Is Hibachi About to Pick a Hyperliquid Path?

Transcript

this is going to be a very fun conversation we're going to hit on hyper liquid I've seen some tweets recently like saying ZK is at an all-time low but all I see is adoption straight up to the right I think uh you're getting there one one day at a time one proof at a time maybe the next hyper liquid is is using ZK right and what does that look like and I think that conversation is only going to get louder welcome back to the rollup your home for technical talks and open conversations that shape the modular future nothing on today's show is to be construed as Financial advice you can see our full disclosures list at the roll up.co Ventures sit back relax and enjoy today's episode after a quick word from some of our awesome Partners innovating Finance Global institutional and onchain deep liquidity

to drive Capital efficiency now maximize yields on institutional grade liquid staking with M unlock trillion dollar of capital via F Bitcoin all of this secured by Mantle's treasury of billions in stables and E Capital Pools this entire ecosystem is coordinated by a vast network of season PCS that are aiming to create the next generation of Banking and Financial technology this is mantle yeah gmgm welcome back to the rollup today we have the special privilege of Hosting our friend Jake from risk zero it's been a

special guest as of recently we've also got uh the head dishwasher uh co-founder uh tibal from uh Habachi and we've got verun from hash flow uh this is gonna be a very fun conversation we're gonna hit on uh hyper liquid uh and we're also going to be talking about what hyper Li gets right and what it gets wrong and how hash flow is building some uh some very very cool uh ZK enabled technology to basically build the ZK hyper liquid guys thank you so much for coming on why don't we start with some intros uh ver maybe you could you could start uh you know it's your first time on the show tell us a little bit about yourself and uh what got you into building very absolutely thank you Robbie and uh yes

so uh my background is uh essentially uh I'm former aerospace engineer uh got into crypto in like late 2015 early 2016 right and then and then uh uh over time you know ended up K hounding hash flow but the key motivation here is basically I always viewed crypto as the internet of uh uh what the internet is to money uh so and uh and this is for the first time you can actually basically build truly decentralized money that is cannot that cannot be manipulated or controlled by by any single Authority right so and that to me was a uh something I really

resonated with right so which is what got me into the space and then obviously I've been a builder all my life so uh sooner or later you get back to your old instincts so here we are love it tal tell us a little bit about yourself yeah yeah of course um I well so my my first kind of uh journey into crypto was actually uh when I was in school I I studied Finance uh at NYU um so my path was very much designed to go the Trad fire route and I think I felt a lot of frustration seeing kind of all the inefficiencies that uh were being taught in school as just normal um

anything from like settlement and just the way that you know transactions happen in traditional Finance it just didn't feel uh like that was a viable future so I very quickly turned into uh looking elsewhere and I think that's when blockchain tech and and crypto is just kind of starting to show uh defi to the world so so I really made my my first full dip into uh the industry during defi summer um from like a Marketplace point of view um you know first looking at uh nft marketplaces then spot marketplaces with with hash flow as well and then now I kind of graduated uh to to look into perss I feel like perss are uh very much NASA class of the future and so I'm just you

know very excited to kind of be building um towards this derivatives route because there there's a lot of opportunity there so yeah um you know still still kind of kicking um in in the in the east coast um and um super excited for this combo love it I'm right there with you man um cool and last but certainly not least uh We've Got Jake from risk zero and Jake i' I've uh I've been liking saying uh that ZK is one of the three uh fields in crypto that has found product Market fit we've got mean coin speculation we've got stable coins and we've got ZK and it feels like ZK is one of the few few uh applications here that

has found product Market fit um you guys are one of the leaders in the risk the inventors of the risk five ZK VM one of the leaders in the ZK space uh Jake welcome back how you feeling today good man good man and you know I mean I've seen some tweets recently like saying ZK is at an all-time low but all I see is adoption straight up to the right um so I think we have over 20 people dedicated to going to production with RIS zero now um people in the ZK you know landscape are really committing to this like RIS 5 ZK VM architecture as we see more and more teams building one um and we see efforts like um the the foundation currently you know spending

tens of millions of dollars to formally verify system so very exciting [Music] times yeah we had a great podcast with Alexander Hicks on formal verification and and generally uh the implementation plan of ZK into the Bas layer guys today we're going to be talking a little bit about the convergence of kind of all three of your backgrounds except Aerospace Veron sorry but uh Finance defi uh ZK blockchain uh this is all very very Cutting Edge in the sense that you know we have some of the most um viral technologies that are coming out and they're capturing a lot of Mind share and a lot of attention and a lot of

market share and then you know we realize once we look a little bit deeper into the tech that you know maybe there's not as many validators as we thought there was maybe this thing you know has a little bit more of a surface area for attack than than some may realize and you know that's very very crucial to understand in order to put large sums of money into these protocols and so people are call you guys uh the Bal and Veron people are calling hash flow the ZK hyper liquid before we even get into like whether that's true or not or what that even means maybe we could talk a little bit about hyper liquid since you know we're going to be going

with this naming and and ultimately what is kind of the problem space that arises when we build some of these you know fast quick relatively affordable protocols what what is the tradeoff what is the sacrifice and then ultimately how how have you guys uh squared the circle yes I think uh just to define the problem uh we we can simply take a step back and see uh fundamentally before uh we had the defy and whatnot uh you had crypto obviously had Bitcoin ethereum and so on so uh and eventually in 2015 16 you started this trend of new Bitcoin Forks emerge which later down the line you had ethereum and then new coins

being issued on ethereum which basically created this Marketplace for people to trade these cryptocurrencies right so which then led to exchanges to coinbase Kracken binance to emerge right so and at the time they obviously addressed the scaling need uh that that was an issue otherwise by just simply transacting on the blockchain right so your exchanges essentially took custody of your funds they let you trade on the platform and they pretty much operated like a bank right so which basically was kind interesting because you so of went about crypto to debank everybody except then you had new institutions emerg that function exactly like the bank so and then and then uh fast forward few years

you obviously saw the risks or systemic risk associated with that model and I think often times people blame it on custody and exchange is earning your money and that's a bad thing that's like one issue but I think a bigger issue more so than custody because I think custody is fine so long as there is an accountability and you know that the exchange is sented and you can verify that the exchange is solvent because one of the biggest issues with most of the exchanges that existed is that you cannot really prove that an exchange is solvent unless and until you see every record of every transaction that they perform now obviously they cannot share those records because uh that puts user

transaction and data at risk so uh as a result it's kind of thing even if you think an exchange should publish yes they can publish their reserves they can show this is how much balance we're holding in our wallets uh without showing every single transaction but then there is no guarantee that these reserves are enough to meet the liabilities that an exchange has towards its users right so and that was the big issue with FTX as well where you could verify an FTX wallet and say okay oh cool FTX has half a billion in money but then you had no idea that they owed people two billion right so um unless and until you could have actually seen every single transaction then you going go wait a minute there is something

fishy here right but you can't really you couldn't really do that and which basically takes you to it's not just the proof of reserves you also need uh to prove that those reserves are sufficient to me liabilities so I like to use use the term proof of sancy which basically means anybody can verify at the exchange is Sol right so and this can also broadly be applied to uh any institution that is responsible for managing money in general right you could go all the way to8 crisis and so on where people who are managing lots of you know money for somebody else in general there should be a way to verfy them in the past we used institutions like you know big McKenzie and whatnot to perform

these Audits and publish reports and turns out ZK is like the mathematical version of meckenzie who can just just do that uh and allow anybody to independently verify these things without uh trusting a sing you know any Authority or a human rather right so that's kind of where the ZK became interesting and obviously one of the paradigms that we attempted to solve this exchange problem is to put everything on Chain by like like you know what let's just do every transaction on chain because that's the only way to attack of Sol for custody but then that brings another set of problem is that you are now constrained to the L1 restrictions so I like if your

L1 scales then great if it doesn't then you know you're basically bottlenecked by that so you sell that with exchange on ethereum uh then serum salana it's good but at the same time now if you narrow down on Exchange specific applications especially when you need like high performance or high frequency trading or something uh where you're trading at like sub millisecond latency and so on one that requires lots of performance it's really hard to build that on an L1 because by a LW of physics you can't really beat build a decentralized system that is as performant as a centralized system right so there is some sacrifice who always end up making right

so depending on the application I think our philosophy here is that centralization decentralization mem so I took really at the end of the day it comes down to what aspects of the system must be decentralized or even better better put are the things that need to be needed independent verification those are the things you should basically optimize for apply ZK all that everything else can remain centralized and optimized for user experience right so and that's the philosophy we've been leaning into when we built uh hash xos and then hiachi is a product that is using xos right and I think hyper liquid as you mentioned what are they solving so they've attacked the same problem that we talked about but

their approach has been more along the lines of building an L1 that actually puts the exchange on the L1 right so now obviously at the moment they have three or four nodes so it's fast but over time you'll be interesting to see how they maintain uh that that same performance as they end up scaling adding more nodes and things like that does it actually affect the performance or does it degrade it and and so on right so um that's I think just a general framing is that both hyper liquid and hiachi are trying to solve the same problem in the sense that how do you make an exchange uh basically prove it solvency anyone can independently verify solvency but

the approach is slightly different where we are essentially using ZK to to address uh to keep the performance on par with any centralized system but then use ZK to be able to uh verify the exchange solvency with without moving the centralized Parts uh on chain as hyper liquid is basically building everything from scratch on chain right but then the trade-off here is obviously uh you might compromise some performance over time as you expand the number of nodes validating it right so my time will tell I if they even do expand the number of nodes and and add more decentralization ultimately that is the trade-off is is you know the the

performance is great the speed it's affordable yeah uh low fees however the they're trading off decentralization and you know you could argue that up to a certain point you're trading off security if you're trading off decentralization and so you know at at what point does that become uh not worth the risk of of security well um it would be great if you could have your cake and eat it too and that's the goal with ZK is that you can have just as much performance but also maintain the levels of security so that every everything is cryptographically secured um and not just by two or three nodes and so I you know it is is ZK at that point like you

know the the the counterargument to this is probably that you know ZK is still slow and expensive to generate proofs and to verify proofs so how are we going to create a ZK ified hyper liquid that can run at the same Pace with as low a feed you know the same the same levels of performance uh that has everything ZK proven so Robbie like I like one more thing on the hyper liquid bit so uh while we can argue that you know they may not decentralize the nodes but at the same time I think one of the things to keep in mind is that even if they don't decentralize the nodes if people have full visibility into the chain and the

transactions then one could make a case that that is sufficient because now you can audit what the exchange is doing and if they do something malicious you instantly know exactly what's going on right so so obviously the counter point to that is that uh if the nodes who are in fact adding transactions on chain are the ones who have the power to manipulate the transactions then a user can essentially also not know if the transaction that was entered on chain was correct or not right so in those case Things become like how can a user exit from the chain which is something we addressed the CK part but those are things if hyper liquid addresses without even adding extra nodes that is still would be interesting to see how do you

make sure that these transactions were not manipulated by the nodes that run it even if the node sample size is like small right so anyway that is to say now I think maybe Jacob you want to take the other question or or I can go into it yeah when we think about like ZK right now I think there's three things that come to mind with performance it's throughput latency and cost um when I look at cost right now I think we're there um you know a deck swap is currently around like a 100th of a cent this Rivals the fastest cheapest chains I think actually chains under load this is currently cheaper than anything remember with ZK we don't have like fluctuating gas cost it's a flat cost

for compute so the more transactions you do um you know we we scale larly it's not exponential like in most blockchain systems um for throughput we're also there um um with systems like you know Rio zkm from the very beginning um you know we understood that the way the internet scaled was horizontally and that's the same way ZK is going to scale um so we can go as you know wide as possible to just keep up with a risk five executor which theoretically is just as fast as you can go right like we can't create faster risk five execution um you know computer chip companies need to do that the third thing is latency um latency I think is still the one lagging

behind you know we think about this a lot and you know we've designed rer to be like the fastest uh you know I think we're the we have the minimal latency in the market currently um but it's it's still going you know we're hunting this whole idea of realtime proving right now for single things but with something like Habachi and you know if you want to use xos um what you could do is you can provide this pre-confirmation um and get this you know Central exchange latency um you know single round trip that's as fast as you're going to get the same sort of latency you know you experience on salana or one of the leading l2s right now um and it's still going to feel really good as a user the drawback there um is just going to be withdrawal time but you know with something like cash

flow or you know what Habachi is doing right now um this can be 10 15 minutes to withdraw your funds from the exchange and for some chains that's the same time you get on coinbase so I don't think the users are going to feel a big drawback I think one other point I wanted to touch on real quick was Veron had a had an interesting point about you know hyper liquid when they add more nodes they're going to slow down why is that because all the nodes need to communicate they all need to reexecute the execution uh you know to validate and come to consensus the really cool thing about boundless and what we're building you know over here at rer is we actually say more nodes more power right if one person you know adds 10 gpus that's more

that's 10 more gpus that we get for verifiable execution um if you want to add 10,000 gpus we now have 10,000 more gpus and all these applications sort of get the pull from this ever growing Global liquidity pull of proving and could you could you break down realtime proving just you know for listeners that may be unfamiliar um because this is this is a very common theme when it comes to some of these ZK conversations yeah so real time proving um I think it's it's sort of a liquid definition that's you know sort of hardening hardening up one thing to consider is that it's different on you know whichever blockchain you're on realtime proving refers to the idea of

being able to generate a proof in verify it in a single slot of a blockchain uh the real big use case that you know I think people are hunting right now is once you get under 12 seconds it's really 11 seconds for ethereum because you need 1 second to propagate the proof and everything um you now get asynchronous composability between all rollups on ethereum and this is really nice um you know I think we've been talking about the fragmentation issue between rollups for over a year now um this essentially could eliminate that and you could get Atomic bundles and atomic transactions um across rollups really making each roll up blend together um

and the user would would wouldn't even start to feel the difference on top of I think you just get more efficient market forces um you know for trading and stuff like that yeah absolutely and this is I mean this is part of the ZK Vision um it seems like we're going to get real time proving and and you know Jake as you said you know this is different on different chains it all comes down to the block time and so on bitcoin for instance if you know opcat was there we could verify proofs we would already have bit we would already have real time proving right um and so ultimately you know we're we're going to start to work towards the chains the l1s faster block times um and and Implement real time proving there as block times are become

quicker and quicker real time proving is more difficult um because time from the blockchain perspective is is uh moving faster uh ultimately you know this unlocks this entire bundle of of uh Atomic interupt as you said where you know we can unlock basically these contract calls I can go you know for instance if I have to execute a Liquidation on a I can go make that swap on Unis swap bring it back liquidate that Capital all of that can can occur within the same block um so realtime proving is gonna unlock unlock a lot of these use cases um how how F how soon do you think we'll get real time proving yeah I

think at this point there there's a path to real time proving um I would be you know to all the listeners be careful about people claiming realtime proving before it's practical realtime proving does not only refer to generating a proof um you know under the single slot it's from requesting the proof to you know propagating it and in a system like the zkm that like you know a lot of people are adopting there there's many steps here we need to execute the program we need we then need to generate the proof sometimes you need to wrap the proof to actually fit on chain um and getting all of that under you know 12 11 seconds is is going to be tricky I think

the actual proving I mean we might see as soon as the summer um but a practical you know usable implementation um I still expect to land sometime you know this year where um I think you know there could be a possible way with the current risk zero zvm um I'm personally very excited about Bas um you know we're partnering with the people over at irreducible um to build the world's first risk f ckvm built on top of NES and you know it's all theoretical for now but the speedups seem to be a magnitude um or you know a 20x faster um which would land us right there in directly into a practical implementation

of real time proving another phenomenal podcast Jim from irreducible is just super such a smart guy and and was really just talking through like the the mathematics of how we're able to speed up proving and verifying ZK K and ultimately what he's doing with binary uh binary uh proofs and binary Fields is just it's incredible stuff um so definitely encourage everyone to check that pod out as well um ultimately like a this is a lot of infrastructure a lot of theoretical talks with theal I'd love to kind of get your perspective as you know you you've been someone who's kind of traced this especially coming from you know traditional Finance as well you know you probably have your your your finger on the pulse when it comes to

like what users actually want so you know we can talk about infrastructure all day but ultimately it has to find product Market fit you know we have to we have to ultimately get users excited and and involved with this so how are you thinking about taking this infrastructure the ZK and and uh every everything that Veron has talked about early on in this podcast and and making it and wrapping it around a front end and an application that ultimately is some is one that users know and love how has that experience been and ultimately what are you hearing from users such that you know it makes them really excited for this product yeah I think the history that verun kind of walked us through at the

beginning is really at the foundation of why we built Habachi uh you can look at the last five 10 years of exchanges in our industry and and see kind of like the the the wins and the failures across the across time uh whether it's from the centralized exchange point of view or decentralized exchange point of view you but but really kind of um we wanted to approach this uh like you said from the point of view of the user like what do they really care about and you know everybody will uh lean towards you know ux and and speed Etc um and and really when when you take uh when you look at a product from that lens um you really have to look throughout the stack

and see okay what is the best option I have for this particular um you know uh level the stack whether it's like at the transaction level at the order book level or you're going down to settlement all the way to like top of funnel like where can I receive funds from what chains can I easily adopt different ecosystems Etc so um that's why we chose DK really and I think Jacob had a really good point earlier he's talking about scaling right like we have to think about scaling all the time as well but that linear that linear scaling component to it mean uh that allows us to essentially have as much as many transactions as possible within any kind of point in time and then that doesn't impact or affect our ability to put everything on

chain at the at the settlement level is is really really key for us so you know kind of just taking a step back uh building with an offchain order book one it gets us all the benefits of centralized speed um and and Traders care about that you know I we spent a lot of time trying to understand what you know a good exchange means for all different types of Traders and you know being a derivative exchange um speed is you know at the top of the list for n most people that I've ever talked to that trade uh derivatives so you really can't compromise on speed that's that's for sure so you know um the second thing is obviously that proof of solvency

component like you you want to make sure your funds are safe you know I think that that is the The Exchange kind of uh tradeoffs that we've had to live through over the the years and I think finally uh with the evolution of ZK and all this Tech that's coming out we don't have to make that tradeoff again uh we've kind of gotten to the point where I I truly believe like we we've reached kind of the end game of exchanges like this is how exchanges are going to be built moving forward because you really do have the best of what's available out there um at every level of the stack so um overall I think know I'm really excited to scale this product up and and to keep adopting the different evolutions on the ZK level but

everything else as well right I think um one really interesting thing about hiachi is that we can kind of adopt any chain really easily because again the order book itself is offchain right there's no there there is very low barrier for us to start accepting you know uh different collateral assets different deposit assets um Etc so when you ask the question about like what does the user want um you know right now we're kind of still thinking about the users that have already uh adopted you know blockchain Tech and they're probably the the same cohort of users that we've had for the last few years realistically so when that next uh adoption cycle happens you know what are what are the products that are going to

be right at their fingertips um and it's most likely going to be the ones that are super super easy to adopt um that don't require a bridge that don't require uh users to understand what a chain even is right so um we're thinking about things at the top of the funnel uh from the adoption point of view but we're also making sure that we hit all and and all the essentially checkboxes in terms of what a good exchange means all the way from speed to you know that that trustless environment Redstone is def's modular Oracle supporting over 100 clients on over 50 chains offering 1,000 assets Redstone is the leader for Bitcoin

staking and yield bearing assets they offer the most robust customizable price feeds available in any model join the Redstone exhibition [Music] today absolutely ultimately from from a chain abstraction point of view like that narrative isn't dead like you know we're still going to want this unified environment this one where we can basically deposit any asset to your point and then we can go and we can swap that we can trade that and we can use that as collateral in this environment that allows us to have access to all of these new tokens all of these new assets and products and so as we kind of you know abstract away more and Jake I want

to get your perspective on kind of like moving from what I would call a chain Centric view to more of this ZK offchain State view But ultimately as we move towards that you know it it'll it'll open up all these new cases and all these ux improvements to your point as long as we don't lose the decentralization and verifiability I think this is a great great thing because everything becomes kind of like out there and we can just connect it we're no longer siloed based on oh you're on this change you have access to these tokens these assets and these apps no everything is just kind of getting getting unpacked and put into the void and then it's going to be up to ZK to

connect this jacob I think you can probably put this this mental model together a lot more eloquently and and kind of you know highlight what theal is talking about from the app perspective maybe you can talk about it more from from the infrastructure and the mental model yeah so you bunch of points to hit on here one is you know Robbie you sort of mentioned like the verifiability and traditionally you know we think verifiability comes from decentralization because in these distributed systems um you know a small group of people could at times you know impact liveness if you're in the validator set and then in some systems as little as 33% you can start to impact

safety and with ZK these um restrictions really lift very nicely um one is with safety as long as you assume a well implemented proof system there is no longer any safety assumption you cannot verify the program if you run it incorrectly um the soundless property of ZK um the second thing is liveness in ZK you only need a single party to ensure that your protocol uh remains alive it's exactly how we've built boundless you know every step of the way of you know the proof life cycle any party can come easily in and fulfill a step of the process to you know tibo and your point

Robbie you know back in you know 2019 2021 one there was this whole idea of the fat protocol thesis right how do I get a vertically integrated stack um to where you know I have all this value capture you know we saw ecosystems like ethereum and salana Avalanche polygon all pop up you know this forc developers into a you know sort of ju toos of now I have to choose an ecosystem based on grants arguably um how many users I have the efficiency of these blockchains that really all these blockchains are making a set of tradeoffs um when it comes to certain

properties then we kind of you know fast forward to the last I would say two or three years where you have the modular ecosystem let me build these hyers specialized pieces um a lot of the times they're you know app chains you know Cosmos things like Celestia sequencers like you know espresso um but then we've had a really hard problem sort of connecting them all together and in general it's just been a really hard problem because now you have all these hypers specialized pieces these are critical you know infrastructure pieces putting them all together is just just hard I think boundless the future we're thinking about is what I'm sort of calling the universal protocol era let's lift all of your logic into a ZK environment and to tbo's point we can

start to settle on a lot of different places Habachi is doing this uh you know rashen from in the team are doing this um and one use case we can point to right now that's you know sort of live is is I layer so boundless is helping power igen layers slashing um sort of like Stak tables and this is you know the core logic of the protocol but now that we have this zkp and you know all this logic is sort of in a ZK environment off chain um you can now use igen layers reaking regardless of where you are because we have this nice little little proof that we can verify on any chain um and I think this is the way we're going to go you developers no

longer have to choose a chain they just deploy to all of them um and this really helps people you know access the most users access the most liquidity and this is exactly how we've you know built boundless from the beginning is not only you know how do we give people I think you know ZK becomes the best verifiable execution environment in the next year but how do we make sure it's super easy for them to deploy on any chain so those people as an application start to get the best based properties from whatever settlement chain you want to do and make it as easy as possible I love it th I mean this mental model has really shifted my thinking too because ultimately when you when we break out of this chain Centric view we realize how

like how Petty some of the problems app Builders were facing what like the idea of what chain am I going to Deploy on is no longer a question that we have to answer um you just deploy to all of them I think that's a very very powerful statement for Builders uh verun is this something that you've already realized and you know if so how have you been able to kind of take some of these some of these realizations and kind of add on top when you guys you know ultimately were building uh xos and and uh and Habachi you know so I I think Jacob was spot on right so uh like yes you can build applications on chain but then every single time when we were doing

even har RFQ at the beginning like deploying to ethereum then we okay now need to do arbitrum now we need to do salana now we need to keep adding these things it was just a was fragmenting the user experience significantly but also uh I just don't think it was like even easy to maintain as as a backend uh as an application to do all those things and it was just getting chaos you know messy right and I can only imagine the pain a user must go through but when they have to go and they see like 10 different variants of ethereum uh on on a single wallet and then they're oh wait which one do I click right so I think the end the end result the users they want a seamless

experience where they can just go buy and sell what they want to do without thinking about anything else like any backend complexity that exists today like that should not be a thing a user should be seeing I think one of the things the reason why we even ended up in the spot is because we started smart contract development and then we started translating the smart contract into a front end like one to one so your front ends start feeling a lot more technical because they're basically interpreting the smart contract quite literally right so the the design principle that we like to cons you know we've been approaching this with is anytime there's any complexity that you see that is not something a user has been used to figure

a way to hide it like the end thing the user should see is only something that they are already familiar with or if not it's something they can relate to but don't create or add any blockchain Jon for example we don't even talk about ZK in the front end when the users using an exchange like they don't need to know about there's a ZK environment that is powering any of these things right so uh what we need to make sure is that we have a system in place that is going to protect the users we can obviously talk about these in academic setting why people actually understand the technology they will scrutinize what's going on but for an average Trader who wants to come and buy and sell they just need to know the exchange is liquid and

is fast and it's going to give them exactly what they're asking for right so they should not be worried about bridging or figuring out how to redeem their funds or or uh making you know everything else is is a secondary thing when it comes to uh the end users right so and I think that probably applies to many applications who are building uh on this technology and uh I should double click on Jacob's Point earlier uh around real time proving so as an exchange one of the interesting things is you know the way we work in the back end is we are basically running a central limit orderb where transactions happen then you batch these transactions then you compute a Zer knowledge proof on these

transactions and then you submit that proof to uh respective chains for settlement right so now really uh the latency at which you know five years ago when Stark was the only uh ZK thing in town you know I think the Prova times were like nearly eight hours and you could argue okay that's a long time and really what it impacts is how quickly can a user withdraw their funds from an exchange right but I think with the modern systems like resero you know suck syn and so on you can basically do these things in like 10 minutes or less with ver zero right now we able to generate proofs every 5 to 10 minutes and that is a sufficient time for for like no users

going to keep withdrawing money every five minutes like that just doesn't happen it's very very rare right so so in that case like five less than 10 minutes is already a time that is sufficient for anyone to build like reasonably good applications without worrying about that right so it actually is already there you know five years ago you've said CK is like not there yet but now I think it just makes it easy to build these things right so which is where be okay fine now we can just create these things and not worry about this chain tribalism that has existed in the past right so yeah I I would I would push back or critique just a little bit and say that you know absolutely right

now we have ZK proofs at the at the scale and performance where we can withdraw safely quickly all of that because you're right when it when the me mentality around withdrawing my funds from an exchange I'm not I'm probably not going to be doing that every 10 minutes right and if I do it's like you know it's just a few of them and then and then I'm done but if we do achieve real time proving yeah then I think the terminology will start to shift and it we'll no longer call it withdrawing it'll just be like a asynchronous interoperability call between different contracts or between different applications and so you know I may find a better price over on this

exchange maybe I'll go and I'll I'll you know perform some kind of Arbitrage where you know nowadays I would have to call it a withdraw and a deposit make it and then maybe deposit back but in the case of real-time proving withdraw is kind of It kind of exits from our vocabulary and it just kind of like you know I can just go execute that and and actually keep my funds on the exchange on the front end you know that just looks like one button click on the back end it's a it's a deposit uh swap Arbitrage and then deposit back but all of that could be Aus skated away from the user if we had real time proving at that scale yeah for sure and and to one more thing there right so we still able to do

it today even with the current constraints I think what we end up doing is we maintain a pool of funds and if a user wants to move the funds elsewhere uh we let them you know we basically allow you know them to move those funds but from using the pool yeah and then the pool is basically crediting or lending funds to the user who just wants to do it instantly you call instol and then you refund that using the proof that gets entered into the chain like five minutes later right so and and you see that all over the space now with with intent-based solvers and and you know bridges that are holding inventory and and and I think that is you know applications can do that in particular and we see that as a microcosm of of the the way the space is moving generally um

because that is the primary use case for or excuse me the primary uh uh design for interop nowadays holsome inventory across is you know the best for this right now moving across l2s obviously if you have a very sizable transaction or if it's on the longtail of assets or chains maybe solvers don't have the inventory um but you could and and you know in the future you'll have a cryptographic proof that'll do it anyway but in the case where you know that's not that's not available you have intent-based solvers ready to fill those orders and so absolutely it is a great bridge while you know we continue to accelerate the math and the ZK you know for the next year or two you know we'll

see maybe intent-based solvers do live on beyond that for longtail But ultimately it seems like these this is a near-term solution that works pretty well ultimately once we have better ZK we will achieve uh cryptographic ZK interop for really the longtail and probably the most popular routes as well you know for sure like yeah what I was describing is like a there's a temporary mandate that already works for 99% use cases but once we get the real time proving then you don't even need that right so then you can just basically go straight into that Jake did you have something to add there yeah I was goingon to say like I don't view the relationship between intense and ZK as binary I actually see an opportunity

currently and I believe there's some teams working on this um to improve intent based systems with CK as well a lot of uh intent based systems sort of work on this optimistic approach and you have you know in the case of across I think it's like one hour and some system it's different of like where approver needs to fulfill an intend not approver sorry uh you know fulfiller I don't remember the exact fruity and slip solver filler uh Sol there's a difference I've seen the debates it's fine um where they you know need to provide the liquidity and then wait an hour and you know they're subject to um you know Market volatility an hour is a long time in crypto and you can actually

use ZK to start you know bumping down that latency and that time they're subject to volatility in the market um that essentially you know will be reflected in the user fee that that is paid so as we approach realtime proving not only you know we're still hunting that and that's the Holy Grail but um we'll see ZK have an impct on these systems and reduce fees for users in the in the meanwhile yeah absolutely I I'm I'm stoked I mean I I every once in a while like this space just surprises me and uh I have a feeling ZK is going to do that here over the coming months love this conversation and uh I I'm really looking

forward to to using this product as well uh because I I think that uh we're just going to see more and more like this so yeah guys any closings before we wrap up I would say go and give I mean we talk a lot about Theory and and and infro I would say go and try hiachi because you know that's where you truly see ZK in action or we kind of feel the magic of ZK on what it could do because we've talked about how it can scale things and everything is abstracted away but then once you actually give it a shot you realize oh wait a minute you can actually reach like low millisecond latency just pretty much navigate through anything and you wouldn't even know that there is some you know a lot of technology that went into it and that's been taking you abstract away

from you so give it a shot and I think uh that's where I think we keep saying the future is pable so so I think uh you're getting there one step one day at a time one proof at a time so yeah I I I would just I would just add um that the timing is like uh is so great like you're saying Robbie I think there's just uh I think we're getting to this point now where people are starting to realize that that ZK is actually you know has the opportunity of of kind of reshaping a lot of the things that we've been used to for the you know years now um so I'm just personally um as my last thoughts I'm very excited for what's to come still uh I I feel like it's still

super early uh and and and I also you know feel like we're in a great position because we've been working on this tech for you know longer than than most people I I the from like the point of view of an application so um we're super super hungry to keep you know uh bringing in like any all the developments and and real time improving and all these things that we talked about over the last 45 minutes I think um it's it's just now that I'm starting to see people talk about this more and like realize oh like wait maybe the next hyper liquid is is using ZK right and what does that look like and I think that conversation is only going to get louder um so yeah uh you know thanks thanks so much the time and um I'm sure

there's more to come yeah and that was that was one thing that verun had mentioned early and I'm glad you you kind of brought us full circle because you know one one of the ideas here is that this technology this xos provable order like that that can go to traditional financial institutions it can go to web 2 like this is technology that makes essentially anything zifi and our traditional financial institution can benefit from that just as much as relatively centralized uh uh exchanges right so Absol absolutely yeah absolutely yeah like settlement and in trafi is a is a

complete other thing but it's even slower than what we're used to in on chain unfortunately so yeah I think that's a really really good point we got keep saying if you look at the evolution of Technology you have transistors right that changed the Wy compute and then and then you had the internet then you had the cloud computing in the late 90s and then iPhone again that basically what Computing to enhance now I think the next next evolution in this stack is provable Computing wherever you basically now have all the Computing in your hands but it also gives people the opportunity to verify or prove that those computations are correct so you don't run into the systemic failures

that you've had in the past and I guess to end us on all this uh a quote from a RIS zero team member um this person like you know he's been in like traditional you know web 2 companies um worked on a lot of crazy things um and he actually came out of retirement to work at risero and his quote the first week or my first week at the company that I love is not every decade you get to work on a cryptographic primitive that changes the world um and I truly really believe that you know we've been chatting about ZK since 2016 last year it really became tangible affordable scalable and you know we're still going to push performance to the

edge but this year is the year ZK finally becomes tangible um people get to experience and traditionally when you finally make it tangible when you make that product that you know really opens people's eyes like a Habachi like things people are doing with ckls um like things people are going to do with traditional web 2 applications and just use a blockchain for that great settlement dis intermediation value that it provides um the hockey stick adoption is upon us I see the internal numbers um we are well on our way to have one hell of a year love it guys thank you so much let's Z KFI the

world let's do it see you next time thanks guys