DeFi by Design EP#99: Unpacking Cairo 1.0 and zkRollup Scaling

Transcript

foreign [Music] by Design where we talk all things blockchain and cryptocurrency while striving to educate Empower and generate [Music] what's going on guys welcome back to D5 by Design podcast episode 99 one away from a hundred what a journey it's been and we're excited today to dive into a topic that we've been going on about for at least six months now perhaps even close getting close to a year which is the development of Layer Two specifically on starknet what's going on there uh what's going on with Cairo the programming language the upgrades uh a

lot of the the protocols and projects there and then also just some general thoughts about Dows and kind of how you can get involved to help Dallas um how how Peter has been kind of Consulting and working in the crypto space for some time now kind of some of his General and specific thoughts as well about the space so looking forward to diving into this one and of course today I'm joined with Peter uh who has been spearheading a couple projects in the space since he joined so would love to hear a bit about kind of your journey so far I'll let our audience know kind of what you've been working on and we'll jump right into it so good to have you on Peter and uh good

morning yeah great to be here uh good afternoon um I'm clearly one episode too early but uh hopefully the the 99 is pretty cool as well um yeah so my brief intro is I work on yagi finance which is uh yield aggregator on start net and then I also do Consulting for web 3 teams feeling that back I started working on crypto in 2018 I came from a mix of an engineering and management consulting background so initially worked at startups like Google and Twitter in California uh engineering and data science roles then got involved with

McKinsey as a Management Consultant was basically building kind of spreadsheets and PowerPoints uh for two and a half years there working in different Industries as a strategy consultant and it was really fun to kind of Step Back From engineering for a bit and really try and pick up all the soft skills that I definitely lacked um you also had the green mathematics which made it even worse right so I really needed that I think kind of business school style experience and then got involved with product management when McKinsey acquired Quantum black to do AI type work start building products built a open source Library there that's now a machine learning framework part of the Linux foundation called kedro and just really enjoy the process of working on a developer tool as a product

manager thinking about kind of how to design the API and so on and also just the impact of building an open source project building a community and so on and so it was a fairly natural transition to then you know getting exposed to crypto where I saw a lot of developers being able to have a lot of impact and that was kind of one of the things that actually led me down the route of going into management consulting and product management where I saw that I was somehow capped as a developer and then crypto I saw people who generally have necessarily those skills but still could like lead companies do impactful projects and and that's what really pulled me in and so 2018 I started doing consulting for various web 3 projects really focusing on this developer tools idea uh specifically focusing more on security

so I started off doing kind of formal verification tooling dived into debuggers played around with using reinforcement learning to find bugs in solidity really a bunch of a bunch of things uh kind of to do with the ebm and then as the space matured and we kind of got into the bowl cycle of 2020 started working more bits more contracts did a bunch of governance proposals did smaller projects for teams working on kind of small Insurance funds or vaults and things like that in defy and then really late 2021 came across starknet and it was a kind of an inflection moment where I felt that I've learned you know a lot about smart contract development from

really working on these projects but also felt like I wanted to learn about l2s I was really excited about what was happening with zero knowledge I saw Stark and I had a completely different programming language that seemed really exciting and so it was like like the perfect moment to basically start a project and so I started working on yagi as a yield aggregator for stocknet yeah yeah I kind of had a similar um understanding of of layered shoes around the same time kind of when arbitrim really launched and I I ended up bridging a lot of what I was doing over to arbitrum um and just testing it out and you know I had been messing around on Avalanche and Phantom and things so I was already

sold on the idea of not doing all my transactions on mainnet but then arbitrary you know with the kind of understanding of the of the security um and the speed and kind of the in inheritance of the ethereum uh ethereum ethereum kind of network if you will um you know it it kind of was obvious that that those were going to be superior to some of these other evm chains that I was messing around in and then yes uh similarly boom Stark net hit um you know and and there happened to be this kind of strong Community around starknet at that time kind of around the same time as you mentioned mid-2021 uh early 2022 um yeah and it's really kept its shape

all the way up until now um you know and building pretty much just a variety of different D5 protocols um you know in this in this new language with the kind of ZK snark Tech um yeah I mean briefly for for our audience to start off I would love to kind of hear why you feel that the ZK Stark and the ZK Tech of Stark net is superior to um whether it be optimistic Roll-Ups or other forms of blockchain um kind of uh execution and settlement layers um you know in in the least technical way that that somebody could understand if possible

yeah um I think the yeah it's trying to go very non-technical I would say at a high level what's really exciting about ZK technology is that you kind of the kind of I think going back to like why are transactions expensive on ethereum they're expensive because effectively we're paying people to re replay the transactions many many times over right and so that's kind of why ethereum is more expensive than AWS right something runs once on AWS but if we play a transaction on ethereum we need a bunch of validators to just keep replaying that transaction and so the idea behind ZK is instead of replaying the whole transaction can we produce a proof is that the transaction happen

and can we find a much more efficient algorithm to kind of run that proof and verify the transaction happened without having to rerun it so we still get you know a large number of validators who are validating these transactions but the validation process is substantially different and it and it happens by kind of validating these proofs and so you can then get into what are the different types of kind of their ZK snarks and there's there's the ZK Starks which are from two different systems and Starks is really what uh starkman uses in particular and and and that were actually invented by the team behind Stark where and the kind of high level comparison between Starks and snarks you can you can actually look at them in many different categories and it's fairly

interesting how it will play out in terms of which ones will matter and I I would say it is somewhat use case specific because you can look at things like proof size you can look at things like transaction costs bandwidths you know all these different things but um the the theoretical potential of Stark of starks is greater but the the challenge is really of building a ergonomic ability to to write Stark proofs and so the the kind of intuitive comparison is like when you think about ZK evm ZK EVMS are typically written in snarks because the mapping between like what we're doing on the evm level and snarks is very direct versus if you're working with Starks it is much harder to map it

into kind of the programming language Paradigm we're familiar with of mutable memory and so on that you know solidity provides access to and therefore if you want to I could get the extra scale potential of starks unfortunately you have to kind of build your own virtual machine and so that's the hard work that the starkner team has done and really building Chiron building the whole stack behind Cairo to build a stark native programming language to kind of help unlock that scale and and you know it seems like it seems like that would be a lot of Canon necessary work like why are we going through that trouble of creating a whole new programming language uh when we have solidity and like shouldn't solidity be kind of the JavaScript the smart contracts or should we just use solidity for everything I actually think the answer is no and the

reason um even though we may feel that for example today uh you know optimism arbitrim all these platforms are fairly cheap right you can you can deploy smart contracts on them and they're just so much cheaper to use than domain and ethereum I think what we've historically found is that people exhaust kind of the scale that they're given and so if you look back at what happened on Main and ethereum we started off with people experimenting with with with with all sorts of applications right we at some point we had you know we had Landing protocols that were storing loan books on chain right when that was affordable and just eventually as we start to kind of saturate block space we naturally find that we saturated whatever efficiency is given to us and so I think

and now if you're a good solidity developer right like well as as we've seen with openc for example that that contract was handcrafted with assembly so all of a sudden we're working solidity but we're really not because we're actually jumping in we're looking at Huff you know if you're doing any anything serious with Mev you're probably coding in Huff uh if you're writing really kind of serious high volume solidity contracts are probably jumping into yule and so kind of the the ease of use and the familiarity of solidity is something we just end up for going anyways and so at that point it's fairly compelling to say hey no we can we can jump on the Cairo bandwagon and here use a language that's actually more high level even easier to use and so that it has additional safety properties in the language which you can dive into

um and and actually be more efficient so it just feels like an impossible trade-up but it's one we're able to make today yeah and I think that kind of thought process and messaging is going to be very important for really just onboarding people into Stark net and into the this kind of Realm of of thinking right I mean ultimately people don't really don't really care that much about you know um you know the ins and outs of let's say like the the technological languages in the program but what they care about is the speed and the cost um and also by by an inherent nature of the security as well uh and so you know I think bringing this messaging for across the board for Stark net teams

into mediums like such um you know like the show or just kind of in easy to understand um threads and posts and articles and things it's going to be very key for kind of onboarding more people into understanding first why you know why this is happening why why do we need to um have Trevor why do we why should I use start now and you know I think you make quite a few good points uh and from the developer side um most of the the developers that I've spoken to uh with regards to Cairo also kind of are in the same boat right they they think you know it's just a superior programming language as as as a whole um you know and currently obviously there's there's uh stuckness still in

development as far as its uh speed and transactions and its capabilities but um you know my kind of thought process is we're about a year or so out from ZK uh evm polygon zkum Stark net uh CK sync and and the other kind of bigger ones um you know really taking stride and then I think that really gets us to a point where there's two things that I'm quite interested in hearing your opinion on first is what is the world of Roll-Ups look like like we're uh are we gonna have like dozens and dozens and dozens of different Roll-Ups and you know are we going to have um different optimistic ones EK ones you know and potentially we can even dive into like how that they might interoper interoperate with each other but then on the more you know kind of uh fun side if

and I think that I would like to understand an explanation too as far as decentralized of these Roll-Ups we saw our Arbors from launch their token optimism to Stark net has already um hinted that they will uh and that they've already you know gone ahead and distributed some to different teams and things so each of these Roll-Ups is going to have to decentralize their sequencer which we can kind of unpack what that means and decentralize and solve the kind of like the nothing at stake problem so are all these Roll-Ups just going to have tokens as as well if we exist in this world with a bunch of Roll-Ups so yeah we'd love for you to unpack kind of is you know what's the future of these Roll-Ups you know how are is there going to be just tons of

them how they kind of interoperate and then yeah how will they decentralize will there be a token uh for all of them and kind of you know what what's your thoughts on that front as well yeah great question and and one that you know I wouldn't say I have the perfect answer to I don't think anybody does but it's it's let's try and speculate I think there's a couple of directions this this thing is developing in and there's a couple frames of reference to use um one frame of reference is that is the sort of scalability lens of Roll-Ups which is you know you make a roll-up you're effectively compressing you know n transactions into one costs go down bandwidth goes up and so on and so it's

purely a scalability play and one where you're basically trading off um you're potentially introducing some latency into the picture right so if you're minting an nft on a roll-up it takes a bit longer for that nft to eventually kind of appear at the lower level of mainnet and so and the reason I'm kind of explaining it this way is I think because it partially explains why we have this idea of fractal scaling where we have L1 we have L2 but then people are already building and thinking about building layer threes on top and the idea is that we can kind of keep going further and further on this kind of pure scalability spectrum of let's just compress more transactions um into into into less state

while trading away potentially you know some security Properties or some latency properties in terms of having that guarantee of being able to eventually withdraw that withdraw that money on layer one so that's I think one angle uh in which Roll-Ups are pushing which is basically like whatever Network they're under just scale that network but then there's kind of this computational diversity angle which is I think also super interesting and I think we're just not at that stage yet where that's really playing out because most of the Roll-Ups that we're seeing are really scalability focused Roll-Ups they're really focused on just General computation and scaling ethereum as the narrative but we've started these discussions certainly in this Darkness Community those discussions start up when talking

about layer 3 Roll-Ups because the question is okay you know starting as the base Network are there some games or are there some other applications that require their own app specific Roll-Ups right and then the question is you know why do they require those and then one idea is like computational niches the idea that there's some virtual machine layout that is just better for a certain application right for example for games you can optimize for specific requirements maybe they need a lot of storage and then you just you can really over engineer the roll up to focus on that aspect of how to store it specific State we'll have obviously um kind of you know D by DX and other other types of order based systems Roll-Ups that are that are targeted at that you know that's kind of how Stark X

was actually born um which is kind of the precursor to Stark net was actually um a very specific type of roll-up that was developed to and help develop amms and the different types of um sort of different types of Dex is basically um so that's the second point and so and then there's kind of the third point which is which is the interplay of those Roll-Ups because again if we go into a roll-up and now we're in a world where the Roll-Ups are are there because of computational niches not just as general platforms then then all of a sudden we want some composability like we're losing some composability right because if you have to go to one roll up to play your game and then to another roll-up to maybe access defy and that you want to invest those gaming assets to get

generate some yield now you have a problem right so we have this like third question of how do we start composing rollups how do we start interoperating and so obviously we have you know things like you know shared sequencers and um you know there's all this kind of optimization on on L1 around you know how to make roll up transactions cheaper so I think there's like a lot of research going on in that direction I think we have the perfect layout but but that's certainly going to be another trade-off of you can get very specific but then where do you lose the composability is the rest of the ecosystem um I think the honestly like the future is just going to be the interplay of these three factors and the trade-outs between them and and so it's very it's gonna be very domain specific I think in

gaming we're probably going to see much more silos um than we may see in D5 where we really buy the composability a bit more um but um but yeah it's ultimately very domain specific and hopefully we'll see a couple more domains right maybe you know around the nft space there's going to be specific needs emerging there um maybe we'll start to finally see I think Enterprise blockchains is another angle which with with we've kind of forgotten about that part right we I it was kind of around the time I started in blockchain 2018 I think Enterprise blockchains were still part of the narrative this idea that you know 20 Banks could agree to deploy a private blockchain between themselves I think now that that narrative is is kind of falling by the wayside right

we found there were solar challenges with that one was actually there's no connection back to public blockchains you know we don't really we haven't yet developed Bridges they're really secure at connecting two l1s right um so there's a challenge there but actually now that you think well what if what if that private blockchain could be a roll-up let's say on top of an L2 or top of an L1 all of a sudden it starts to make more sense because you can you can restrict that role of to have a very limited number of validators or whatever rules you want around that Roll-Up and you can and you can still access the underlying assets right so all so now those Banks could settle with ethereum with stable coins and and have and and not have to build it all from scratch right not have to recreate the whole D5 stack from

scratch and so I think private blockchains are potentially you know have a way of coming back through the roll-up narrative so I think those you know those are just some of the factors they'll play out as far as your second question around the tokens you know it seems like yeah it seems like it seems like most Roll-Ups are you know will need a token to decentralize I think we could definitely see non-decentralized Roll-Ups that are going to be you know potentially for games like not every roll up maybe for a game needs to be decentralized um and and potentially we're going to see maybe more creative structures especially when one team develops multiple Roll-Ups perhaps you know they don't each need a unique token perhaps there's some Synergy there and and then

again in terms of in terms of that kind of vertical scalability piece you know if um if a network decides okay we need another roll-up for people to make certain latency transaction cost trade-off like maybe the same native token could just could just be used at that level as well so I think a lot of considerations there but I think for now and to be honest like I think the you know one of the one of the appealing things with apps app specific networks right it's sort of if you if you make a really successful smart contract you do a big token launch and you have a massive dial now if you want to 10x that um what do you do well so you look at D5 llama and you say well the only tokens that are bigger than mine are the ones from like l1s and l2s

and that's probably that's a lot of work to build a whole ecosystem around that but what's kind of in between building my own L1 and building a smart contract well it's building my app chain right so so that's why I think a lot of teams also from Financial perspective probably look at options as an opportunity to really continue growing as a company and so so I think we'll definitely see a lot of tokens associated with Roll-Ups effectively in the in the short term at least yeah we we've already seen um our specific chains like you're saying for um order book taxes dydx and D gate is another one they created their own ZK Roll-Ups specifically for their own order book Decks that actually uh sponsor and good good good long-time

partner and friend of ours at the roll up um and and I think gaming is is also clearly clearly a a huge one too because like you said the financial aspects of the game uh perhaps less important than the uh speed and then and the user experience versus and the security versus somewhere like um you know D5 uh when you want to have kind of those money Legos working together um you know and in addition to that make sure that everything is fully secure and then on the three points of scaling I mean those are those are really um you know they're intuitive when you say them but before you kind of think about them um you know it's hard to kind of fathom

what that lipstick because it's really deep diving into the abstract um and so it's it's yeah it's a tough speculative call to make but uh inevitably we're going to see more and more uh basically layers of scaling happen whether it be layer twos of different layer ones and then kind of as you're saying you know rolling up even further uh from layer twos to layer three so nonetheless very exciting and I kind of want to touch a little bit deeper on on the token aspect and just kind of ask a broad question like aside from Grants and community incentivization and just you know making everybody happy and hyped and making everybody feel good and giving everybody free money you know like why does stark net need a token

oh yeah great question I think I think generally I think one of the if we look back at kind of the past six months or around around Stark knit and we can summarize the narrative I think a lot of people ran out of technical challenges to point to because because initially you know I think Stark net Stark X was one of the you know just raw TPS and a bunch of other metrics it was it was it was just on the leaderboard uh in terms of Roll-Ups very early on so the question just became you know can it turn into a general smart contract platform and when the Stark net team proved it proved that uh with with Cairo

um they they showed that you know there's a language that can actually make it really easy to build um these types of applications so we started to run out of of of of different critiques of of the community and I think the one critique that remained is really still around scalability I think the the TPS is not matching what we had on Star kicks but there's a lot of improvements coming I think actually very very soon that will probably within the next week or so that I think are actually going to change that and so I think we've caught up on the tpsi but it's taken probably 12 months to really make a dent there and then the second concern was always around decentralization which is like well you know you guys are building all these really cool things but the prover is not open source and

the sequencer is not open source and it's all private and it's like you know what's what's gonna happen with that and so I think that's been like a lot of the focus also over the past you know six to 12 months you know a bunch of those things are open source now there's open source projects that have spun you know the moderate sequencer for example to create independent versions of each of these components But ultimately there needs to be a governance process now to decide right what happens with all of that so I think that's where the token comes into play into the final piece that essentialize is really you know who's going to be working on those things what are going to be the canonical versions The canonical sequencers what's going to be the canonical proofer and the final step is basically you know the token and so

that's why you know Stark now has a foundation that's why a lot of the decisions around you know all the the great things you mentioned like grants incentives and so on are now really moving into different committees as part of the starkman foundation is because we're you know we're trying to decentralize the governance of the network yep that's uh that's kind of the the idea that I had as well um you know just wanted to confirm there because that that seems to be the underlying issue when when Roll-Ups launched without a token is there's this kind of uh centralization of of factors that are very important as well as again there's this kind of thing where uh the validators of the roll-up are ultimately

um not being compensated and also there's there's that kind of um there's no token so there's there's no value at stake in the chain in in the sense that if if the T nose building behind the roll-up somehow you know lost it all or went bankrupt or had a terrible Fallout there wouldn't be uh any repercussions for just kind of shutting down because well there's nothing at stake there other than kind of what's uh what what I I guess you call them hard assets have been bridged or are are kind of on on the Chain so it gives the team and the community something to Rally behind and in terms of actual value creation and then of course yeah the other kind of uh Community marketing BD type of things

are much more enabled um and you can you can onboard a lot more Talent uh through grants and things like this um so you're pretty excited about what's going on with the upcoming TPS upgrades which of course I am as well I've been using uh Bravos and kind of all different types of applications on on mainnet and test net for some time now and yeah I've noticed that um you know the speed is still it's it's you know it's it's nothing close to arbitrary or kind of optimism or even some of these other layer ones but with this Cairo 1.0 going live some of these upgrades happening um you know you're pretty bullish on that so yeah we would love for you to unpack kind of what's going on with Cairo 1.0 what that's all about why it's

important um to you and for the teams that are building on starknet and yeah what kind of expected upgrades that we can uh see come soon yeah absolutely so Cairo One is kind of the the second big release of Cairo um we call the old one Cairo zero and Kairos there are actually historically evolved um as a very low level kind of Assembly Language that was just a way to abstract the creation of these zika proofs and so it started super low level then when people wanted to build contracts on top of it the team decided to create some hard order syntax that resembled python on top of it but it was never designed to be higher ordered language so there were some there were a

lot of conceptual challenges kind of working with with Cairo zero um the ecosystem was still able to recreate a lot of contracts that we see on mainnet and also try some new ideas using Cairo zero But ultimately it was clear that that language was suboptimal um and so Cairo One is kind of the from scratch rewrite of that and I think what's exciting about Cairo one is a bunch of things so really first it's it's designed from the ground up to kind of have the a layer based approach where there's CR in between and Sarah's kind of the intermediate representation and what that allows the team to do is to really stabilize this sort of Intermediate Language but then I'll allow the higher level language to

evolve very quickly and so that without necessitating basically Network level changes right so especially as we move to decentralization that's just gonna be very very useful as a property and so kind of the that's sort of from compiler architecture perspective it's really a decoupling of Stockton as a network and then Cairo is the language that that enables the second piece is kind of just you know getting up to par in terms of ergonomics where I I my experience with Cairo zero was that it's it was more difficult to to develop and then solidity I think Cairo One definitely gets up to par then the more exciting is Cairo One actually incorporates many many ideas that solidity should be incorporating and basically feels much more like a regular programming language and so that's where it starts to get

exciting Cairo one was modeled after rust so a lot of the syntax um is similar to the language that has become a favorite language for back-end developers in in the crypto space recently right with the admin of Foundry and other you know libraries on top of rust and so it's a very very common language but then really tailored for smart contract development and then there's a bunch of extra things that you know even I'm discovering recently digging into the language such as you know turns out it has linear types for example that is actually something that very very few smart contract languages have the first one to really incorporate linear types and get popularity was move and obviously with the kind of whole resource model

so that's also very exciting which is that Cairo one is actually now taking a Step Beyond rust and and incorporating more advanced type systems that are really going to help in terms of making contracts more secure um so that's kind of the pure ergonomics piece and then there's also the I would say the development the developer experience piece which is hard to explain but kind of one of the one of the reasons that people are so excited about Foundry in the first place is that you know so that he was really slow to work with and so Foundry really allow the development cycle to get a lot faster purely because of just technical speed of how quickly you can run a test or how quickly you can compile the code and so on and with Cairo we're seeing an even order magnitude improvement over I

think what we saw with Foundry and so that's the thing that gets me really excited we're working with Cairo really feels like I'm just working with you know C plus plus or python or just regular programming language all of a sudden not this really slow kind of smart contract environment and so that's from purely a developer perspective I think cardio one is really exciting and of course all the things that's enabling on the back end for the network is is kind of the the big unlock that allows kind of the TPS improvements and so on to continue cascading and there's a lot of you know there's a bunch of pieces in the stack that are that are that are being experimented with now I think are going to unlock further scalability so for example um the the Lambda team is is building they're building a way to compile Cairo

code to mlir and so mlir is a system built to effectively help more effectively uh compile tensorflow at Google and it's a sort of very very Advanced compiler framework where as long as you can compile things to mlir you can compile things to to to work really fast on hardware and so now we move to a world where we don't need to kind of emulate the way cargo works but we can really run Cairo in the same speed that we could run on a regular computer program that's doing kind of non-smart contract related things and I think that's going to be a complete game changer for developers you know the more they they discover it yeah yeah that's awesome yeah that's uh a whole leg up against kind of what else

is being built out there and I think a lot of a lot of the reason for uh my General bullishness on Stark net comes from comes from this aspect uh and you know I'm I'm kind of more on the non-tech side but I've I've done quite a bit of research in speaking with you and other people who are who who build with Cairo all kind of have the same thought process as far as a little bit tough to understand this is Cairo zero I'm speaking of but once you get it uh the the possibilities become uh much more Innovative and Limitless than other languages now with Cairo 1.0 kind of more familiar with um devs who understand rust or you know just overall an easier transition uh you know that

kind of uh conviction is is strengthened even further and I'm curious is account abstraction uh enabled because of Cairo uh is that like a direct correlation uh and briefly if you could explain you know account abstraction for for audience and why it's important uh before kind of diving into that one yeah um so I think one of the fun things about the Striking ecosystem is that it's it it's it's it's trying to work towards over determined success and what I mean by that is that there's not like one thing it's it's not like oh you know Stark net has Starks and then it's just the minimum like the minimum

implementation of everything else like they they weirdly try to innovate across many different levels and and so I think actually uh account abstraction is is somewhat independent from Cairo in the sense that you could you could see a kind of abstraction implemented in saluting in fact there's there's there's obviously active proposals for account abstraction to to go live on many other networks um that's being implemented on many other networks um with with sort of practical challenges around adoption and so on I think the difference with starkman is they they got it done really early and they basically established that as this is the canonical way that accounts are going to work so what is account abstraction really it's a it's a it's a decoupling uh of the account and the the signing mechanism I would say in simplest form

so in in in in in traditional in traditional Nautica abstraction model or the the end user account uh model you basically you have an account and it has like a built-in signing approach where you need a private key and there's like a specific signing scheme you need to use to validate transactions and then account abstraction kind of decouples those things it says you can have an account that account can just be a smart contract and that smart contract you get to Define how transactions are signed as the kind of developer of that smart contract and so now if you if you still just want to use private Keys you can build that with a sound abstraction that's totally fine but then you can also say hey

uh let's you know let's allow this user to log in through Safari and let's use secure an enclave and let's access that key and then use that key to validate their transactions right so all of a sudden you can you can make accounts of things that traditionally couldn't have been accountants um whereas if you had to do that with if you had to do that with um ethereum could have before account abstraction you would always have had to generate that private key right so you want to create a throwaway wallet you have to generate a private key if you wanted to let users log in from another type of service then the service on the back end would have had to create that private key right it couldn't have directly tied into the whatever the authentication mechanism it wants so it's really a decoupling of the you know

that allows many many different ways of signing and validating transactions that are much more ergonomic right and and sometimes those things are just you know there's going to be easier account recovery because you can involve your friends sometimes you know if you're playing a game and you need five dollars to play the game and just buy the nfts you need like maybe you don't need a hardware wallet for that right maybe you just want a really lightweight mechanism and you you know worst case you lose that account for some reason or or you're willing to kind of take on some centralization risk right that this this game provider uh maybe they're managing the account on your behalf maybe that's totally fine because um the only thing that matters is if you do get a a bigger nft that you can then migrate that to an account that's more

right so just just it just opens up the space of trade-offs so so kind of to summarize you know it I don't think it's it's necessarily you know it's coming from from Chiro One I think it's an independent feature that networks could adopt percentages really early in using a kind of abstraction and so I think we're seeing some teams like cartridge for example just really uh you know cartridge our building a a platform for people to access different games you know it's you can think of it as the Steam for starknet and so their approach with account abstraction is they've been able to incorporate you know your secure Enclave on your Safari or on your iPhone to authenticate into that platform right so those are the kinds of things that the

Stockton Community has been experimenting with for a while now yeah yeah it's good to know that that there's a difference there um you know a lot of I'll I would say 99 of the account abstraction conversations um that I've had have been with sharknet based teams or uh industry leaders so um you know it's nice to know that that could be separated and I think really it's the future of wallets I think the metamask experience kind of sucks um and even like rabi and other wallets are you know they're better but they're still just not the same as far as um different functions and what one of the biggest ones that really struck me was did a podcast with uh Rohit from

Jedi swap months ago and um you know he touched on setting automation parameters for your wallet so where you know you could have a setting in your wallet where you know let's say every week you set it a parameter where your your wallet sends you know a thousand usdc to your your you know um I don't know to to a different wallet or or to or it it can automate a transaction for you um you know without really you having to do much but kind of set it up um and this and the ability to have your wallet be a smart contract I mean you can do so many things with calls and with functions and with different

um types of interactions and transactions that just aren't really able to happen with a uh kind of uh a normal world so I I kind of view a kind of extraction wallets as like a smartphone and um the the the kind of the setup that we currently have is like um uh like a a wallet that's in your pocket uh with credit cards and cash so uh very exciting to hear those takes and to hear some of the teams that are building about uh using that technology and improving the experience for users which ultimately I think is what it all comes down to uh I think the users might not really care about um the difference between Starks and Starks I do I think it's cool I think it's very interesting I don't think

they'll care very much about the exciting ideas of why Cairo you know why Cairo 1.0 is great right but they don't really care but like when they can do these easy transactions on mobile wallets like braavos or plena or Argent um you know and it feels so smooth and it's cheap and easy and there's tons of innovative functions you know then I think then I think they'll care especially when they're profiting off it so I think it just kind of enables that next level of adoption um you know and that kind of brings us full circle into what you're building which I think could be very very big I think I think yagi could be one of the bigger projects on starknet by TBL as a whole

um you know I've seen beefy on on ethereum as well as on um all the other teams get always have the best rates um and be very very reliable and beefy's another yield aggregator for those uh who may have not heard of it so yeah I'm curious kind of what teams you're thinking about building with what are some of the yield strategies that you're thinking about implementing whether it be Dex LP positions um you know derivative decks um kind of debt pool type positions um you know new token emissions start net token emissions kind of what what what are your thoughts around you know some of the yield opportunities that are going to be available in Stark net soon or you know are currently available now

yeah um yeah I think you're I think you're absolutely right uh to kind of fall on from that I think that kind of abstraction is one of the most exciting user-facing things and that's ultimately all that matters really is what do the users get at the end of The Thing and um and I think our camera section is kind of part of you know General transition that we're taking you know going through from users actually having to actively send transactions into the Hot Topic which is intense which is basically more generalized versions to Define transactions you know perhaps instead of saying I want to trade it you know uh sell this asset on uniswap I just want to get rid of this asset and whatever I can get the best price really moving kind of at a more higher level abstraction and then I think the next

version of that is really objectives and kind of what I call passive Finance where you you what you really you don't actually want to be taking any actions right you ideally want to just deposit your money and have it grow over time and then withdraw it when you whenever you need it right and so so that's kind of what Yogi's there for um for people who are familiar with yield backgraders already um it's it's the the kind of core differences with another yield platform right that um like urine for example or other platforms like beefy is that you know one we're based on Stark net so we're gonna fully focus on Native started yields you know big Believers in Stark net so um that that's that's a core part and we'll talk about what those yield

opportunities are the second piece is actually confusing the unique properties of Stark net to do things that were not possible on mainit and so one of those things is really moving more of the logic on chain so the goal with the Ikea is to really make it as trustless as possible and so the goal is to you know use as little oracles as possible to use this little off chain management as possible and really have it truly run as an autonomous contract and you know I'm a big fan of kind of index funds and traditional Finance world because they are very clear like they do what it says I'm thin right an index fund describes exactly how it's going to allocate Capital what assets going to use and you know kind of what you're betting on and then you know weirdly turns out that actually for most

people investing in index funds tends to be much more profitable than trying to find you know active managers and I think the same is going to be true in crypto where most people are just one to you know look at kind of what the strategy is and just deploying that strategy so I think that's what that's what that's what Yogi can become with the properties of starknet by making computation really cheap by making storage really cheap we're going to be able to just put a lot of algorithms that currently happen off-chain from any yield aggregators such as hey we have 10 10 10 kind of funds and we have some predictions on how they're going to return let's optimize the allocation between those 10 funds I think we can put that on chain with yagi and then to answer your question about specific yield opportunities

um the teams partnered with before on on testnet were ex-bank originally to build a lending wrapper Vault uh Jedi swap to build a two-sided liquidity Vault so this is a way to basically um participate in am liquidity provision without having both tokens so if it's a die eat pool we built a two vaults one where you can just deposit die that way you can you can deposit eth and then both depositors will get kind of merged together and we'll make a collective liquidity share for them but they both get the same effectively return profile in the end and then also on the more exotic side um did a little collaboration with Realms to build a staking fault for their ecosystem

and and I think those directions kind of very much describe what we're looking to do on mainnet in terms of really yeah working with D5 teams to build aggregations on top of you know finding the best lending yields across the different options finding ways to provide liquidity in amms um and then also potentially partnering with gaming teams I think that's again which Stockton's going to make that much more possible on mainnet what we saw is that there was not a lot of opportunity for people to really automate their participation in games and it's something I think people have appetite for we saw a prototype uh with abegashi there's like a Abe gashi uh Petting Farm where you could it was like a daycare where you could you could

actually deposit your abigashi and then it automated the petting action I think with Stark knit hopefully we'll see much more sophisticated gaming economies I think we'll actually have vaults where you are going to have your you know game token resource and you're going to say you know what like I don't have time to play this game all the time but I do want to have some automation happening while I'm not playing the game and so deposit intervals for that reason and um and and so yeah so I think it's just really about uh and then the kind of the third piece which you alluded to as well is obviously rewards and tokens and so I think to the extent that starkner teams are gonna be you know launching tokens or creating incentives that are based on tokens obviously it'll be natural to build those strategies that you know

automate the the harvesting or of their Awards and so on so that's definitely something you know in terms of a yield aggregator my big belief is that your aggregators have to be very responsive to the market you know it's it's kind of a very different contract to build compared to an amm or an amm you kind of build your contract and then you're done you're going to think about the next version versus with a yield aggregator you constantly have to be putting out vaults that really capture the best opportunities available so that's kind of um so so part of this is like you know these are the things I'm looking to build but part of it is actually um really seeing how the ecosystem is going to develop and we're going to respond to that amazing man I look forward to seeing it uh come to fruition especially with the

up the upgrades in CPS um and uh tarot 1.0 as well as just new new projects and developments on starknet so um yeah man it was an absolute pleasure having you on um and uh I I look forward to seeing what you build again and yeah thanks for your time yeah it's it's been a pleasure great questions and uh yeah I'm I'm I'm I'm I'm very excited for the roll-up ecosystem and I think you know stocking is just one of many teams in it and I'm also rooting for a lot of the other teams to succeed as well uh don't get me wrong I I'm I think this is uh you know the Journey of scaling ethereum is this very exciting one and one that's not you know it's not simply about again doing more of the same but it's about all

these different niches yep 100 tons of Roll-Ups tons of DK Tech tons of scaling I look forward to it as well so um you're doing keep doing your part we'll keep doing ours and and hope to see it uh come back in a year or two's time and things will look a lot different awesome cheers buddy