Hyperliquid’s Bold Vision for DeFi | Jeff Yan | S9 E3

Transcript

That's why centralized exchanges have been so successful so far is that you get to basically cut a lot cut a lot of these corners or or I guess you just kind of like get to not think about a lot of these problems. Like you know me is one of them but like yeah there just like a lot of unsolved technical problems like how do you how do you coordinate these validators like do things that still allow for a seamless trading experience right but I think all these can be solved and it's also like it's it's like a fun thing to solve. Also, I think it's like smart people naturally are attracted to hard problems and we'll solve them all. And so, yeah, I I would definitely bet on the global financial system being on a decentralized platform like owned by the people. Quick shout out to

Permissionless 4 coming this summer to New York June 24th through June 26th. We'll have speakers from all across the industry including Hayden Adams, Kyle Samani, Guy Young, and more. Get your tickets today by following the link in today's show notes, and make sure to use code Bell10 to save 10% on your purchase. See you there. Hey everyone, quick disclaimer before we get into today's episode. Nothing said on Bell Curve is a recommendation to buy or sell securities or tokens. This podcast is forformational purposes only and the views expressed by anyone on the show are solely our opinions, not financial advice. Our guests and I may hold positions in the companies, funds, or projects discussed. All right, everyone. Welcome back to another episode of Bell Curve.

This is a great one. I'm joined as always by my trusty co-host, Monae. Monet. Hello, sir. And we've also got um Jeff of Hyperliquid or do you go by Elfe? Uh what's the preference here, Jeff, for this problem? Jeff, I don't know where he came from. Uh not sure, but it's it's a nice nickname. Um okay, we have got an enormous amount of ground to cover today. Uh so we we ended our last episode with Hu and Rob Lesner actually kind of on the subject of Dexes. And so we thought there'd be no one to better discuss the future of Dexes in general with than than you. Um, and maybe before we get into it, could you just give us uh a really quick maybe two-minute overview of the history of Hyperlid and how you guys have gotten to

this point? Uh, because I think in the last year, you've not only dominated in terms of volumes, but the architecture uh, and kind of the future vision of Hyperlid as a perplex towards more of an L1 has changed uh, a decent amount. So, can you just kind of remind viewers, you know, how did you start and then how did you get to where you are today and then we can get into some of the weeds of the architecture? Yeah, sounds good. Um, so I'm Jeff. Um, I lead Hyperlid development and we're a super small team. Um, mostly Americans. I myself grew up in the US. uh went to school uh at Harvard, studied math and CS and after school went to do market making in Tradfi, so at a firm called

Hudson River Trading. Um learned a lot about markets and thinking about them rigorously, but ultimately was pulled into crypto and um yeah, wanted to build basically DeFi kind of spoke to me even in 2018 when um there were basically no users yet. Um, so kind of got way laid and ultimately ended up just doing prop trading in crypto there and from that we kind of were plugged into all of the markets uh both CFI and DeFi and had a pretty big presence and when FTX collapsed um that was kind of the sort of impetus to start thinking about what ultimately became Hyperlid.

basically saw we saw the problems with FTX firsthand and witnessed um a mental shift where people kind of realized that you know crypto is all fun and games until something bad happens and then you then you realize that the value props of crypto itself actually ironically um are sort of antithetical to how these centralized entities are. Jeeoff, can I can I interrupt and ask maybe a slightly heretical question? Is there anything that you think FTX did particularly well? Because I remember at the time FTX was a beloved uh exchange especially for traders. They had a lot of good stuff to say. Um and you know I'm front running a question that I

wanted to ask you later but it seems like the go to market for a lot of new exchanges who've been able to make or splash on the scale that Hyperlid has is you start to kind of innovate around uh listing assets more quickly, right? And this is kind of everyone from Binance to I'm not comparing Hyperlquid at all to FTX. At one point they did okay as an exchange, right? But that's kind of an early step of go to market and iteration. So I'm curious from your standpoint, it's always actually someone a question that I've wanted to ask uh someone that does understand market microscure better than myself. Did FTX actually have good trading technology or was it just Alamita on the back end?

Like what's your view there as a trader? And I'm curious to see what you thought they innovated on versus what was smoking mirrors. Yeah, it's a good question. So, so we actually never touched FTX at all. Like I was always just suspicious that um Alamita and them. So, basically you just hear all the stories. It's like they share the office, right? There's like really like same people running the two. Um so, you know, obviously hindsight 2020, but I will say we we never touched it. We we just it was for us it was just like a red flag from the beginning. Um and so I I can't speak to the you know like using the exchange but um I I think they did a lot of things obviously they were very successful right so um crime doesn't help you

succeed it you need something like some core product that you build and I do think they they had some things there so like they're I think they did things like you know they they kind of studied all the other exchanges and kind of present handed the data in a very easy easy way to their users saying like I don't know if I don't know if you guys remember the there was a like a basically a CSV that they published every day with like volumes and like they had some formula for like the amount of wash trading happening on this venue and they like kind of ranked stack ranked the exchanges. I think that was like super good like transparency for um users and um and what I mean their their UI was I I think good. it was like very clearly designed by traders and uh I

think that's that's an overlooked point which is that like building a trading interface is actually it's it's like a very complex product and you need to you need to cater to many different users and you kind of there's kind of like one canonical front end especially if you're building a centralized exchange and so um there are a lot of just micro decisions on you know sort of balancing the needs of different people who probably want like opposing things right like the pro users wants like all these like bells and whistles and like um really basic users like first-time users don't even know what they want but if you put all the bells and whistles they'll just be super confused and so um they obviously kind of I I think they executed that

well awesome sorry I didn't mean to derail there so back to back to Hyperlid um and maybe talk to us about the origins there and then again uh how that evolved to Yeah. Yeah. So going back to that, so basically with FTX's collapse, we kind of saw this need for uh sort of user focus or something that users love but also something that users can trust and really aligned with the original ethos of DeFi. And so yeah, it kind of started as a per DEX. Um, in building it out, we kind of realized that basically all the infrastructure in crypto is kind of not production ready for this specific use case of like uh high throughput, low latency, global exchange, even if you're

just focusing on crypto, but let alone like all of finance. And so we we basically never backed down from the challenge. we're just instead of compromising like okay let's just like drill deeper like can we build the next layer of infrastructure below the one that we're currently building and ultimately ended up at uh building an L1 itself it's like I guess for for the less technical users like a blockchain basically so um if you look at Ethereum Salana even Bitcoin they're all just kind of these public ledgers where different nodes around the world can agree on some world state and what that state is just depends and like how it evolves also depends on the different chains and like none of these were suitable for the sort of ultimate vision

of Hyperlid which is to be um like a trusted financial layer for you all of finance to kind of be built on top of and I guess we'll probably go into this later as well. Um, so yeah, Hyperlquid kind of grew from just a perex to financial infrastructure powering a perex, but also other stuff. And then this other stuff kind of came along as we were thinking more about really moving finance on chain like per is a really big part in crypto but just having like single asset denominated derivatives like linear derivatives uh you know lets people express their views long short but doesn't actually let you kind of build all the building blocks and ultimately

um move people's sort of financial lives onto this new and improved platform. And so part of that came with like spot launch like launching spot trading. But then really like I think the last piece that took us a while to kind of arrive at and like really commit to building was the hyper EVM and obviously like standing on the shoulders of giants here where like Ethereum really pioneered this sort of general purpose smart contract system and letting and all the tooling and ecosystem around it. And I think it's like super cool and that was really what ultimately led us to decide that like the EVM is the way to go. Just like there's like a lot of developer mind share and just culture around building applications that trustly

interface with other applications through smart contracts and uh and just bringing all of that expressive language and technical knowhow to just like have that work on hybrid liquid. Um and like that that was basically um probably like a half year long endeavor and we the the vision here was like not just to like tack on an EVM so that people could play around with it. Um, but really it was like for us it it was it was a super important component into how like how users would ultimately end up interacting with these natively built primitives like order books, per swap trading, etc. And just letting just having that like fully composable link

between the EVM and what the other stuff which now is called hyper core um was was it was like a very tricky interface to get right. It went through like many iterations but uh I think the where it is now is probably not final either. it's like con constantly evolving but I think people are starting to realize that the power of so I've correctly built abstractions there and like maybe one one example um just I guess this is much longer than a two-minute in um one recent example is the the USDT launch which you know was in collaboration with layer zero so two very big crypto protocols and kind of like coming in and just like plugging into hyperlid so seamlessly because they're they're familiar with EVM chains and adding it

as like another EVM chain in their mesh network. Um and like basically they really leverage this abstraction so that um they can do all the trustless bridging and minting natively on hyperlquid without bridging risk and then let users on hyper core sort of directly interact with their asset in a way that they don't even need to touch because that part is built into the protocol itself. Um, I think that that kind of like that kind of like aha moment, I think there'll be many more of those as um as people really like uh kind of familiarize themselves with uh yeah with with what is going on in the ecosystem. This episode is brought to you by Marinade. Marinade is a high performance staking delegation platform

that brings both liquidity and security to the Salana network by connecting soul holders with the best staking rates on Salana. Follow the link in today's show notes to learn more. I would uh I would love to hear kind of maybe if you have more thoughts on um what that that sort of close integration between the the perpex and the uh spot markets versus the EVM will have going forward. Are there any like use cases that you're super excited about um from that kind of new architecture where an exchange and a and a L you know L1 EVM are are really intimately linked?

Yeah. Um yeah, there are many examples. Um trying to decide I think I can go into one that's like it's a bit complex but I think it's like really interesting. So another big component of the L1 is is staking and it's a proof ofstake um consensus algorithm. So users need to basically delegate their hype to one of these validators. Um it's permissionless to spin up a validator. So anyone can self delegate the 10,000 hype and then basically spin up a node and broadcast like hey here's I'm here and like you know delegate to me and then uh everyone kind of proposes blocks in proportion to how much stake they have delegated.

Pretty standard for proof ofstake networks. Um what's interesting about staking on hyperlquid is that more and more the staking kind of has a different utility for different users. And so one part you saw um relatively recently was that um staking now is a is an input into trading fees. And so it kind of aligns users and stakers u which I think is is like a really just positive thing for the protocol as a whole. Um yeah it's an input into trading fees is you stake and you have lower trading fees. Exactly. Yeah. Uh it's no other venue does like that's something Binance has a similar thing where you need to stake a certain amount to hit a certain tier but on Hyper it's like you can you can trade as much as you want you can stake as much

as you want and those two together like each provide their own like I guess discount into the trading fee and it's all computed onchain. So interesting. Yeah. Yeah. Yeah. And uh but but like another even more I would say like cutting edge application of staking is that like HIPP3 was announced recently and there's still a lot of community input going into it and testing and whatever on test net. So it's not close to being live yet but it's it's a very big milestone and this is the permissionless per um sort of like uh I don't know like uh scaffolding like framework if you will. Um and staking is a big input to this because with per

unlike spot like spot order books are kind of are very well defined and so most exchanges actually start as spot exchanges and only later on venture into derivatives but um per I think like the whoever is setting the oracle and market price sort of have have a lot of power to kind of like mess with the markets if if they were to choose and staking is kind of the solution here which is that usually staking only keeps validators honest but in a hyper liquid staking also keeps sort of like per deployers and like oracle updators. I think the Oracle updators will mostly be like other protocols but you still kind of need to hold them accountable like the protocol needs to hold the other protocol accountable.

Um so yeah staking is like a way where the the deployer of the perked market needs to put up a massive amount of stake and during the unbonding period of like unstaking they could be slashed. Now this could be for various reasons like within consensus the the slashing is usually for malicious things such as like voting voting on two different blocks at at the same height just like clearly malicious provable bad stuff. Um, with per deployments I think it will be similar. It's like you specked out this market and you kind of you violated the spec that you put out and liquidated a bunch of users like clearly very bad and there needs to be disincentive to

doing that otherwise the whole no one would trade per so I think going going back to the original question like how does how does the EVM integrate? I think there'll be many per deploy players who don't who don't want to or can't afford to put up this capital, but they can operate a really good really interesting per market that users want. So this per deployer could basically create an LST and the interesting thing here is like LSC would be on Hyper EVM composing directly with hyper core. So like not not even like a not even like a multistake that then just like goes and stakes but rather like the contract itself can stake. So it can be like

fully composable onchain. Um we get a bunch of users to kind of stake to it. the governance of the thing can like just be the contract itself and it can say like okay once we hit this threshold um we will send like the contract will permissionlessly processly execute this transaction to deploy the per users can say okay I'm okay with that like um I want to put up my my my hype uh stake it in this way and the deployer of the protocol can actually then share let's say like some percentage of the fees on this per market back to the LST um constituents And this whole thing, the the the key thing here is like unlike other protocols where the LST is basically like a like a smart contract that then

has permissions like separate multi permissions to like do stuff with the funds, it's actually like fully on chain which I think is like super cool and um so yeah, I think I think that's one example basically. like like these uh perps kind of like um backed by their users I think would be like one cool way in which the EVM and the hyper core can interact. Jeff that's super interesting. Um, okay. So, you know what I found myself wondering, you know, when we're describing this and maybe to just, you know, again rewind and summarize sort of the history of of Hyperlid, uh, which is it kind you guys kind of started as a perplex um, with a bridge to via arbitum, right? Like kind of the, uh,

that's originally how a bunch of the, um, the assets flowed into Hyperlid and now you built your own sort of instantiation of the EVM um, and our bonafideed L1, right? And uh you know I'm curious when I think about what you guys are experimenting with um a lot of other protocols especially on Ethereum might have thought about building an L2 right um and when I look at you know what like Monad is trying to do today or base uh you know they're all kind of experimenting around um what changes they can make to the EVM because that's kind of the bottleneck for throughput um at this point uh but they've chosen to build this in L2 with a singleized se or a centralized sequencer. you guys have gone in another direction and are innovating and experimenting with your own sort of souped-up version of the

EVM, but you also want to build as an L1. So talk to me about that design design decision. Why L1 versus like an ETH L2? Yeah, mon is an L1 too. So I think it's I think the examples you give maybe like the the common thread is that they're like vanilla EVM execution as opposed to sort of expanding the sort of execution logic. Would that be a fair summary of the question? Yeah, that's a that's a fair summary. I mean, I think this is also a challenge of the L2s in general as well is like the backward compatibility which is obviously that you you want to kind of build on this uh you know ecosystem of developers uh who are used to building in the EVM but also you want to make some changes so you can increase performance because there are some real limitations around the EVM. So

um yeah um it's a good question. I I think yeah there are many sides there there are many answers to this like one of them is a practical one it's just that like you know initially the priority was just to get something that users wanted at all and the easiest way was to not worry about an EVM but I think at this point we have the luxury to kind of zoom out a bit and say like what what is the correct thing to do given that all of finance sort of should happen on one platform and I think I think even even sort of retrospectively if we could have changed anything I don't I don't I would have um I think I think the EVM is great. I think it's really good for like maybe a good analogy I think is

like the EVM is the EVM is to like all of finance as AMMs are to like all of trading. So it's really good to cover this like super long tail. I think people underestimate like just how long the tail is of like things that can be done in finance and a similar like AMM cover super long tail liquidity where like if there only like five people in the world who want to trade some asset like an AMM is like probably the best way to sort of coordinate that trading. Um so I don't I don't think it's good or bad. it just like has its use cases and I think if you want if you want a platform that can house all of finance it cannot just be built on on an EVM no matter how performant the EVM is uh and

maybe like one example which you kind of gave we we gave this example like way back just when we were like even starting hyperl because there were a bunch of questions like why isn't it just built on an EVM um and the example is just like if you have a perex and uh with with perplex it's just the the number one concern is risk for the system and the hyperlquid has I can like write down these like just like mathematically prove to you that like hyperlquid is solvent like the purpose every purpose market is solvent at every block and of course there can be bugs in implementation etc but I mean like theoretically there's there's like there's no way for the system to have bad debt assuming you implemented everything correctly and um but I would

not be able to do this for an EVM base per deck because there's no there's no execution guarantees like you can you can never uh with an EVM it's always there's always some sort of like auction system or otherwise to determine which transactions go through whereas uh on a custom execution environment you can say well like at you know when this block happens like we need we need to do these things to make sure the the per the per balances like kind of are all positive or whatever condition you want like um positive with unbound or whatever like you you can basically like you can basically say like this logic needs to execute on every block no matter what user they're submitting. Um I think that's like super important for derivatives. So yeah I

think like longtail spot trading um no no need if if it ain't broke don't fix it but um but yeah the when the ambitions are bigger I think you kind of need to look beyond and kind of you have to question everything and say like is this really the best way to do things and very cool. Um, all right. I want to uh change ts a little bit and um uh describe how you're thinking about handling me uh in this in the system and you know I thought it was really interesting actually to sort of uh you know bring stake there actually was an old paper I'm going to blank on the name I'll find it and include it in the show notes. Um uh Dave Oha of Osmosis actually wrote about the idea a couple of years ago about um because basically what fees are especially on an L1 is uh

spam prevention. Um, and so it's a really interesting idea to basically say, hey, if you're a staker, it also makes sense that you're probably a good actor in the network and we don't need to charge you the same amount of fees, uh, you know, for, you know, DOS attacks. It's super interesting and innovative. Um, one thing that we've also seen as well, and I know something that you think a lot about is prioritizing, um, oracle updates, um, in general. And so, you know, I'm curious when you think about this world of MEV, which has been really really thorny at least in um, environments like Ethereum um, Salana as well. That's kind of one of the trade-offs in uh you know an L1 which is truly permissionless versus a more controlled environment um like L2s and you you're starting to see um I don't know like the optimism versus

arbitum approach is differing quite a bit but both of them are basically doing things that you could never do on the L1 right so I'm curious how you think about the design around MEV within your ecosystem and maybe talk a little bit about Oracle updates specifically as an area that you're focusing on. Yeah. Um, a lot to unpack there. Um, where to start? I think I've never I always answer I always ask three questions in one. It's a horrible habit of uh doing podcasts for years. Yeah. Yeah. Yeah.

Um, I I think the I I think ME is it's thorny, but it in some sense it's like it's like it's like a really popular research topic, but it kind of it's like a good problem to have. it means that there's there's kind of like real flow going through the platform. So, um I will say I think like Salana kind of had a good like people give them a lot of [ __ ] for like whatever like not thinking about MV, but I actually think they they did they did a really good job of just like getting a lot of people to want to use the chain and like yeah like let's get people to use like let's build something users want first and then any anything else that comes up is second order and I think MV is kind of like one of the second order things. I'm not I'm

not trivializing the problem, but I actually think it's solvable. Um I I don't remember who I I I'm als I I saw I saw some tweets recently. Um you can maybe add them as notes later as well, but the the sort of like you know like multiple proposers uh sort of like committing hashes like like basically like block winners committing hashes of um transactions sent by multiple proposers and then like dduplicating them and then just sorting them by by priority. I think I think Dan Robinson also like talks a lot about sort of just priorities kind of all you need and and I I think like these ideas together kind of solve the one block me problem where it's like malicious proposer can extract value. I will say I think going to like the hyperlquid part

of the question I I do think me is like a bit I think it's it's a bit pigeonhold when you think about it as like purely um within a single block. I think the problem is bigger. Uh and the the analogy here is like if you're trading, let's say you're trading on a centralized exchange and some market making firm has a microwave from Chicago to Tokyo and maybe not on a microwave but like a microwave and then some like underwater fibers and stuff and can like you know can basically like pick off your orders like and then let's let's say the central exchange was like operating actually on a blockchain and had like 20 blocks per second like you don't actually I don't think as a trader you care whether it's like, oh, the MEV

came from um the block proposer seeing my transaction and reordering uh some taker in front of it or if it came from this training firm having like super fast connectivity and being able to pick you off. I think like I think a market maker is basically like a user and like that user gets screwed no m in either case and from their perspective like their a their reaction is the same is to like widen their quotes and that ultimately hurts the retail traders, right? So um from my perspective me sort of me is just like a subset of some broader problem which is like toxic flow and we think a lot about this. Um, so Hyperlid was like the first I I want to say first definitely the first crypto thing to do this as far

as I know and then probably the first Okay, I'm not I'm not going to. So I I I think one of the first v large venues to gain traction to basically do this thing where um takers are kind of treated symmetrically across the board which is that their flow is dep prioritized and um the net effect of this is one is like less volume overall but two is like the volume is is sort of better. It's like both sides are kind of always happy even if they let's say you gave every everyone every actor like one second to sort of think about whether they actually wanted to make that trade. I think like you want that percentage to be as high as possible and

um I think with this change it goes up dramatically because otherwise a ton of your volume is like a toxic taker picking off a market maker and just ultimately like retail eating the cost. So um that's an example of like how I think like MEV should be handled which is a little bit more agnostic to the dynamics of the ME but more like the user perspective of like you know don't get screwed by the protocol. Jeff, this is super interesting and for those who are a little bit less in the weeds of market microsstructure. To summarize what you're saying here, I think it's this has always been really challenging for market makers to work on chain, right? Because if you post quotes, they become stale very quickly. And if you have very sophisticated takers, there is a relatively large uh marketmaking firm who's recently moved on to Salana who is

doing this, which you know shall remain nameless, but uh basically they're acting as a taker. They're picking off stale quotes from market makers and so MM then have to quote wider, right? sort of widens out bit ask spreads and retail ends up paying more for their execution costs. So I've got a lot of questions for you on that. Can we rewind and I actually want to you started to get into something really interesting in the handling of MEB that I wanted to talk to you about. So you referenced Dan Dan Robinson's priority is all you need. So this is relatively the approach that's being taken on optimism. Uh, so they have something like an EIP1559 base fee and then they just sort transactions in order of who's paying the highest priority fee. To my understanding, that's something that's really difficult that you couldn't necessarily do on ETH L1 today because

each proposer is random, right? So you'd need like a very um, you know, very fat builders right in the middle of that or a relay system or something like that to enforce it. But even so, from my understanding, it's not super easy to try. what I heard you just describe actually is sort of a and this was a a question that I had for you around um you know your thoughts on multiple concurrent proposers and what's going on on Salana right because the Salana approach to multiple concurrent proposers is this flies in the face of what we know about colllocating right which is hey if we're going to build a centralized box you want to locate as close as you can to that box in Tokyo or Frankfurt or whatever uh and that's the end of it but the my understanding is the ResNick totally vision around this

is that if you actually have a bunch of different proposals all over the world then that you can actually quote uh or you can you know do blocks faster somewhere around 20 milliseconds I think is their idea you I just heard you unite all of those ideas and I don't understand how you did it so talk to me about how you can bring priority is all you need to an L1 type environment and then if you could layer in your thoughts about multiple concurrent proposers and how viable that is as a um you know consensus strategy. Yeah. So yeah, I don't want to take credit for that. I I was actually just reading um I was trying trying to just paraphrase what I thought this paper was saying. So maybe maybe I misread it, but I think it's like the Salana I think the Anza folks and yeah, Resnick and

Anatolia. Yeah. Yeah, I think maybe them as well. Um I really should uh you know learn the sources, but um it was just this paper and it was I think it was it made a lot of sense to me as so I'm actually not an ME researcher. um we we kind of just think about things from first principles and build stuff and often people later on come and like give some fancy names to something that we did and like you know maybe also like someone else was thinking about the same thing in parallel and we just like didn't didn't know um but um so I I might butcher some of the like so like technical terminology here but um but basically I don't think I I think in in a world where networks are super fast you don't care about like additional

hops and like sort of additional time to like basically between proposing and executing a block. Um I I don't really see any downside to having multiple proposers just kind of like you know all commit transactions and then the kind like unveil them after the sorry so the different builders propose transactions and there can be duplicates and then the the proposer basically bundling these and then the the builder is kind of unveiling them after the fact and maybe there there's some like slashing if there's malicious stuff like it all seems kind of like very solvable to me though. Granted, I haven't thought about it too much.

Um, and yeah, I think this this adds like latency because you need to listen to multiple builders as a proposer before you hit some threshold where like you're okay building the block. Otherwise, if that's not allowed, then you can just take one of them and then all the MV is back. Um but yeah, I mean other than these sort of like these these what feel like constant factor latency increases uh and maybe like probably increases the sort of networking overhead of the network as a whole. But again, I think these are all just like like hardware just scales super well. So I feel like this should just happen. And uh yeah, and in my mind it feels like yeah, then you just once

you have all these once you have different sources of transactions in a single block and you just order them by priority, there's there's like no way for like the game theory just completely changes and there's there's like nothing you can really do to game it. Um that being said, I think like basically another one I points is that I think just having like really fast blocks also kind of solves the problem in two different ways. like it makes like the amount of me in a block scales with the time between blocks because um like like imagine you just had like one block a day or something then like ETH's price probably has moved like you know plus minus 10% then like obviously if you're the one building that block you have like so much wiggle room to like pick people off but if blocks are

like one millisecond for example then it doesn't really matter like how dishonest can you be in a block like all you can do is reorder stuff but like nothing happened in that millisecond so like you can't really mess with it so I think those are like the two limits you think about and there's like clearly a continuum there. And so in some sense the like yeah the the faster you're willing to make blocks the less of a problem there is to begin with and then but that's intentioned with the fact that the faster you make blocks the harder it is to do all these like fancy things of like multiple proposers etc or multiple people proposing inputs to give to the proposer. I don't know what that's called.

Um, so I think that both Yeah, there's probably just like some there's basically like some threedimensional surface you're thinking about here of or like threedimensional volume of like uh and then and then the axes are like how fast blocks or like how geographically dispersed or like how much like network tolerance you have and um the amount of like fancy stuff you're willing to do and like there's like a lot of pretty optimal points uh you can you can pick. Uh so yeah, I think MV will be a solved problem and I think I think protocols really should like focus on getting users first and I think um but on the other hand I like really appreciate all these researchers doing this work. It's it's really interesting technically and I think uh it's really cool to see how far the space has come since Ethereum

which I my understanding is like there was not much thought given to me at all just like when Ethereum was built because they had so much [ __ ] to figure out and so um it's just it's really cool. I think both are super important. think um probably prediction would be like in the next few years. It's people look back and say like it's funny that we thought this would like kill crypto, but it's just like just one of those technical challenges. Hey everyone, quick shout out to today's sponsor, Marinade. Since launching in 2021, Marinade has expanded their suite of products to provide solutions for both DeFi users and Tradfi, including liquid and native staking, as well as direct enterprise integrations. Marinad's platform is optimized for institutions to securely expand their exposure to Salana. Whether through the provisioning

of custodial or non-custodial yield, integrating staking rewards into Salana financial products, or leveraging Marinade's bestin market rewards rate for revenue and treasury management. Whether you're in DeFi, Tradfi, or somewhere in between, Marinade has something for you. To learn more about Marinade, follow the link in today's show notes. I guess one followup from like hearing about um like multiple concurrent proposers and just the the challenges of like um you know maybe trying to reduce latency as much as you can. Uh in a sense it feels like uh it might never be possible to match the sort of like speed of a centralized exchange that's just everything's kind of sitting in one one company's server.

And uh I think in a way that's like kind of who who Hyperlid is really competing against right now. that's u kind of more of like a small fish in a in a very big pond with all of the sexes. So, um, I'd be curious like kind of, uh, how how you view Hyperlid's edge against some of these other centralized players where maybe you'll never be able to compete just on pure speed, but um, kind of what what else u is going to make the trading experience on Hyperlid competitive or maybe even, you know, much better uh, than than you could get on a centralized exchange despite being a little bit slower in a way.

Yeah, I think the slowness I don't I don't think users think like a 10 millisecond response time is 10x worse than a 1 millisecond response time. So I don't actually think that's going to be like the ultimate heal. I just feel like dexes will dexes and blockchains will get fast enough and users like that will not be a differentiation point between centralized decentralized. I think what will matter is that with the centralized exchange, you're you're really trusting like this one company and you're not there's an obvious sense you're trusting them which is that you're trusting them with your money which which is a super big sense in which you're trusting them but you're also trusting them to like kind of continue to innovate and um you know I think I think they're good

but like I think there's no way a centralized exchange can compete with like a globally distributed community of builders and there's there's like two aspects. There's like the cooperation of just like like you look at hyperlquid there's we we do like zero we we have zero idea how to market we we just we do not have a marketing department like we it's just like completely organic and community run. I think that community does an amazing job like better than the marketing departments of like all these centralized exchanges that like that's that that's like what I attribute like Hyperlid's amazing success to is like all these users kind of coming together around something that they believe in and really like organically sort of shilling it from the bottom of their

hearts not from like uh not from like a K paycheck right so um and and it's still so early right so like if you imagine that going from like a niche crypto Twitter thing to like something much more global you you like cannot compete with that as as a central entity and um obviously there there's a lot of difficulties with like doing something in a decentralized way and I think that's that's why centralized exchanges have been so successful so far is that you get to basically cut a lot cut a lot of these corners or or I guess you just kind of like get to not think about a lot of these problems like you know me is one of them but like other things just like there's just like a lot of unsolved technical problems with like

how do How do you coordinate these validators like do things that um still allow for a seamless trading experience, right? And uh but I think all these can be solved. It's also like it's it's like a fun thing to solve. So I think like smart people naturally are attracted to hard problems and we'll solve them all. And um so yeah, I I would definitely bet on the global financial system being on a decentralized platform like owned by the people. Um, yeah, just like with all the there a lot of really good points like the ethos, the original ethos of of DeFi.

Jeeoff, I want maybe we can transition into, you know, some like ecosystem design. Um, and you know, one question like a theme that Monae and I are uh kind of poking at throughout this entire season is um DeFi architectures which lean into more modular design. So we've been talking about this within the context of something like Uni V4 where instead of designing you know like this is the V3 instantiation of the protocol you know you are trying to attract a whole bunch of different AMM designers right within your kind of uh they've got a little bit of a vertically integrated structure at this point or uh something like Morpho instead of one you know instantiation of a borrow lend uh pool you actually have a bunch of different um uh you know isolated borrow lend pools and I think that's what I kind of

just heard you say that a core advantage of DeFi is that instead of relying on, you know, obviously a very competent uh team that you've built, but you're actually going to leverage the creativity of everyone to build on top of uh Hyperlid in general. So, can you kind of paint a little bit of a a picture for that? What do you think about um the idea of kind of building everything h inhouse in one team versus um you know, kind of trying to attract uh other people to build things that you guys would never really thought of? How do you think about channeling that creative energy within the Hyperlid ecosystem?

Yeah, our hyperlook has evolved a lot over the two years in which it's been growing. So, initially it was like you said one team doing something is super efficient. So, like let's just let's just iterate and like just build something that users love. Um and at this point it's completely the opposite which is that our thinking as the core team building sort of maintaining the protocol uh sort of de working on like protocol level stuff is that if something can be built by someone else it should be built by someone else and we take this very seriously because we're a super super small team so we're actually only 10 people um and that

includes technical and non-technical um and we just don't have capacity to like do almost anything and and I think that's like a really that's actually like a blessing in disguise. Um basically if if a protool if a protocol can be set up in a way that it's it's like well architected so that people can come in and actually express what they want to express and plug into a permissionless protocol and do it. That's that's very powerful. basically they get to they get to leverage everything that exists underneath and basically just focus on their component of it.

Um I think the the flip side here is um like you can make this mistake by by building a protocol too early and this has happened a lot in crypto as well. I think people get really excited and they see there are these web two examples like platform businesses are the biggest ones, right? So um you really need you you need a critical mass of users and you need to really users and liquidity I guess in the case of DeFi and you need to really abstract the you need to abstract away exactly what the builder needs and that it's it's hard to know what that is until I guess until you have users and sort of like genuine demand. It's a bit

of like the chicken and egg problem of markets in general. So, hyperl I I think hit this like inflection point maybe half a year to a year ago and uh I think there there is just like this critical mass of people who want to do things on it and that's like really cool as a builder because um I hear this time and again it's like it's like builders kind of maybe they they were doing stuff on other chains general purpose and like come hyperl like wow it's like all these people just it's just it's like oh like you know alpha alpha alpha like pre-alpha version of um my protocol like launching and it's like and one hour it's like the the deposit cap I hit is I I said is hit and it's it's like all this kind of there's just like so much latent demand to like do things

and like people know what the problems are. So also it's not like people are coming in and like trying to create a problem before solving it. It's just like you come to Hyperlquid and there's like all these things left open like portfolio margin like why can't I why can't I um borrow against this asset and trade per well like it's because a user could build it build it and like it's you know there's like other stuff happening at the protocol level that's um more base layer and so like why why why shouldn't a user build this and like teams coming in and like trying to build that similar many examples across the board. Jeff, what do you think about uh you know, one thing that I find really interesting about Hyperlid is Hyperlid is one of very very few protocols who've built um with this sequence, which is

mostly what you see is protocols kind of build as a general purpose uh type platform. If you look at something like Ethereum, right, the idea is this is incredibly neutral. Come build whatever you want here. Um and that worked when there weren't a lot of options. So kind of Ethereum and Salana both took that approach. What you're seeing from newer L1s is to say, "Hey, we can't do that anymore. There needs to be some sort of we need to solve the cold start problem of there being nothing to do with for users once we launch this new competing chain." And so, you're starting to see chains like Barra Chain um launch with a native stable coin and a native DEX basically so that users have something to do, which was a little bit of a lesson learned from Cosmos maybe. Um, so I'm kind of very curious what you think

about this idea of first-party applications versus purely thirdparty applications and what the right mix is. Because if I had to maybe poke a little bit at the hyperlquid strategy here is let's say I'm an app developer and I'm like hey this is amazing right? I'm going to this uh ecosystem where there's already a huge deck and tons of users and volume etc. But how can I be sure that um you know hyperlid won't roll out what I want to build as a feature right because they have so much power in the ecosystem already. So, how do you kind of thread that needle in between, okay, I've got the users and the liquidity, but maybe competing a little bit with application developers who you'd want to build in your ecosystem?

Yeah, it's a good question. Uh, so I'm not super familiar with the with the bear chain roll out, but I assume when you say native decks, you mean like they built the DEX. They built Yeah, they built the deck. Yep. Yeah. So, that's coined. Yeah. Yeah. Yeah. Yeah. But but that's like all sort of like EVM applications, right? like they the team deployed some smart contracts and kind of enshrined them in that way. Um yeah, so Hyperloop is different there. Um the the native side is not like it's not like the team decided to one day like just put on their application developer hat and like build some applications. It's more like I think that like what people call native applications on Hyperlid are in my mind more like infrastructure. So I think what Hyperloop liquid does is it

basically I view it as a general purpose smart uh sort of like smart contract platform from the perspective of a builder it's just that you have all these native things to interact with like just like on Ethereum there's a pre-ompile to let's say like compute a hash right like it's a very expensive function if you wrote if you wrote this hash function in solidity it would take a lot of gas and like you would never use it um but the protocol maintainers were like you know what like computing this hash is important enough. Let's just uh let's just like build a build a pre-ompile to let people call it and cheaply compute it. Um, so I think the line between like like no one says the Ethereum developers like are, you know, building applications like you know what if what if you wrote a really good hash

function in Solidity. It's like no one's like complaining about that, right? It's just like there's some things that like are very clearly infrastructure and like good for builders. And I just think Hyperlid redefineses that or like pushes that boundary a bit to where I think it's like more correct if you're really focusing on DeFi. And so like in my mind like a pers like a very liquid orderbook pers exchange where you can anyone can just like plug in and it basically with like almost zero slippage um no long short majors. I think that's just like a public good and there's no reason like that's not something that's not that's that's just like so important as a building block that it should be almost like a hash function as like a pre-ompile in in

solidity. Um, and like I don't think I don't think it's that bad that like a user can't come and build some like AMM version of this like BTC per and be like man like man like you know this I'm competing with the chain. It's just like that's just like not something that user should be building. Users should be like leveraging these like core infrastructure things like build actually interesting things for for end users. And so I I I don't think there's this like competitive element. And I I think we've made I think I think the the trajectory has been good. Like basically the stuff that sort of like hyper good labs works on these days are all trending to be basically like protocolized things, right? Like permissionless per um like

permissionless spot deployments came a lot earlier. But it's all these like it's like things to enable other people to do things. So I'm like, yeah, like someone else could build a protocol on the EVM of course, but I think that you kind of have to draw the line somewhere. And so the thinking is always like what is best for builders and that is I I think that's like very closely aligned to like what's best for the protocol. Um going back to my previous points just like you have a glob you have a global set of like talented developers and like you should just you should just enable them. And maybe one one more thing I should add here is like I think one thing Hyperl does super differently from other protocols is like we actually like we in designing how the protocol works like a really high priority is that

builders should be able to monetize uh basically like participate in upside in a substantial way on on what they do. So like spot the players for example can keep up to 100% of the fees collected in the token in the base token which is to say like 50% of the the total fees on their pair and like that's something that no exchange even if it's a central exchange like they they will never offer um a team that right so um that encourages like really high quality innovative assets to deploy natively on hyperlquid um per same story obviously not live yet but like same story there and I think this this theme builder codes like if you build a UI you capture like more fees per trade than the protocol captures. Like these are these

are all I think things that do move the needle on like building trust with builders and sort of like just making it clear that the protocol is here to support you and to basic like we we want you to build a billion dollar business. We don't want to we're not trying to get in the way. We're trying to like make that possible. This episode is sponsored by Catria. Catria is the first zero knowledge rollup to enhance the capabilities of Bitcoin block space, enabling Bitcoin applications and unlocking Bitcoin's full potential as a programmable financial platform. With Catria, developers can now tap into Bitcoin's liquidity to build scalable payment systems, privacy solutions, and more. Citria is currently in its test net phase and is approaching its mainet launch. To learn more about Citria,

follow the link in today's show notes. I'd love to ask like a followup on that. I think it's super interesting how the Hyper EVM is obviously like a sort of just a very primitive like base layer protocol designed for people to build on top and it's uh been really cool to see how hyper core features like the spot uh markets but then also now uh with builder deployed per it's kind of moving towards more of like a platform uh business take on it. Um I guess I'd be curious like how do you view for builder deployed per specifically? Do you think that there's like any sort of competitive pressure maybe implicitly between uh markets that are coming from from new builders making them versus uh

from the sort of like hyperlquid core team? And then do you have any ideas about how how you're thinking of like maintaining that high quality unified coherent UX across not just markets that uh you know we kind of all know are safe because they're u from the the core team but also from some of these like more longer tail builder deployed markets. Yeah, good question. Uh so the intention is not really to have a comp so it is it is a it is a permissionless protocol so people can deploy purpose on sort of like what they see fit and it is the the protocol is permissionless. So um I kind of can't I can't exactly predict what

will happen. Um that being said, the the motivation behind it is that basically crypto pers are like such there there there actually like so many innovations that go into you know hyperl per you know we're standing on the shoulders of giants. I mean like per themselves are like amazing innovation and I'm sure there are like a lot of other things like you know but it is but it is super like narrow and I think the time has kind of come where um I think it should be up to builders to come in and say how can we apply this primitive this like infrastructure I view purpose as kind of

like an an infrastructure element like what what can we build with this infrastructure so that's kind of the motivation I don't I don't expect people to just put up like the exact same markets as validator operated per say like hey like let's just try to compete it just doesn't really make sense. Um I think I think there there are like various axes on which people can innovate but um the the ultimate goal is for all of finance to be on hyperlquid. I really think it's like it's transparent. It's 24/7. It's low latency. It's it's audit auditable. It's just it has so many things going for it that it really is like I should it

should be like a 10x upgrade on like legacy financial systems. I think there's a long way to get there, but I think these builder deployed per step in that direction. I I want to um I know we're a little uh I we got a little bit of time left here and uh we got so many more questions still to to ask you Jeeoff. Um one one theme that we're uh very interested in this season is this idea of vertical integration. And when you look at a lot of the dexes in Ethere, and um crypto today, uh you're starting to see a little bit of that, right? Which is like unis swap is a really good example of a protocol that started at the AMM layer and they're kind of going up and down the stack, right, towards the distribution layer of having their own, you know, UI uh like desktop UI and then

their own wallet and then down the stack to their own chain. You can look at like the recent uh competition between Pump Fun and Radium in Salana is another good example. Jupiter is the same thing. And so, you know, I think especially like UI is super super critical when it comes to exchanges, which is maybe what is the the sort of driving force here. But, you know, I'm curious on, you know, on your thoughts in general on the idea of vertical integration when it comes to uh dexes, whether or not, you know, how Hyperlid thinks about that, what areas of the stack are the most important for you to have, you know, direct control over. And then, you know, you mentioned at the beginning of this episode that or actually I heard you in sorry, ZeroX research episode that I was listening to before this, you know, there's so many

things that you need to get right. You need to design the right environment for traders, for market makers, but you also have to get the UI stuff right, which has been a real challenge in the per space so far obviously until um you know you guys have done a really good job on moving the ball forward there, but talk to me about how you think about vertical integration in general and then UI right like what is what is important to get right from a trader perspective here. Yeah, I think the so I think vertical integration is is obviously super good as like a business decision. Like if if you're just like a traditional company, you want to own as much of the stack as possible. Um, I think it's like part of that is just like it's it's often unclear where the value occurs, but

another part of that is just reducing the overhead of if you have like many people you you need to rely on to uphold their part of the deal, then you're kind of your ultimate user experience is only as good as the weakest link in that chain. So, and I think that's also true in crypto, but I think there's a bit of there's a bit more tension in crypto, which is that sort of owning the full stack is also a bit antithetical to I think the like what DeFi should be.

Um like I think DeFi is more resilient if there are multiple possible choices at each layer of the stack. And so if any one of them, you know, is compromised or just goes down or like whatever like loses starts going in the wrong direction and like stops prioritizing the right things that that that like ultimately the you can find some path through these um like different options at each step, you can like find some path that's like still pretty good and that like creates a much more robust financial system. So it kind of depends what you're optimizing for. Um from hyperlquist perspective, I don't I don't think it's like a huge basically

like we we we never even seriously entertain the idea of building everything in finance. And so um I think when the like when the vision is that big it it's nec ne like necessarily built by many people and and I'm very I'm very bullish personally on like coordination through protocols not through like management like you have like a 10,000 person company you probably build all all the finance like everyone's super good and coordinated and whatever but like I I just feel like that's that's a I just don't see that happening in the real world. And so like the next best thing is like 10,000 people coordinating across you know a thousand companies like each building

their own pieces and like um and a thoughtfully designed protocol to coordinate those efforts and I think that's that's where I see Hyperlid being and yeah yeah I think it's just like the the end state is so massively impactful and just like there's just like so it's just like such a valuable enterprise as a whole like finance that I don't think it matters if like hyperl doesn't need to capture all of it. That's like a super greedy perspective. I think um I think if you're like if you're like yeah I'm I'm doing this like one super small segment then yeah maybe you actually care about squeezing out every last bit of profit for your shareholders or whatever. But um when you're building a protocol to

power all the finance, I think the the perspective is more like um it's just like like let's just like grow the pie and um you know if if this thing if this thing is to succeed like everyone's going to be like so happy that um yeah doesn't matter. There's maybe one area of like um sort of vertical integration that we've seen across a lot of exchanges that I'd be curious for your take on which is um we've seen a lot of like centralized exchanges try to capture some of the economics from the stable coin balances that are held on exchange.

So Coinbase has this uh very lucrative distribution uh agreement with Circle where they get a lot of the uh the carry. Um, Binance several different times has tried to either partner or kind of like white label their own stable coin to capture some of that uh huge amount of of like yield that's just latently um being generated on all the the stables that are on their exchange. And I think that's a really interesting angle because it's taking money that's completely outside of it's not like taking from builders or users. It's just taking money from basically the US government and injecting it into your ecosystem. So, um I'm curious if like Hyperlid has any uh if you have any

thoughts about, you know, how to capture some of this uh economic juice from all the USDC on chain uh on Hyperlid, whether that's, you know, agreements with Circle or with Coinbase or, you know, maybe even having another kind of white labelled stable coin or something like that. Yeah, it's not currently something we're working on. Uh I I think your question sort of highlights the point that there's just so much lowhanging fruit that like it's kind of hard to believe but you know there yeah there are like billions of dollars like of stables sitting on Hyperlid but up until now the priority has not been that like that has actually not

been the low lowh hanging fruit like um yeah that's maybe like more than hund00 million of yield at this point But also the the protocol makes annualized like what like five to 10x that you know I don't know the exact details like not enough data points exactly to extrapolate a year but um but I think you get my point which is like just there's just like so many things to do and um yeah I think I think it'll be cool eventually for um the protocol slash users to be able to have all all sorts of benefits And I think this is what kind of what happens when all of finance moves onto one chain is you get to kind of

um if if that chain is designed thoughtfully and not greedily like not not as a sort of company company maximizing extraction but rather um a company maximizing just cool [ __ ] that can be done then uh yeah I think I think all this stuff becomes possible. Okay, Jeff, uh, we're nearing the end of our time here, so maybe closing question. Um, and building on the question that Monae just asked, I'm curious what you think about the role of stable coins versus, uh, the role of Hype the token in your ecosystem design.

Um and specifically I think I uh you know Hyperliquid is a really interesting example here where you know even if you look at something like Ethereum like the way that they thought of value cool for ETH the asset was that it was going to be used as this store value or medium of exchange kind of money and in the early days you saw a lot of uh even still today right if you trade NFTs it's typically NFTs are denominated in something like Ethereum but a lot of the early use cases for something like unis swap is the pair was ETH uh right in a lot of these AMMs like B1 um and gradually that's kind of moved moved over time uh or shifted over time towards more dollar denominated uh trading pairs on L2s especially as they kind of try to go out and get new

customers from the world who mostly hold uh dollars instead of something like ETH and so it's interesting you know Hyperlid maybe developed in reverse order than a lot of exchanges on on place like Ethereum or Salana right where you started from day one with USDC uh and dollars at the core right because that's what your customers wanted but now you also have a token hype Right. And it seems like the way that you think about value acrruel there is buybacks. But I'm just curious how you think about this like have you ever thought about like making hype kind of more of a a trading pair or a core part of the ecosystem or how do you yeah kind of balance this desire for customers to trade with uh dollar basically dollars versus value acrruel for hype.

Yeah, I think the money thesis of a native asset is probably not something you can force. So I don't I don't think you as a team you can go and say you know what like I think I think I'm going to build money and like it's just like but money is like one of the most elusive things in the world. It's it's humanity's I think like one of greatest inventions. is like it's like the greatest coordination invention and I think like the source of a lot of human progress but also it's not something you can directly build.

It's you can kind of you you can build it kind of by building a country like the US had did a super good job right like the US like the dollar is a super good product. Um or you can build it like Bitcoin like like Satoshi kind of did which is let's just build some interesting tech and this technology is like so groundbreaking that um you have a chance to create something that's that's like money but um yeah otherwise I don't I don't know. I mean I think I think hype will be money if all of finance moves on the hyperlquid. Um but but yeah, we're kind of focused on the ladder because that's actionable and uh um so I think that like trading

against hype like I don't know I'm not I think people talk about it a lot but I I think genuinely what like people think in dollars in crypto at least um and but I think broadly around the world like people kind of pre prefer to denominate in dollars and I don't think that's some like if you say like if you come in this platform you're trading in hype then I think probably what happens is people stop going to the platform. So, um it's not and and I think it's like ultimately it's kind of what people want. So, I I think the sort of like technologically they're both assets on this decentralized ledger and I guess time will tell.

All right. Uh Jeff, we could go for another hour here. Uh but you've already been super generous with your time and I know we got to wind it down. This was great, man. Thank you for coming on the podcast. If folks want to find out more about um Hyperlquid or follow you or just uh keep updated uh with the work that you're doing, what's the best way to do that? Um yeah, there's a Hyperlquid Twitter you can follow, Hyperlquidx and I'm at Chameleon_JF. Okay, this is the real alpha. Why Chameleon Jeff? I love to ask people with interesting Twitter handles. What's the uh what's the alpha behind that? Um I think they're really cool animals. So, um I don't I don't know how well known this is, uh but they're they're like they're just like so interesting.

They're um you know, like they're one of the only animals where the the eyes can point in different directions, um like independently controlled. And they're one of I think they're one of the only lizards that have like two claws forward and three claws back, which shows like a very interesting like evolutionary trajectory. And they also have like a really powerful tongue which is like pretty cool. I Yeah, it's just like and they change colors obviously. I mean that's the that's the well-known thing. So, um yeah, I think just like they're I've always been really fascinated by them. They're kind of like aliens on Earth. So, um yeah, that's that's it.

Nothing more than just respect for the animal. So, listeners of Belro, this is why you got to stick around to the end of the episode always because uh you thought you were getting an interview on Hyperlid here, but you're also going to learn about Chameleon. So that's why you guys stick around. Jeff, this has been really fun, man. Thank you so much for coming on the show. We'll have to do it again sometime soon. Yeah, thanks for having me. It was a fun conversation. Cheers. Thank you. All right, Monae, what an interview, man. Um I can see why everyone's so hype on Hyper Liquid. That was great. Yeah. No, that was that was fun. Definitely got a little more bowled up after uh hearing Jeff. Same. Uh yeah, I mean I thought, you know, it was interesting. We kind of went into this interview and I was very

focused on at least kind of vertical integration and not thinking about some of the other themes of the season, but it felt like we ended up talking a lot about this kind of idea of modular design and fragmentation. And seems like Jeff is highly focused on leveraging not just the creativity of uh his team which has done very well um executing on the vision thus far but also leveraging the collective uh creativity and builder and entrepreneurial talent um of crypto. And you know, it's kind of interesting. They've they've approached Hyperlquid from the exact opposite direction of more general purpose platforms like Ethereum or Salana. And they've kind of started with the monetization and the product which generates the fees and now they're trying to bootstrap the ecosystem around it as opposed to some of the earlier smart contract platform

which really started with the ecosystem and the community and then encourage them to build a bunch of ideas. And so I'm curious when you think about that approach, how do you weigh the the pros and cons of the way that um Hyperliquid is going about it versus a more general smart contracting platform like Salana or Ethereum? Yeah, I mean I I think you mentioned in the show like uh that's Ethereum and and Salana to another extent like they had the luxury of uh being early. So um you know they they could just kind of build it and the builders will come and and you know build on them because there weren't you know a wealth of other opportunities but uh I think that in a way hyperlid they kind of they were very smart because uh they recognized you can't just launch an

ecosystem or a platform from day one. there needs to be a hook to acquire users and um there you know per uh markets I think have been been that hook. Um and it also gives the platform like some really solid economics to uh to get people excited about um from the start. Um, but I I think it's also it's going to be very interesting because it feels they're still quite early in um sort of transitioning from uh really a product focus just having the best per exchange that you can in DeFi to more of a platform focus uh being somewhere that builders can come and and you know make

their vision happen. Um, so I think it's yeah, there's still seems like uh Jeff was like really he had a an eye to all of these challenges that you you have the tensions between those, but uh it's still very early in that process. So it you know it's tough to see if um they've they've run into these these challenges yet. Yeah, I I think that Hyperlid actually was way ahead of its time um in figuring this and that's actually the trend that I think you see in not only new mostly new L2s in general uh but also new L1's that are launching. They're saying hey um you know it's a completely different market structure than it used to be when Ethereum and Salana launched. And that's everyone from Barachchain launching with

their own decks to Honey, their own native stable coin. Um that's new platforms like uh the megas of the world where it's not just like hey everyone they're very uh concentrated accelerator programs where they're maybe kingmaking is too strong of a word but they're really putting their resources behind a couple of very high quality teams and so you're seeing different iterations on this and I I think this idea of first versus third-party apps is going to be a huge uh a huge topic when it comes to ecosystem design in general and I think that Hyperlid basically absolutely nailed the meta and they were way ahead of their time. Um then the I think the rub is it's ultimately going to come down to that question that we were kind

of asking and I thought you had a very interesting question to ask which is um one there's hyperlquid's probably going to have to reassure its uh its compliments um that it's not going to try to compete with them right so okay there's hundreds of millions of dollars of fees which are being generated from stable coins like is Hyperlid going to try to take those or is it going to let someone else build um and so I think maybe it has to demonstrate uh that it's willing to let builders come in and take money on the floor. Um, you know, this just kind of sitting out there on the floor. And then also, I thought you had a really interesting question around how do you ensure the standards of third party app developers, right? This was actually a challenge for the app store in the early days and why Apple, you

know, it'd be interesting to go back and look at Apple's first party versus third party app strategy because in the beginning it was a little bit more permissionless and then it's like whoa uh there are a lot of apps in here which are not creating a great user experience and you know I'm getting complaints from users that when I download this third party app that I'm getting scammed or it's a bad UX and so they got much more stringent around standards. Um, and so it'll be interesting to see whether or not uh, Hyperliquid goes from a more, hey, anyone can sort of build here to a more prescriptive approach in terms of basically just generating a good uh, unified user experience. Yeah, I think it's it's really interesting you mentioned that because I think they're almost they're like kind of doing both of those at once, like two different

strategies. So the Hyper EVM fully permissionless to deploy, at least for now. Maybe they'll change that in the future, but um it's just permissionless. And then for their builder deployed per which are still on test net, but uh expected like soon I think. Um there's like this big um staking requirements. So you're putting up some capital to secure the market. Um and then you also have to like bid uh to to essentially like have the right to make a market. Um, so I think it's it's interesting that they're kind of doing a more curated approach for some of these like hyper core features like the per markets. um but

then also doing the permissionless and um I think there's still going to be some challenges that you see with this platform model even with the more curated ones like uh I know Jeff said like okay I don't expect uh people to be launching you know another Bitcoin per market that competes directly with the ones that already exist um but what if uh you know a builder deploys some new longtail market um that ends up becoming very popular um you know is is that something that uh like the sort of core markets are going to try and compete with? Are there going to be several builders who are all chasing after the same market? Um how do you deal with the sort of like fragmentation uh in liquidity but also

just in kind of mind share and and uh user experience? Um so I think there's there's going to be a lot of uh interesting challenges when the uh when the rubber meets meets the road on this. Yeah, it's going to be fascinating to watch um watch their approach to that. And you know, when we were discussing ecosystem design as well, I think that the question of stable coins is probably the most interesting one. Uh because as you as you pointed out, yeah, USDC actually is a huge source of revenue for Coinbase, you know, through the revshare that they get through the center consortium and circle. And so, um this will be and actually a huge distribution strategy in general for uh stable coins is leveraging an exchange. So, it'll be interesting to see what happens when a

native stable coin comes to Hyperlquid, which I'm sure is probably inevitable, right? Is this something that Hyperlid is going to try to take a rake on in the same way that Coinbase does or whatever? It'll just be interesting to see how that ends up playing out. But the other thing that I thought was uh was interesting was talking about just ecosystem design around hype, the token, and uh dollar assets as a trading pair. And I thought Jeff was I yeah I largely agree with Jeff here and I think that the the division that they have is super interesting. So you know I've had some this has been a core question of the rollup ccentric roadmap in ETH and it's now increasingly coming up in within Salana as well which is what is the role of soul token. Uh but you know the question there being okay

there's a little bit of a tension here because we wanted to ship ETH up to L2s where it was going to be used as a trading pair and that created a lot of the buy pressure for ETH in the early days but L2s are like hey I want to go out and attract new users and they want to hold dollars and trade with dollars as a trading pair and that's obviously how hyperlid started and so their solution to that was well we do have an exchange which is generating a ton of fees so we're going to do buybacks um and so that was kind of like the burn mechanism as a value cool method for ETH in the early days. So, it was interesting to hear um him stratify kind of the the difference in between how he thinks about dollars in the ecosystem versus the role of hype the token. Uh

but I'd be curious what you thought about that and whether or not we can extrapolate that kind of clean division in between, you know, across, you know, like the Ethereums or the Salas or the L2s of the world as well. Yeah, I thought I thought he was dead on about when when we were talking about like, you know, is your token money? Um, and Jeff said, "You can't really you can't build for money. It's just not actionable." So, he was like, "You know what? We're just going to focus on shipping the best experiences, you know, the best um yeah, the best like products and and platform that we can." And I think that that um yeah, it feels like Ethereum maybe has gotten too focused on ETH's money. Like, you know, I own a lot

of this. I hope it becomes money at some point, but it's really not not super actionable. So, I think that was like a really uh refreshing and and um kind of like the right viewpoint to have. Um, one thing that I I think is really quite different from Ether Salana and I'm not sure it's really replicatable from these just more general purpose uh chains is that um Hyperlid has this sort of like native app side of things. They have the hyper core pers where you can get you can stake to get reductions in uh trading fees or you can stake to be able to deploy a perp as a you know the builder deployed per market. Um and you

even need to bid in hype to be able to like sort of like win the the right to deploy a per market. Um and I'm sure there's going to be more of these in the future as well. So they the the native apps um not only it's like another revenue stream to like kind of preund their buybacks, but um they can also kind of attach more interesting utility to the token beyond just staking to secure the network. Um it it's it's almost like they have like restaking built into the core functionality of of like the hyper core platform. So I think that's that's super interesting. And um you know Salana and Ethereum they have a lot of restaking um developments

happening. So potentially some of these other utilities are going to be um it's going to be happening but it'll be from sort of the outside in uh of people sort of adding on these utilities. And it's not really clear to me yet whether ETH is going to be the preferred collateral for restaking or whether it will again be something that gets won by stable coins or or T bills. Yeah, it's really interesting that you bring that up and you know this was maybe a poorly worded question by me in the beginning but I I'd never really questioned this before.

It's like why didn't Hyperlid just why didn't they just ship as an L2? And I think if you were cynical, uh, you might say, "Hey, this is, uh, trying to capture this idea of the L1 premium." But I walked away from this conversation thinking that that was not the case at all. And I I agree with Jeff's take um, which is a simple take in me land. And I think the simple takes, especially when it comes to MV, are the best, which is this is an engineering problem and we can mostly solve this problem. And I would encourage we'll link the the paper that he was referring to which was I believe uh Max Resnik and Anatoli's blog post from the weekend over with uh multiple concurrent proposers and how that could um you know you could have a very distributed but also very low

latency ecosystem and I agree with him. I think you have to make a little bit of trade-offs but you can solve that problem. Um but I came away thinking they were very intentional about the design of Hyperlid as an L1. Um, and they were also very focused on giving users guarantees that you could not get with other centralized exchanges. And it was interesting a couple of times that he phrased having a very liquid per market almost as a public good or utility. It's like utility liquid liquidity as a utility that others can use to build on top of is a super interesting idea and it run I think that makes a lot of sense very pragmatically from a hey I want to build in this ecosystem. What can I build on top of?

there's kind of this idea of a liquidity layer which unis swap wants to be uh you know sky and spark spark wants to be there's a lot of lido is designing this liquidity layer as well everyone kind of wants to be that um but I think kind of using you know a unit of liquidity as a resource for builders to build on and the L1 design underneath it I thought was pretty cool um it's super cool I think it's um maybe it's still not appreciated enough yet though that um like a lot of the core maybe like winners on EVM chains and and all all chains so far it's been these these DeFi apps that are kind of building the liquidity primitives um you know lending markets lending pools like a um&M's like

unis swap and curve um they've been kind of like the core success stories of of DeFi so far and I think they've they've kind of succeeded because there have not been performant order books and there haven't and perform at perk markets which are like kind of users preferred ways of trading and levering up on chain yet. So if these are being provided as a public good um it's like a challenge cuz you can't just copy paste the same models um that have worked on other chains. Maybe you can to some extent, but you're going to have to adapt to the fact that um you know this core app is providing a lot of

the functionality that has been forcing users into these other primitives so far. Um, but then maybe it's a benefit too because if you know people can actually start to think a little bit different and take some more risks, build something new, um, and not just focus on the same the same primitives that we already kind of know work because they have revenue and TVL and such, but like um, they're not as they're not as exciting anymore. They're not really breaking new ground. Yeah, I agree. I I think one thing that we didn't really get to touch on which uh maybe we'll have to have Jeff back on at some point but was the the idea of the intersection of RWAS and um some more uh in like the incumbent competition uh competition right so

obviously you have sexes like Coinbase and Binance within crypto but you also have um you know stock trading platforms like Robin Hood that are moving deeper into crypto as a space and you know I think that obviously uh the exchange layer is super super competitive not only in you know Trady but also in crypto and so you know it' be interesting to think about the the competitive advantage that something a chain like Hyperlid has versus those incumbents and um you know one the last question that we I wish we had gotten to talk about a little bit is the go to market and customer acquisition strategy uh which frankly I would be surprised if you would reveal that on an interview anyway but this is what I've managed to piece together looking at a couple of

exchanges come up and I'm curious to get your take on But it seems like every exchange innovates on how fast they can offer uh tradings or uh different assets um and list them and then eventually it morphs into something else. So the example is like Binance in 2017 they just kind of listed everything right and then they buttoned up and then they started to compete on you know the big uh trading pairs where all the liquidity is. U very similar thing with FTX and I don't know if we're far out enough from FTX where we can you know say good things about them. obviously bad exchange, don't like SPF, they stole customer money, you know, not absolving that story at all, but they obviously did have a good product at the time that a lot of people liked. Um, and one of the things that I

remember them doing is they would just get per markets up faster u than everyone else. And they would innovate on uh they kind of did early innovation on prediction markets like with that Trump kind of yes no that's binary structure on betting on uh Trump uh versus I think Biden at the time and um they had pre-token trading markets and that type of stuff and hyperlid innovated a little bit with um HLP where you know it was kind of this pot where users could dump their money uh and they would act as a market maker for basically any trading pair and that was a huge part of the strategy for how they won market share in the beginning and so it seems like their strategy after this is to build out more financial services and become a chain, which is a big departure from what other exchanges have

opted to do. Uh, and so yeah, I'm just rambling here a little bit, but I don't know if that's also your kind of looking at it from the outside perspective on how exchanges have chosen to go to market. Um, but I'm interested to see basically what comes next for Hyperliquid here. Yeah, I think I mean I think it's to like sort of at a high level the way that I'm looking at it uh at that is like kind of finding the niche in the market where your competitors are not able to serve as well as you can as like a agile upstart or for other structural reasons. So I think uh hype they listed stuff super fast because they're just like a super small team. They're hungrier than these big thousand person

exchanges. Um, and they just got it done. Um, they also are a decentralized platform. So, people who want to trade in a permissionless environment or just trade from their wallet and not have to make an account with an exchange, um, that's something that you can only do in a decentralized platform. You can't do that on Binance. Um, so I think that that's another edge that they've had. Um, but yeah, I think they're they're probably approaching uh maxing out like those those edges like everyone who just really wants to trade in a DeFi uh environment and that's the reason that

they're trading there. They're probably starting to uh get most of those users get kind of saturated. So I think the next thing is going to be you know figuring out uh what value can you get from this atomic DeFi composability between the pers and hypervm. I think we're still super early in seeing what actually what useful products are there but um I think it's it's a very interesting design space and I I do think we'll see some really interesting products coming out. Um, and then you know potentially like um is there like an institutional pitch to say you can self-custody and maybe that reduces your risk when you're trading or like you know some other unique features that

Hype can provide that that others can't. Um so yeah, I'm I'm super curious what's going to be their next edge that they they really lean into. Yeah, I think that actually a structural advantage of dexes over sexes in general is that the long tail is actually much longer and much more important um than people thought it was going to be. And I remember a couple of months ago, Brian Armstrong of Coinbase came out and said, "Hey, we need to rethink what our listing process is going to be because this just we're overwhelmed here." Um and I know Binance actually um is a little bit sensitive. There was a, you know, when memecoins were a thing, um, it was actually, it went, obviously the

sex listing is still very important and they have a lot of leverage, but you kind of saw for the first time actually sex liquidity was kind of the exit liquidity, um, and all and kind of you would already have most of the price appreciation um, on a DEX environment before it graduated to, you know, sex counterparts. So, yeah, I think that, and this really depends on your view of memecoins and assets in general in crypto, right? Some people view that as very fluffy and uh very ephemeral and not going to last. My uh my own personal perspective on this editorialize a little bit. That is the beginning of a long trend. And I think you're going to see more like the volume of new assets getting issued on chain go up and up and up. And I think actually a a key feature of dexes in general is that it's much

easier to, you know, quote unquote list. Um doesn't mean every asset's going to go up. Doesn't mean we're going to get a 10-year memecoin super cycle. That's not what I'm saying. I'm just saying that I think the volume of assets coming to market is going to be much higher and it's a structural advantage for a decentralized over a centralized environment. Um so I think we didn't really talk about that on um on the episode, but I do think that's one advantage that Dexes have over their centralized account. And you you can see that we'll show the chart after this, but there's a um dex versus sex um you know percentage of trades and it's been creeping up into the right which is also cool and something that I think we want to see in crypto. Yeah, I think they're

I think they are leaning into that a bit too. Like um there's nowhere else where well other than you know Dexes and and Ethereum and Salana and such, but um in uh you know sort of a product experience like you expect from a centralized exchange, you know, a high performance order book and per market. There's nowhere else that you can uh do permissionless listings. So, um, yeah, I think, you know, it's it's clearly definitely a core part of their strategy and and going to be really interesting to see how that develops. I agree. All right, my man. This was really fun. Um, to tease some of our next episodes. So, we talked about uh, stable coins a little bit in this episode. Um, we're also just at the end there getting into borrow lend. Uh, so the next couple episodes we're going to be hearing from

pioneers in uh, both of those spaces as we kind of move through the major use cases of DeFi uh, today. So yeah, this was a really fun one and uh yeah, I'll see you here uh next week. Yep. See you. Can't wait. Cheers. [Music]