are all really interesting insights.
And speaking of Hyperliquid and their focus
on Audible and Spot, I'm also curious to hear what you guys
think about Cloud Texas or just Texas in general integrating
more of DeFi AI agents to allow their user base,
especially the more retail user base who are not very sophisticated, to use these AI agents to create like automated market making strategies
so that they can also earn more fees as a sort of like their own AMM instead of relying on a traditional AMM.
And what do you guys think about the space
and how it actually is going to involve more DeFi AI agents
for not just making trades, but also market making strategies
from the retail side?
It'll be interesting to see an AI agent run a market-making bot on top of one of our
clubs.
I don't know if we are there at a stage where people are trusting their money with an AI
agent, or maybe I'm misinformed.
I think it's all right for, as things stand today, it's easier for users to use the AI agent as an interface.
Instead of clicking buttons, they probably issue commands, and that's what happens.
But maybe, like you said, we do see a future where the AI agent can pick between different risk profile boards to then go deposit your liquidity.
So that is a more plausible future, I would say.
That is a more plausible future, I would say.
Yeah, I think just to add on to what the last speaker said,
I think it'd be interesting to have, again, more vaults
where instead of people just taking their airdrops
and not knowing what to do with it and just selling it,
where they can deposit in some sort of market-making bot.
I'm not really sure if I would trust an AI agent to do this,
but something like that to basically put your crypto to better use and add liquidity to that token.
It would be interesting.
I've already seen a bunch of teams kind of tackle that having an LLM be your trading
interface for perps.
for perps, there's a lot of Telegram bots that allow you to do that as well.
There's a lot of Telegram bots that allow you to do that as well.
So that's an interesting avenue for that, where you can basically talk to an agent
and say, look, I just got this huge airdrop.
I want to hedge that position.
What is the best way of doing it?
And then that agent would kind of give you some options for you to pick from.
So that would be an interesting avenue as well.
This is assuming the user knows what hedging means.
Yeah, I think generally, the way that I see agents, or let's just call them automation agents for
the sake of a better argument, the way I see automation agents being enabled in DeFi and specifically with indexes
is actually less of the intelligence part is outsourced
to agents, but more of the automation part is outsourced to them.
So what that could basically mean is setting up really complex triggers on
what would need to be triggered for you to buy into a position.
Like I know Phish is posting technical analysis and all of those things these days. And maybe
he has some really complex technical analysis that once the 21 are moving average crosses
like the 50 are moving average or whatever you guys look at, just trigger a sell on the asset. And
you know, whenever the reverse happens is when you trigger a buy order. So essentially looking at
the type of different conditional orders that you can set up on centralized exchanges, but being able
to dial up the complexity of that by quite a lot. I think that is definitely one of the things that AI agents, sorry,
automation agents can enable. And yeah,
I think AI agents is a term that's pretty loosely thrown around.
Whereas in reality, a lot of these agents are just automation agents,
which isn't something particularly new, but it's good that, you know,
it's being rebranded.
So we'll probably end up seeing a lot more modularity and a lot more
integrations that you can build out with different tools um with that being said um i definitely love to switch
gears in the last segment of um this episode and you know um talk more about the convergence of
amms and clubs which is something that you know both uh fish and for what she spoke to briefly. And, you know, think about the eventual goal here, which is
not necessarily AMMs versus clubs, but the eventual goal is AMM and clubs versus centralized
exchanges, which, you know, brings us to the important topic of, you know, how do we collectively
as AMMs and clubs, like the entire tech sector, how do we take on centralized exchanges? What does
it mean to be on priority with them
and attract the next generation of crypto traders
to come on chain to trade on DEXs versus centralized exchanges?
So yeah, so would be keen to hear you guys' thoughts on that.
I would say focus on building good apps
and the users will come.
And once you have to take a flow
the liquidity just flows in that direction the only reason binance liquidity is so sought after
is because of the huge user base they have and not really anything apart from that so
i would say most apps need to focus on getting the users first and then worry about liquidity
need to focus on getting the users first and then worry about uh liquidity later yeah i'd echo the
same i mean it's kind of like the robin hood model right you have to own the end user and
you're starting to see every dex um kind of do that like it's not enough to be just a liquidity
platform anymore you need to also uh have the end user in mind. That's why I think Uniswap has a wallet or an app now.
Jupyter has an app.
They're all building apps where people can actually interact
and have basically the retail users
trading on their platform.
It turns out even a lot of people,
they swap through the wallets nowadays as well.
Phantom makes good revenue.
Rabi as well, phantom makes good revenue um rabbi as well metamask and basically like you know i think market making is now becoming where who can capture the most retail flow um but to answer your question abdul about how amms and
can live together i mean you, I think they definitely can.
And when Bonad does go live, we are planning to launch both an AMM and an order book.
I would say the AMM will be more higher priority for us in the early days because, you know,
in the early days, like there's not much, there's not much like assets you can offer
for an order book. It'll take time to get those assets.
You need to have collateral for market makers to market make with.
You can have aimables and people can immediately start providing liquidity
without you worrying like where to get that liquidity from.
And yeah, I think for the most part, like, you know,
the thing that people will look forward to when Monad launches is first of all to buy Monad tokens, but also to buy, you know, new tokens, right?
And I think in both cases, AMMs will definitely be the predominant model backs in the game.
model backs in the beginning.
But as things progress, I can start to see
things like especially the majors
like the Monad token, when it does go live,
you know, people will want to get
a better price execution
and therefore it might be better for them
to get that from an order book instead.
Makes sense.
Yeah, sorry, Kevin, did you want to say something?
I think that there should be external... See how Binance won the market, even though there were already a lot of mature marketplaces there when Binance came out,
was that they were very good at listing very trendy new crypto assets very proactively,
so that people can have more toys to play around.
people can have more toys to play around.
And this applies to the growth of Upbit exchange,
the largest crypto exchange in Korea.
So they also started very late.
And back then, Bitsum and CoinOne
already captured a lot of market.
But Upbit's strategy was that they just integrated with Bittrex
exchange so that they could enable like kind of like CX aggregate a lot of this retail
long tail markets within to Korean retail users.
So I think one of the things that,
like as an exchange platform to thrive,
one of the things we can do is that we can bring more interesting asset classes
that are not available on other venues
and that will attract a lot of users
and kind of like dopamine, which also can be something like RWA, but also something like more DeFi native innovations.
like 10x leverage index for S&P 500, which is impossible due to regulation, but not only
regulation, but also in terms of heavy operation. But if we can, because of the composability and
programmability, we can have that kind of more progressive instruments with DeFi, right? So if
we can bring that, which is also usually not really feasible to
provide liquidity with AMMs through order books, I think we can bring a lot of attention and
trading volume that other CEX cannot or has not been able to address. So I think that is also another way for on-chain order books,
or not only order books, but on-chain DeFi ecosystem to thrive.
Instead of just competing with the same old assets like crypto-native,
meme coin, and just more leverage trading on BTC.
new, like more, just more leverage trading on BTC.
So we need to like grow the market,
like market like TAM itself.
Yep. I think that makes a lot of sense.
And as you look at the DEX sector,
gaining dominance over centralized exchanges over the last year,
if you look at the chart, it's actually pretty bullish.
I think at last I saw, I was somewhere, you was somewhere trending up and it was at like 30 to 35%. And over time, if you look at what ends up becoming kind of like a chicken and egg
problem with exchanges is like you have liquidity on one side, which is usually typically provided
by market makers. And then on the other side, you have retail flows, which are attractive and, you know, kind of, um, summon the, the liquidity
from market makers. Um, and the reason why it's the chicken and the neck problem is because
retail wants the liquidity to trade on, you know, major assets like Bitcoin and Ethereum.
But at the same time for market makers to post liquidity and, you know, maintain, um, tight
books, they would want a healthy
amount of retail flow so that it's profitable for them to market make on whatever that exchange
is. However, one of the interesting things that we saw last year was, you know, beyond
just these two things, one more thing that ends up becoming a moat for exchanges is token
discoverability and token creation, which is where, you know, like Pump.fund
was the biggest DeFi app of last year,
you know, pulling in millions of users to Solana.
I think they've made over six, 700 million in revenue
over one year, which is, you know,
unheard of for any DeFi app that has ever existed.
And, you know, as we kind of look at that
and then, you know, how Pump.fund has progressed
over the past year, where I think about a month ago, they launched PumpSwap, which is their own MMM.
So now tokens graduating out of pump.fun, get bonded and then eventually have their LPs created on their own decks versus Radium.
Right now, I think they have the highest volume in all of Solana, which also kind of ties
in back to the point that Phish made about owning the user and owning the entire user
journey.
So I wonder how you guys think about using that as an angle and competing with centralized
exchanges, because obviously the biggest benefit of being in an on-chain environment is that
not only can you trade assets, but you can create them from scratch and let users create
those assets. not only can you trade assets, but you can create them from scratch and let users create those
assets. Versus if I go to an exchange like Binance or Coinbase, I cannot create a new token.
And then eventually, given how a lot of token launches have gone over the past year
and retail users just end up getting burned, people feel like they have a better shot at
trading meme coins or any of these coins that are launched by retail people. It's also super
relevant for what Austin's team is building with Doppler.
So Austin, I wonder if you have any perspective on this.
And then Fish, I know you guys also have the Token Mill product.
We'll love to hear your thoughts as well. So I go first.
Yeah, Austin, I think we're not able to hear you.
Maybe if you can rejoin in Phish, yeah, you can go first.
Yeah, I think new token launches is, well, obviously a very lucrative area.
Like everyone wants to be first and they're happy to pay high fees for that.
And that's why, you know, despite Pump.Fun's dominance, you still see like almost every week a new token launching platform.
And hint to hint, we are also launching TokenMillv2 sometime this summer.
But yeah, this completed domain of DEXs, this is a very decentralized thing.
It's something that I don't think Binance and Coinbase will be able to compete in this area.
This is strictly where DEXs and DeFi can really win and dominate.
So yeah, I'm very bullish on it.
I think it's not going to stop.
I don't think that the problem is solved.
Like, you know, despite many people's best chances
of trying to, you know, gain market share from pump fund,
there are still different ideas.
There's still a lot of problems to solve.
Like, first of all, the fairness thing
and also snipers, etc, etc.
I think, yeah, I think this is something that will continue to innovate.
It's obviously very controversial. People don't like it. People like it.
But the data shows that there's just still huge demand for it.
Makes sense.
And for Vashi, PBGA and Clubber, if you guys want to add, feel free to as well.
I was just going to say quickly on the competing with sexist part, I think, composability,
I think is one of those things that Dey can lean on um if you look at kind of like what really like picked off
a lot of like the salon activity uh it's obviously building like really good products but also some
of the airdrops like gito was like a huge airdrop that added a ton of capital the ecosystem so
people can start using that capital to try new apps, trade on new apps, et cetera.
One of the biggest hurdles that DEXs have,
that centralized exchanges don't have,
is just the actual fiat onboarding.
And the only clear way for DEXs to do that is,
you know, bootstrapping the ecosystem with, like,
building good products,
having good airdrops for users of the ecosystem
to then use that to start trying new applications.
And building the ecosystem that way
is also a huge advantage that DeFi can lean into.
Yeah, I also think composability is the key.
So like in on-chain environment, we don't have to actually, like one dev doesn't have
to own all the end-to-end user experience, I think, because like even without that, we can't,
the ecosystem, the whole ecosystem combined can own this like end-to-end experience.
So like even without like when Pump.fun didn't have like pump swap, they could still use
a radium as an infrastructure so that they can still have a very successful token launchpad without worrying about
the trading infrastructure. So I think they can always choose to own their swap end-to-end feature
at any time, but the fact that we can build based on this composability is, I think, one of the key benefits that on-chain DAXs have compared to CXs.
Yeah, that definitely makes sense. And yeah, with that, I also see that we're almost at time.
Unfortunately, that Austin couldn't add his two cents.
For those unfamiliar with Doppler,
Doppler is building a token launch pad
that improvises on the model being used by pump.fund
where instead of having the price start low
and then go up as it would in a typical bonding curve,
the Doppler protocol actually uses a Dutch auction where the price starts higher,
goes down and then eventually goes up again
into the bonding curve,
basically solving for the problem of token sniping,
which ends up creating a really unfavorable experience
for people that are trading meme coins.
With that, yeah, I would like to thank all the people
that joined speakers today. It was
a productive conversation, so I appreciate
having all of you guys on, and also to all the
people that tuned in.
It was a pleasure chatting with you guys about this
and really excited about the possibility
of a lot of DEXs like yourselves
on Monad, going head-on
with centralized exchanges in the future,
and creating a kind of creating experience that people haven't experienced before.
Thank you, Abdul. Thank you, everyone. G Monad, everyone. See ya.
Awesome. Thanks everyone. Monad.
Come on. Take care guys.
Thank you for having me. Bye.
Bye.
