Building a ZK-Based Blockchain w/ Steve Pack (o1Labs)

Transcript

[Music] hello and welcome back to the strange waterer podcast once again thank you for joining us for today's conversation on this show we've spent a lot of time talking about the recent developments in the ZK cryptography ecosystem as we take our first steps into what we all hope to be the next great crypto Market cycle we've gotten the opportunity on this show to speak with many of the companies who will eventually join the Giants of the previous Cycles giants like coinbase or Alchemy and this is happening now

because now in 2024 it's becoming painful ful clear just how transformative ZK is for blockchain briefly here's how I'll summarize it ZK cryptography can be used to prove that a specific computer program generated a specific output when the program runs it also generates a proof which when validated mathematically guarantees that the result was valid and produced honestly computationally proof verification is is orders of magnitude simpler than actually running the computation and so ZK allows a resource limited system to access the resources

of a non-limited system blockchains inherently are resource limited systems because keeping system requirements low is essential to decentralization so to put it all together ZK can be used to augment the resource limited environments of blockchains with with near unlimited computing power and with modern resources the world is wide open because it doesn't matter how much computation your application needs as long as it's got a ZK proof it can exist on chain as I said today this Insight is clear to so many people that we're

seeing an entire subcategory for Venture Capital but for 01 Labs Developers of what would eventually be called Mina protocol this Vision was clear back in the prepandemic times founded in 2017 Mina protocol answers the question what would happen if you designed a blockchain with ZK woven directly into its DNA which brings us to today's guest Steve pack head of product at 01 Labs after walking through Steve's background at cloudflare we learn what Mina protocol is and how it works then we'll talk through how Mina and ethereum

will continue to grow and to grow together and make sure you stick around for the end where Steve talks through how 01 Labs will bring ZK technology to optimism and the op stack this is a super fun episode and I know you'll enjoy it one more thing before we begin please do not take Financial advice from this or any podcast blockchain and ZK cryptography will change the world one day but you can easily lose all of your money between now and then okay let's bring on Steve Steve thank you so much for

joining us on the strange water podcast hey Rex happy to be here thanks for having me of course man I've been so excited to be talk uh talking to you guys at 01 and about Mina protocol I mean that's like the OG like beating heart of uh ZK and blockchain so super excited to dig into it but um before we get into you know crypto and blockchain and all that stuff uh I'm a huge believer that the most important part of every conversation are the people in it so with that being said can you give us a little bit about your background like how did your career start and how did you find crypto yeah sure so I've had a varied career I was a full-time software engineer and engineering manager for between 15 and 17 years so definitely come at it from that like you know

Hands-On keyboard Tech Guy um but I always had interest in in various things um I was an entrepreneur after that um shipped a a hardware product um and it was after that that I got into like really sort of Silicon Valley product management I would say with uh 5 years 5 years at Cloud flare um so that that was kind of like the profession profal Road leading to a full-time career in crypto um but yeah like many of us you know had been sort of in and around the edges and like experimenting observing um you know all of that sort of throughout that that period um actually if I remember back the first the first sort of crypto thing was um colleague of mine at a at a finance startup saying Steve there's

this new thing called Bitcoin and its value can't go down because there's a limited Supply and and I quite patiently explained that like well it can because there's supply and demand and the demand can go down and that could be because of Technology risk or because of real risk or people just don't want to use it and one of those things like I I was right but uh was was kind of wrong um in terms of like those those risks don't matter um you know and it went on to be wildly successful obviously so that was my first um little exposure um I remember actually at that same role someone into crypto generated ether addresses with like people's names in it um you know I've got an email somewhere with a with

a wallet with my name in it like or at least whatever ones fit into the you know the letters they used like yeah five for S I don't know what a t is but yeah yeah exactly with with an eth that is like in some email server um yeah and then like just like you know like I said I was I've always been a tech guy right like you know before I became more product focused and so I'd always observe the the technology and I'd like take a look and be like okay this this stuff's cool like but you know what what can I do with it and you know I kept sort of dipping in and out I guess like each time there was a new wave honestly like you know like many there like I was a bit of a victim to the hype Cycles early on um and yeah it

just became this trend that like each time I would dip my toe back in there was more and more you know real world utility more interesting use cases and and and and more and more stuff happening um so yeah that was the sort of crypto on the side before it became crypto full-time well I that that's super interesting so full disclosure I joined crypto in 2021 so I'm very young but thank you um and I entered uh you know I have a corporate finance background I entered thinking that like this technology here was here to change finance and the kind of deeper I've Doven into it the more I realize like yeah I don't know I mean maybe this is here to change Finance um but like quite frankly I think like the interesting parts are about like literally

transferring value and immutable ledgers and not like crazy esoterics um but I'm still here and like even more passionate and and believe more in this because to me the story of really ethereum but let's just call it blockchain is the story of decentralized computing and it sounds like you you your Finance friend who by the way I would love to know where he's working so that we don't put money there but um it sounds like your Finance fan really your Finance friend really tried to bring you in from the you know hard Money store value like all this stuffff and you were not interested in that but with with each hype cycle came more and more of the capability and like this vision of distributed compute

that is like quite frankly kind of obvious today and so I'm curious like what was the first moment where you realize like oh my God this is going to be an important primitive and an important system just like the internet is and like I need to Pivot to this as opposed to kind of like a fun toy that like you may or may not be willing to spend some free time on but like you also have a job man that's hard to answer cuz it's like you know it wasn't sort of an epiphany or um and you know I think a lot of people maybe in crypto maybe this is true for a lot of people but not everyone maybe says it it's like there were various times where like it was clear to me that there was something here whether it was Bitcoin as a you

know store of value or is in and a hedge against you know Fiat mismanagement um and then the sort of like yeah dist you decentralized Computing um you know all all that stuff right and and my um commitment is tested right each time like you know there's a blow up whether it's regulatory or whether it's you know just the like the the Celsius types of things you know um but I guess like yeah as time goes on right like my um uh not commitment but yeah like my level of belief that like this is here to day and it solves real problems um you know gets stronger each time and you

know I'm I've been based in the US for a number of years and you know all it at least in the for the bitcoin news case like you know all it takes is to go onto the treasury website and take a look at the the level of issuance to think that like yeah like people are asking real questions about like is you know is this sort of um way of managing a budget like sustainable and if not like what is the Hedge against it and think their answer that many part answer to the problem and many part answer to the like how to hedge against it but I think more and more Bitcoin is is is part of that answer yeah makes a lot of sense and I guess before we like really transition into crypto um I would love to talk a little bit about your time at Cloud

flare um because I think that what we're building here is about core infrastructure and the next evolution of the internet but since I have the opportunity I have to ask someone who like really saw the Insight of like the current version of the internet um like what what insights and what lessons did you like pick up at Cloud flare that you think are really applicable to building in web 3 and yeah like how did that experience like really give you a leg up uh versus people who maybe didn't work at like these big tech companies or maybe came to crypto like straight out of University or without real experience I think there's three like three things that really come to mind which is like you know product

management at at a big company um there's the like crypto insights I got and then there's like just being at the sort of pointy edge of like the big bad internet and like the technology um so you know firstly product wise like you know like I I said I'd be in software I'd I'd run a company like I I knew something about product management but just um you know KFA had great discipline of like you know like sounds obvious but talking to customers and over over and over again and like any doubts you had like you answered by talking to more and more customers until you got that feeling in your belly that like this is the thing that I need to build and you know I I sometimes talk to

PM candidates or or other product folks in the industry and it's like there are insights you get from looking at data and it's like you know does the blue button or the red button convert better and like you know those are nice and that's fun it's nice when you can like really just do but most insights come from like you know talking to developers talking to people like whoever your customer base is and hearing the same thing 10 times getting that feeling in your belly like turning that into a product Insight that you share with those people and then you go build it and so just that like you know Relentless customer focus um and prioritization right you can only you know of those things that you find out you can only work on a few of them you know that that's what I got I guess from

that it was kind of my like you know Silicon Valley Product training I think um and very very grateful for um another one was and this helped make the jump was um like crypto companies would be Cloud fler customers um but like not in the same way that many um uh other companies are so like lots of like Cloud FL started as this self- serve you know like um individuals could use it there's a free plan there's a $20 plan and then there's like Enterprise I use it for free yeah right like um it's great service um but like so many crypto uh companies started by putting down a credit card and then all of a sudden like that credit card was running up like thousands and thousands of dollars

of usage fees and like the sales team were like what the hell is this and you know they kept finding it's like M RPC Gateway and like you know um block Explorer and what and and they didn't know what it was right and so there was kind of like this who knows about crypto and you know there was a group of us obviously who did and um so we sort of it sort of like you know the company discovered crypto as a good customer base and a rapidly growing one kind of by accident and then you know started to invest like a bit more in it so I guess I could see like the growth trajectory just by by that right that like this whole new customer segment came up and then like the longer I was there and as I was in more senior roles and started

to look at like strategic sort of Partnerships with crypto companies um you know we were getting asked to like look at what would it take to um to add sort of some of cloud Flair's distributed compute footprint to the validator set of of some um you know crypto companies and as part of that it was like okay what is the what is a crypto validate a workload compared to like uh you know the the capacity of the cloud FL Network and um what you know as part of doing that like we needed projections right where does where do you see the volume going and just seeing some of the projected volume from like a number of these um you know projects it was just clear to me then like yeah um even more like commitment gets stronger

each time each cycle but it was like yeah this is this is here to stay and it's it's going to keep growing yeah and out of curi like no need to disclose anything that's like sensitive but when was the era in which you on the inside of cloud flare which for those who are not uh knowledgeable is basically like the like skeletal system of the internet right at what point did at what era did you start to notice that there was actual like serious like you know Global levels of usage from crypto companies was this the Ico boom was this like the 2020 era when was it no it was more 2022 like that was the the sort of that that bull market runup um you know we we had crypto customers before then but they were kind

of just like individual customers um but this kind of like where is all this usage coming from like who are these people putting down credit cards that go from $20 a month to thousands of dollars a month like you know yeah that was that was more of a 20 um 2022 era um actually I said there were three things I think I only covered covered two the the third um and yeah this one's um this one's kind of different like yeah Cloud flare is at the edge right like literally you know often described as an edge Network between the internet and like some some web service and as that you kind of get to get familiar with what's on the other side like what's in out there in the big bad world of the internet and and the answer is like scary in terms of a whole

lot of abuse a whole lot of bots a whole lot of attacks a whole lot of like you know the worst possible content you can imagine and people doing the most terrible things you can imagine and you know you got to you got to find out a way to navigate that to protect your customers from it but like you know Cloud Flair tried to also you know like walk this line of you know free speech and and following the law and um you know complex thorny issues and um you know crypto also has to deal with them but I I feel like there's kind of two camps right there's the kind of like we're going to ignore that cuz we're full full scale Libertarians vers the like well in the case of if we're a u if you're a US regulated company then you know you

got to follow the law and uh yeah I guess that's one thing that I was more exposed to there than I think a lot of folks maybe um you know who have been crypto natives the whole time yeah and and uh recommendation for the audience for me something that was incredibly eye openening was reading the book tracers in the dark which is really a puff piece uh of the founding story of CH chainalysis and so we can have discussions on if they're good actors or not but like you need to understand what goes on the internet and like how crypto is involved in it and um I don't even really understand how like Cloud how involved Cloud flare is in in trying to stop that but like the there's a lot to

like really understand about what's going on and if you want to be in this space like it's important to know what we're floating right at the edge of um but yeah very very interesting point it's like everything lasts in a tradeoff but got to know what you're trading trading off against yeah yeah yeah yeah good point so sorry last question on cloud Flair and then we'll move on so you said that they they were first starting to notice like oh my God this is like something real around the 20212 time frame same time I did right and so my question to you is like are are you able to kind of like register SPF happened to all of us and we all experience it in different ways and for True Believers like us like that

was just well like there's going to be Bernie made off there's going to be Elizabeth Holmes like we had one like no big deal for a lot of people it was like this is the proof we need to show that not one single valid thing happens in this community um how do you think cloud flare understood not just SBF but the entire like bare market and like the crash and what was it like for cloud flare to go through their first cycle no it's a kind of a nonevent like crypto as a as a percentage of you know Cloud FL is like not not large and you know the cloud FL service is you know like dealing with traffic right like rather than tvl or trading volume whatever so like honestly it was

probably some customers sort a sort a dip you know like um in usage it it not really an event wow well I think that's also a lesson to us that like with our own Echo Chambers like we're still thinking about SPF and literally no one else has ever thought of him since he goes arrested right cool all right so let's talk about um like crypto and I think uh in order to get us to Mino protocol can you tell us a little bit about like how did you end up at 01 Labs uh yeah 01 Labs 01 Labs Big O yeah got it you know as I was sort of um you know looking for opportunities to like go full-time in this thing that I was you know passionate about um you know there's a there's a lot of chains out

there there's a lot of projects out there um you know so o1 Labs was very um you know closely tied to the Mina Foundation were the original incubators of the Mina protocol and still the primary contributors um and I just excuse me I respected a lot of the like you know the technologists so the people that I met there that there was it was a true crypto crypto and ZK OG company that like had launched a chain that like was working on you know the um the newest of the new and like I said you know product gu but from a tech background like I've got to I've got to be interested you know in the technology and um you know me meor is is is fascinating and so um yeah it was just a good mix of like a you know a really

interesting technology base with an amazing team good culture and you know I I I saw good opportunities for mea protocol to really like you know make an impact in in in the crypto sphere yeah awesome so I guess now is a great uh moment to like anyone listening to this podcast is already familiar with what blockchain is and definitely familiar with like what ZK proofs are but can you talk a little bit about what Mina protocol is and how they use ZK proof as a primitive within the protocol to like really change how we think of blockchains sure um maybe like start at a kind of high level and and and kind of go down from there like you know if

you've sort of interacted with ethereum your you know your experiences like everything's in the open right like you you docks yourself every time you make a transaction you know you pay5 to $100 in fees depending on um congestion um You probably send your transactions through a centralized service like INF fur or something and um you know very large validator set very large proof Estate Network but you maybe some questions about how decentralized it is and how many like of how much of that validation is done by by small by by parties um you know don't want to be all negative about like you know that like the there's a whole lot of positive stuff but like that's that's also the experience right like today and um so

Mina uses ZK to address some of that so one um you don't pay for gas in the in the traditional sense that um that that you do on ethereum and we can depends how deep we can go but um one mental model is in ethereum like you send a transaction and validate is execute it and so if it's complex you got you got to pay in gas in Mina you do the transaction locally generate a proof so it can be as complex as you like the the trade-off is it'll take a bit longer before you can submit the proof but once you do there like a constant size um verification cost to that so you know you you don't have these huge spikes in in fees according to how complex the thing is you're doing it's it's it's pretty constant there is some variability based on traffic but you

know not not to the same scale um because it's EK there's optional privacy at the application layer so applications can Define like you know these parts of this transaction are private and so you don't dox yourself every time you know every time you do it um and there are still like RPC gateways like similar to infura in in Mina but they're um kind of um optional it's a lot it's a lot easier to run like EUR own node and soon there'll be like a well there is already a fully in browser node but it's not widely used yet so yeah it it sort of uses zero knowledge to like address some of the core issues you know that a lot of the you know know crypto Community

sees I think with blockchains today and you know and did it did it super early um you know which is which very cool yeah so early you've had to go through a name change ah yeah um yeah Koda was how it started um not a super interesting thing though it was just like it was too similar to some other product and it was early enough it was it was better to change so yeah cool so I think like the best way to understand what is like different and also mag about Mina is to like first like let's talk through how ethereum works right so I love the metaphor the world computer because like really what's going on in ethereum is you have the evm which is just like a

Computing space and then you send developers send their code into the space we call that a smart contract and then like in order for someone to interact with ethereum they're basically saying hey ethereum remember that code that was deployed before I want you to to run it I know it's going to cost computational resources to run it so here's some gas money that represents those computational resources run it within the ethereum place the answer with ethereum and then I will see what was ever placed in ethereum now that's not how it works with uh with Mina right right and and I guess I I don't want to give the punchline I'll leave it for you but can you talk a little bit about how the uh Paradigm is shifted and like

where the code lives and like I think that's really how you understand where the ZK comes into it all let me answer it in two parts right um the first part is that that model you described with ethereum right like that kind of broke down in terms of usability in terms of how um expensive that was right and that's what led to you know layer twos as the kind of scaling um solution and so to understand meaner it's maybe worth understanding like how a layer two on ethereum Works which is exactly the way you described in terms of say you deploy a smart contract you know someone execut the transaction um you know it gets executed but like what you what like the first the first thing to note there is

that um for most of the l2s today like they're centralized right so instead of having this you know half a million computers out there you know in the proof of stake it's like a trusted party so now you got to trust someone so you lost some of the trustless um and they execute the transactions um depending on whether it's a ZK rollup or an optimistic one they send the results of those transactions to a eum so in the layer 2 version of ethereum what's ethereum actually doing is it's just verifying some zero knowledge proofs from a from a layer two and saying like yep all those transactions check out checks out it it it looks good but still with lots of trade-offs so like it's interesting that like you know the way you described ethereum like it's still

used like that today people do do like execute smart contracts of course but like the volume has shifted to layer twos right and and and and the reason is it wasn't scalable in in the way it was now like the way that sort of layer 2 model is much more similar to mina um except that like Mina is a layer one and you're not having this um you know trusted intermediary that are all at various stages of decentralizing but no one has got close to like you know being there yet whereas Mina already is okay so like what do Ian what do I mean by that and what's what's different so um if you were in ethereum you deployed a contract to ethereum or you deployed it to a layer 2 in Mina you you um you're a smart contract developer you you create

a contract um and when you compile it you get a um a verification key which is um yeah kind of just a uh a way to like later prove that like some transaction was on this smart contract and you send that key to the chain so you're not sending the whole smart contract the whole code you're just sending uh sending a key and then um you know you create a like a website it um and uh that it's like a regular looking website you know people like log in they they do things and at some point they're going to like execute a transaction so when they click like you know go um instead of sending like their name and their balance and everything like we

described before V an RPC to ethereum to be executed um they have the contract that the developer wrote there in their in their browser they click go and the browser does incredibly complex mathematics to generate a zero knowledge proof that can cryptog cryptographically prove that I ran this contract I put in these inputs and here is the result and that might be like you know I I I I did a trade on a on a DEX and I had a high enough balance and you know I like I have a valid wallet that might be like what's you know you what's and then the proof goes to the chain and so the chain now there's no executing that transaction

all the chain is doing is verifying that the zero knowledge proof is valid and all it needs to do that is the proof from the browser and that verification key that the DAP developer created earlier and so it verifies the proof that's a constant size you know constant complexity operation an O of one operation um I think maybe it's I'm going to look bad close but but what you'll notice is that what I just described where it's like the proof goes to the chain the chain like verifies it and and using that distributed validator set um that was kind of like what ethereum does now for the L 2s but without a this without this trusted trade-off of having an intermediary like with Mina it's already a layer one with very distributed

validator set um you know in the full security of no one uh so yeah I I think I hit the main things right you could make your inputs private the compute happened locally only the proof was verified so you get way lower fees and you don't have the trusted intermediaries like you do in the the layer 2 world crypto as it was meant to be it's really incred you we we'll get back to this a little bit later but what what is very fascinating that we're seeing now across all most like legitimate L1 blockchains is like we're all kind of conversing converging on the same like endstate road map which is like you have your base L1 that is really about verification and you invest

as much resources in decentralized Trust lessness incredible neutrality and then we're finding that the magic of ZK allows us to move the actual work away from that L1 Network whether it's in the Mina Paradigm or the rollup Paradigm or what is very clearly coming the achain Paradigm the idea is like move the computation away and just use the magic of ZK to um like project the entire computation into the trustless space there's details but like I I agree there's some there's definitely some convergence there on like yeah the models that are finding to be to work so while we're still in this kind of like walk through Mina phase of the conversation one of the things that um is always like been an interesting

question for me and like one of Mina's key selling points is that it is not only like the world's most lightweight blockchain but um the the chain I guess or or the like the data load always stays at a constant I think it's like 22 kilobytes right now I was just watching a video about how that could drop down to 11 or uh but but my question is like how do you achieve you know a like growing Dynamic like constantly Changing State without like the you know the data you need to track uh changing with it very good question um let me yeah let me answer it in two parts like one is like obviously like the details of every

transaction that's ever executed like over hundreds of thousands of accounts and you know years of of State it's not all it's like the state itself isn't in that 22k right you can't say give me all of the details that were in this transaction um so it's not like uh it's not it's not magic magic some magic but not magic like what it does do though is allow you to um like generate a proof so as a user to say like hey like um I did this trade a year ago I traded this for this here's the proof that I sent to the blockchain and just with that 22k you can um verify whether I'm telling the truth right so it's like um it's enough

State there to um to verify that any any transactions that ever happened is included and as I it's not magic magic in that you can't you can't store terabytes of data in 22k but it's pretty incredible that you can verify that anything within this terabytes of data set does fit in that 22k and um there's a great video like or at least a um analogy our um our CEO uses actually where it's like it's kind of hard to believe but um it's like if we took a like if we took a Polaroid of this conversation and then like held up the the photo of that Polaroid and took another one you would see that like this conversation like was included in the state of the photo right and if we kept

taking them like the it's going to get harder and harder to see like what's in each po oid but you can kind of conceptualize that some part of the data of this conversation or the transaction is included in this this Polaroid and it's kind of like that like and that's where the math gets sort of magical but um you know that's the that's the sort of the 22k part and but the the reason that's really cool is that um you know Mina's um Mina has a state bridge that's that's getting pretty close to production actually and and what that means is that when that 22k gets synchronized regularly to ethereum then um anyone who's um like has a a distributed application in in ethereum a

Dap in ethereum but once privacy aspects they can like you know build that on Mina and the privac the Privacy Centric Parts happen on Mina and because that sort of that 22k state flows along back over in ethereum land if if the user wants to say like I did the trade or I did the vote or I like burned the nft or whatever it is they can supply the proof of that to the ethereum D and it can just look up that 22k and say is this proof in the 22k and if it is I'm good to go and so yeah it's this like bringing together the like the sort of privacy and magic of ZK over in meand to the you know the broader ethereum ecosystem which uh you know the bridge

is in production yet but that's going to be amazing uh when that lands so I really want to talk about um how Mina will like interact with ethereum in the end game and like where you see these two ecosystems uh ex coexisting but before we get there so what you just described makes a lot of sense but I have to ask like what in ethereum would be called the data availability question which is like let's say like okay so I have like this this uh let's say like private key on Mina I've done thousands of transactions I keep my entire net worth on Mina and then like my computer for whatever reason is incinerated and I still have my private key I get a new computer

I load in the private key I'm ready to boot up but like I don't have access to like all the state that I had like generated before I don't have access to the previous transactions um in ethereum that like today exists directly in the call data or we don't need to get into that but my question to you is like how does uh Mina like make sure that the data needed in order to generate these proofs is available for people that need to generate proof that's a good question it and it's one that kind of gets pushed to the application layer so like like ethereum there is some State on on Mina so like if if the state that you need to like recover is is on chain then you're

okay but like it's so limited that for like complex applications it's kind of it's kind of not and so yeah it's down to the application layer as to how they solve that and excuse me like that's in the past being something like ipfs um you know there's lots of Da Solutions coming out now and like the community is kind of you know figuring out like what's the like what's the what's the ideal so yeah I I I guess I'd say like um application specific and you know the community and the ecosystem are still working out like what's the what's the best approach there and without like uh picking any names or like committing Mina to a future road map do you just think it's realistic that like Mina ad adapts like the kind of modular data

availability Paradigm um and like might you know just send that data to somebody like a Celestia um is is that like kind of what an end game looks like yeah yes there will be there will be you know multiple da options for developers to choose from and yeah of course Celestia will be will be one of them yeah no no and again so got it it it just sounds like the answer to this is like whatever data availability Paradigm that's coming out of Cosmos and ethereum will find its counterpart in the Mina ecosystem yeah yeah I think so um so I I think like now would be a good uh part a good opportunity to Pivot the conversation into like the application layer and like

specifically I'd like to talk about um at least from your perspective or maybe from the research that o of one Labs has done like what are the use cases that you guys are like particularly excited about that Mina is going to be able to service like excellently from day one that kind of the rest of blockchain and especially non-z ZK blockchain is really not able to service good question um I guess I'd take it take it in two parts um one is like uh I think there's a tendency in the industry to like look at ZK and just say like ah privacy right and and it's and that's fair because it's like a big problem in blockchain

and it's kind of magical like how it does it and so privacy is a huge huge deal and and I'll talk about that in a second but like at least today if you put like your you know hard product hat on like in terms of capital deployed and like usage powered by ZK so far it's actually been in the scaling aspect like scaling compute and specifically like you know scaling compute for ethereum um you know like ZK rollup so this um concept that you can um compute like arbitrarily large amount like not unlimited but you can can compute large amounts of um you know data so execute a lot of instructions on a lot of data and

in the end just have a little proof um like that is a scaling kind of magic and um you know so that's that's been the biggest impact I would say of ZK like today and you know in Mina for many years and then in ethereum in more recently as the ZK rollups start to like get more and more um uh traction so um you know that's that's a big one the scaling the scaling thing and yeah that's um you know o1 you know does does some work in that area as well um but I think like the Privacy one is more interesting at the um at the application layer right and there's lots like you know you can go down the sort of list

like some of the ones I find well let's just say privacy in general right like that you can can execute a transaction on a blockchain without doxing yourself like that's that's kind of nice um you know other like cool things that I like um uh drug drug Discovery is one where not the discovery part but the the the proving so like if you're a like a um uh if you develop new drugs and you want to prove to a pharmaceutical company that you have a drug that like when um when that drug is put in this model of the body or whatever organism you know it's it's in that it has this effect and that you can prove that with having to give up all the details of your drug like

it's such a nice like fit of of ZK um you know I really like that one um you know but the all of the the the ones that get listed off I think are important like surveys and voting right like you know you want to be able to you know like vote without giving giving up um your identity sometimes same with surveys um you know all of the identity use cases right like why do I have to give up my passport like and and and whatever every time I want to interact with something um that's one there though where I do think that um there are two types of um entities there there is the sort of Crypton native like give me a ZK zkp and and I'm good you can do anything and and then there's the like

okay what are the regulations in my in my jurisdiction and is that zkp going to going to help if the law enforcement's knocking on the door and saying proof to me that You' verified this person it's like oh here here's a field element yeah um but you know no doubt that like this is the way sort of identity like should be done and it'll just take time for the other systems to catch up but you know being able to prove aspects of yourself without having to give up everything um you know oracle you mentioned um I know you said chain analysis not change link but um you know oracles is another good one right it's like um you know we have this such a convoluted system now of like bringing web 3 data to to Web Two

um both like with financial data where it's like have this whole separate proof of State Network that people put stake at risk to prove to to promise you know with their state that that they got this data and it's real and they make it available on Chain versus the Ser knowledge version where it's like here's proof that I got this data from this place right I got the price of Apple from the NASDAQ or like you know I got this trade feed from kren and Quin base and here's proof that I took every one of those trades um you know and put that in a um you know to calculate a price and that's what's on chain and you know one of our um Partners Dia is actually um you know doing this so yeah there's there's lots um I don't know we can we

can go deeper into one particular one but um you know there's yeah definitely a lot of use cases unlocked by ZK yeah well I'll just drop this in here now and we will get back to it uh at the end of our talk but to me this is a little bit in the weeds but um like to me the most interesting short-term thing for ZK is to like just go back and look at all the things that we did optimistically or like in in some sort of trusted mode while you know for scaling or whatever and just say like any of that if we rearchitecturing Mina from the start what would it look

like yeah yeah yeah no I mean I I honestly think that's fair and I think that if we were starting if vitalic was here like and we're ethereum wasn't created and we said like we got to create some sort of credibly neutral Computing space um like it would be so obvious that this would be the direction we're going um I think and to be fair like I kind of I kind of joke right because the reality is ZK was a was a like it existed as an academic paper like it wasn't a usable technology and you know like all of the projects out there especially ethereum you know went off and created something amazing with the tools they had and to their credit are like you know constantly evaluating

what what comes up and you know retrofitting the parts of the stack in in the order that they think is right so for sure so I think this is a good moment to like kind of zoom out a little bit or and maybe look forward and talk about like what is the endgame uh relationship between let's say we'll make a special section for ethereum and Bitcoin we'll make a special section for all other just like l1s that will come and go and then we have Mina like what what is the endgame State um once things stabilize a little bit like how does uh the technology You're Building interact with the world computer yeah good question um how to predict the future um I think like just to sort of yeah paint

the whole picture I think um um and I kind of hope Bitcoin sort of stays as it is right I know there's you know Innovation around the edges with it inscriptions and stuff but to me like it serves a very good use case of a kind of like you know a hard a hard money like not operated by a central party that you know print turns on the printer when when it needs to for political reasons I think like Bitcoin is is a good use case um as it is ethereum like uh you know it is called the world computer but um the direction it's going it's like it's not really the world's computer it's the world's like set you know settlement layer or or the world's like data availability layer which is a

terrible name because no one you know outside understands it but like that's kind of where it is going right like that um it it is the place where everybody else you know of you know the l2s um you know settle settle their data to so like you know is that still the world's computer or is it the world's database does it matter who cares like but you know yeah it I think um you know has sufficient gravity that like it's going to stay where it is in that super important like role but maybe less less compute gets done on ethereum like it gets done you know elsewhere and and and more it um you know just gets good and efficient at like verifying you know transactions from other um layer 2os and then layer 3es and and whatever else um

and Mina like minor I think is is two things I think it's minor is different enough in that like the ways we described earlier in terms of like you know the experience of um you know privacy of low fees of like you know being truly decentralized that it will have its own ecosystem of like daps and you know people that just do their things on on on Mina like you know and you know Mina um will also have l2s right where um if for a certain use case it's acceptable to um have some partial trust in a in a layer 2 operator to get like faster block times more throughput like whatever it else and they settle TOA like that same thing will will exist in the Mina ecosystem but I think like

the big part is you know ethereum you know it's going to be a long time if ever that it has true privacy right and so I think like any any app on ethereum or layer 2 that wants privacy in the sort of you know short or or medium term um mean is going to be a really good um choice for that and that you know succinct 22k um of data being available on ethereum you know does Bridge those worlds together like and you know then the once that's done too like token Bridges and Etc will will come they'll be like a standard bridging thing but the fact that you'd be able to um have applications in ethereum that have privacy like features that are implemented over on on in in Mina world I think that's going to be a really like

important part of of bridging those those worlds together yeah yeah no I I think that makes makes a lot of sense and if you'll indulge me for a moment I will sell you on the world computer vision and I I understand as a caveat I understand what I'm about to say should not be said to someone working on Min a protocol because you're already ahead of of ethereum but I understand ethereum as the world computer first starting with the evm being tur in complete right yes like it's resource constrained and it's not very good and it's not efficient at all but it is capable of doing doing anything a computer is capable of doing including your MacBook including like Google cloud player servers like if a

computer can do it ethereum can do it then we say like okay but ethereum is slow and terrible and expensive and like all the things and then the idea is okay like if we keep this space credibly neutral and slow but we have this magic technology of ZK which allows us to project outside Computing into the evm yes technically ethereum is not doing the computation but like you can call it the world computer because um it's it's not valid until it like kind of passes through ethereum and that is exactly what Mina protocol is doing I don't I don't disagree like you know it it I I think that's uh that's totally fair and my point was kind of like um you know

okay let's let's use this analogy so two types of computers an application server and a database server right like one runs lots of business one is really good at storing data right and today it's true ethereum can and does do both but the application serart it's getting real real really really expensive really quickly right and so that's why that has moved elsewhere and as you say it gets projected back and look even even verifying zero knowledge proofs and and and storing them that's still doing the work of a computer it just kind of looks more like a database right or more like a distributed Ledger accounting system and that's that's sort of all I meant was that it's its will computer role is like yeah becoming more

focused on the data availability and the verification rather than the computer itself but I agree it is thir world's computer no no for sure yeah for sure no and I don't uh mean to like take a counter position to you but just because I think it's helpful I I'd love to to tell you what I think um is the endgame of Mina and how it interacts with back with ethereum and then you tell me what you think but uh a couple weeks ago on this podcast we had uh the CTO and co-founder of space and time Scott disra sorry Scott um and one of the the conversation we had was um turns out that like if you are to actually do the verification of like a snark on ethereum

mainnet turns out like even at a full uh block 30 million gas like you can only verify like somewhere between like 10 and 15 snarks and so if we're really like believing in this world computer vision like there's a actual scaling problem at the L1 where it's like there's just not enough uh computation to verify everything we want to verify and so I said this to Scott and what he said back to me is like well obviously there's just like the way we solve this is not an oracle network but I'll just use it as a um similar frame but like this external network that is uh you know like completely tailor made to just being doing the ZK Pro the verification to like um blah blah blah blah blah but

as you describe mean and protocol and the ability for the 22 kilobytes to be like sent and live on an ethereum smart contract so that ethereum uh applications can like look into the Mina blockchain set and like make uh decisions based on that like to me that's the answer like we figured it out like we are going to need Mina in order to achieved like the ZK future of the world computer yeah now that's an interesting take I yeah I take each you with some of those things there right so one thing I've learned in crypto especially in cryptography right because I'm not a don't have a math cryptography background but um The Details Matter a

lot right so say you made a statement um you know 30 million gas you can verify about 15 SNS it's like well what type of SN like you know what's the like how was the snot created what was the proof system used and like how you how how efficient have has has the verification logic been optimized and the like you know the creation of the snar itself um and then more importantly though like so you get some you can it's very very easy in crypto to get orders of magnitude changes right like so that 15 can become 150 and 1500 and whatever very quickly like you know and things you didn't and and it moves faster than you expect the other one is like what um how much

computation does that one of 15 or one of 150 or one of 1500 how much computation does that snar prove right and this is where you know um folding incremental incremental VI verifiable compute like Mina like 22k snark shows that like that that snar is just the output of some arbitrarily large computation size so like yes there'll always be some limit to what the what the layer one can do but like if you kind of Unlimited in what you can represent in that snar then it kind of doesn't matter so yeah there's a there's a lot to unpack there um you know and I definitely don't take the draw that line

where it's like there's some computation limit on how many SNS you can verify therefore like some separate proof of state verif zero like no like I I don't I don't see that the the things more likely I think is you know the L1 continues like ethereum continues to add like snar friendly things like pre compiles um you know like it it the the EIP 4844 right like specifically makes it more efficient for um l2s to submit Blobs of transactions like there's a lot of things you could do to make it like super super super ZK friendly right and so yeah I I think we're a long way away from both getting to the we've fully optimized what ethereum can do in terms

of verifying SNS and that we' fully optimized like the creation of those stocks and how much compute can kind of be behind them and represented by by that eventual proof no fair point like totally accept the push back and I think um like both sides are totally valid like that welcome to the industry where 100x comes at you and hits you in the face without noticing and also like only one half of the equation is verification the other half is like what are you verifying and that can probably grow unbounded yeah plus it's easy to make a point when your other interviewer is not here right the the person you interview after me will just tear down my argument I yeah no no fair enough so um I we can keep going down this all day but I want

to reserve the last about 10 minutes here to talk about something very exciting which is the work that you guys are doing to upgrade the optimism stack so um hard pivot but would you uh tell us a little bit about what's going on with optimism and with uh o of one labs and I think risk zero is involved a little bit yeah yeah it's uh it's super cool happy happy to talk about it so um you know optimism um you know put out an RFP to um uh to like ZK prove the the op stack and there's a lot of detail um it's kind of hard to to talk at a high level but but at a we'll do our best at a high level like we talked about right there are optimistic Roll-Ups and there are ZK

rollups and and the ZK Roll-Ups have some um benefits in in the technology and you know um I think all of the optimistic rollups can see that like you know like they they could also benefit from from that technology but you know they already have ecosystems like they already have huge amounts of users and smart contracts and you know it's not like you just throw that away and so I guess what um you know what uh optimism is doing is like adding um what are called validity proofs to the to the optimistic proof so old model right it's like submit a batch of transactions someone can submit a proof to show that that there's wrong and those are called fault proofs right um at the same time why not provide a zero knowledge proof

in the same way that sort of ZK Roll-Ups do to say hey look like even if we give people another like seven days to submit like a fault proof like here is here is a zero knowledge proof right now A validity proof that the the last batch of blocks are all valid and they're executed according to the evm rules there was no funny business here it is and like you know that that maybe you can trust that and it's not going to happen overnight but like you've now got two different sort of um you know proof models and you know it's early days like that was an RFP you know both 01 and and risk zero um you know applied and the p is coming to a to a close actually and we're starting to talk about what what productionizing this might might look like but I guess the cool thing is that

you know it's Mina's um proof system which is called kimchi and out the recursion layer is called pickles that we already had this amazing proof system you know that produces the um proofs for Mina and is able to you know um provide recursion to then take that and apply it to like another another chains logic to be able to yeah prove these transactions like you know compress it all down into like one small proof and then and then and then send that off and so you know like um it was just really um it was it was a really nice coincidence in a way that like that same text stack um that we could use um for mina we can use for um op um but another like really cool part about it

is um the way we didn't actually talk about this but the way you write um smart contracts for mina is using typescript um which is um very cool because ZKA can be like kind of mathy if you use some of the like older School libraries whereas we make it a bit more higher level and so that works great for a lot of people right to be able to to to write smart contracts in in typescript it's not that different from doing in solidity like arguably easier in a way because you have types and high levels um but like no doubt there are folks that are going to want to build applications um that settle to mina that that aren't written in typescript to like run in the browser but that are just like general purpose programs and you know part of this op work was

developing a what's called a ZK VM so a a virtual machine that can execute code in our case it's a myips VM um and the code it's executing is optimism's program but it could be any um any exec cable basically that compiles down to myths so you know even though we're using the you know Mina core stack to be able to build something interesting for um ethereum and or for optimism it's going to reap such huge rewards to for the minor ecosystem because we're now on the path to you know being able to run any arbitrary program generate a proof and have that settled demeanor and so it just opens up the like when we talked about you know that arbitrarily large amounts of compute that get hidden

behind like a single snark um you know that's that's a very a very cool um thing and it's still early days because you know the the focus has been on getting getting the validity proofs for optimism to work but you know once we um sort of redirect to make those proofs directly um be able to settle on Mina um it's going to be another like big big happy upgrade for the Mina protocol so we're looking forward to that do you foresee this specifically with the work you're doing for the optimism stack do you foresee this as a um trans a long-term transition from fraud proofs to ZK with this intermediate time where both are like kind of being used um or is this like

really optimism staying like for as its core uh infrastructure an optimistic rollup but having these like new ZK tools that are available to developers yeah that's a that's a great question obviously I can't speak for you know Labs or the foundation or or the community and what they're going to like vote for what what what optimism have published so far it's a it's an end not a not a replacement it's that fault proofs and validity proofs is better than just fault proofs and that like over time as you know the industry the community like is able to assess like the different security and trade-offs of of those things like you know maybe one gets trusted more or less than the other

um and that could mean for example that you know you don't have to to wait 7 days for a withdrawal if there's a valid zero knowledge proof and like either um you know either a bridge trust that sufficiently because for X years there's never been a like a A reorg or or a fault a fault prooof submitted when there was a valid um zkp validity proof there but like that takes time you know and they don't know the future we don't know the future but I think um what's clear is yeah adding adding validity proofs in the form of ZK like just is another sort of security tool in the toolbox you know for um for for for optimistic rollups yeah well it's not even just a security tool right it's

actually like a very useful developer hook and like the big unlock is obviously in bridging but I can think of like many other applications where just being able to verify with just the block header that something is valid like allows you to unlock like really interesting stuff on either a side chain or a side program or whatever and so I think like we can talk about like the architecture of rollups um but I don't know I mean this is just like the clear path forward of like not only what we need in order to get more secure systems but like in order for the optimism to be able to take advant advantage of all this like incredible computational magic that's coming out of ZK world yeah I I

think part of what you're keying into there is um you know which which optimism has spoken about a bit um you know their version is this super chain messaging right that like layer twos can like talk to each other and um if you have a a valid proof that like um you know up until this block you can prove that you're you're up to date and and and so you know recipient chain with this proof can you please go and do go and do something that like and that's still a while away but like that unlocks something pretty pretty interesting and you talked about convergence you know of like um blockchain into a model like ZK Sync has the hyper chain model right and if you look at it the diagram is exactly the same um you know that you have

within the layer twos you have sort of little ecosystems and then they all they can communicate via ethereum but if you do it um you know within within the super chain or the hyper chain like that it's more efficient so um yeah zkps also you know unleashing that that area of innovation super interesting it's not only them right it's starkware it's polygon K evm it's yeah this is clearly where we're skating and I'm so excited for you guys to like get the first ZK proof of um of optimism because I mean like literally if you're a believer in the math which I know you are I know I am and I'm pretty sure that everyone listening to this podcast is like if you believe the math the second that first

ZK proof of optimism is done like the whole Pro fraud proof system becomes like a necessary crutch of trust that we need in order to transition to like the next phase but like I think the second you guys deploy might be like the moment that we Herald the end of the optimistic rollup era yeah it's going to be interesting because like I said like the the way optimism have talked about it in their blogs like and I think this is probably following in ethereum's footsteps is you know when there's so much value locked up in your system like you have to trade very carefully and of course they will but I agree with you is that like once those once those validity proofs start landing and like people try to break them and and they can't and

it's like well why are we waiting why are we waiting the seven days like I think it's going to it's going to you know like you progress the conversation like quicker it's like mathematically it's impossible to generate a fraud proof that is valid or I guess a fraud challenge that is valid so like why bother waiting yeah yeah and you know it will take time before security researchers are happy the community is Happy whatever but I agree when people see Over the Horizon like ah like instan is withdrawals and uh like oh permissionless like you know instan is withdrawals like the pressure to get there and it's like okay what do we got to do to get there like how how many audits do we need like how many like how big is the bug Bounty going to be like yeah I think things will I think we'll

probably accelerate too yeah and then once we unlock single slot finality it's like the whole world is uh changed again but that's uh the start of a rabbit hole that we don't have time for so Steve thank you so much for um for sitting down and sharing everything that you did about Mina about o of one and about your background I mean so eye opening for me and like really helps me understand just how like the the greater World computer like bigger than ethereum but the blockchain community is like really skating towards the same goal so man I really appreciate it thank you and before I let you go can you uh just let the audience know how they can find you how they can learn more about Mina and if they found themselves like really

interested and want to get involved what should they do yeah great so um I'm at ZK SJP on on Twitter um you know metool docomo get started on on writing smart contracts in typescript that um you know that have full privacy uh you know we're all in Discord both o one Labs the Mina Foundation um you know just search for the Mina Discord uh all of your questions will be answered there's a super helpful Super Active community um the foundation runs um like Grant programs like so if you know you just got a cool idea and you just just just need a few thousand bucks to get started like you know there's constantly rolling um programs um one's called mea

Navigators one's called ZK ignite um so yeah lots of ways to get started I think yeah like Discord Twitter and and meoc are good places to startop and how much do you need to know about zero knowledge in order to get started um it like a a lot like less than on any other ecosystem let me say that so the fact that you can develop in typescripts right and the docs you know take you through the sort of what you need to know to get there there's learning it's a new technology you got to learn something but um yeah it's as it's as easy as it can be and actually any developer like can can get started with with with a basic app and then when you you know when you go to the next level the docs and the the ecosystem are there to support well man again I I can't

thank you enough and I encourage anyone who's interested in developing ZK or developing ZK based applications like spend some time with Mina like they are the ogs of how ZK will interact with blockchain and um man I'm just so excited to see what comes next so Steve thank you so much and uh have a good day yeah you too thanks for having me Rex h