How Safe Sequencing Will Power the Appchain Future with Firewall

Transcript

we're taking your podcast virginity man here's a secret that uh nobody wants to say aloud yeah I realized that was a tough one but I think that was a pretty good answer respectfully Devon like who are you to decide what is you know right and wrong welcome back to the rollup for today's video we sit down with Devon from firewall to talk about a concept called safe sequencing Something That We're particularly interested because of the idea of bringing on institutions to rollups right now we think a lot of Institutions will probably come directly to ethereum L1 as we've seen with black rock and other we think that if we can push forward higher level of security on the sequencer level for rollups we can have this Capital be Al today we talk about the vision for firewall Devon story getting in crypto what safe sequencing

is the trade-offs any centralization risks added complexities and more on a technical level hope you guys enjoy let's Jump Right In after a quick word from some of our awesome Partners vertex Edge is now live on say Network say users can now trade 50 plus spot and per markets with the unified order book liquidity of vertex Edge across arbitrum blast mantle and SE the next 6 weeks 850,000 say tokens will be up for grabs as trading rewards every single week exclusively on say join vertex on say Network now on vertex Edge

that's vertex protocol welcome aboard on our journey into the singularity this is the endgame road map for fracks and fractal the new L2 out of the frax ecosystem that is totally based on fractal scaling fractal the modular L2 blockchain built by fra with modular and fractal scaling in mind fractal scaling is based on the mandal BR set of fractal images and explains how l2s become l3s and the entire app chain life cycle goes through its Evolution the frax ecosystem is made of some of the

top builders in the space and has stemmed from its OG days of the frax stable coin it is now in the LST space the LRT space and this fraxel L2 joins the existing Suite of fra products including fra swap and frax lend you can see frax faery the bridge to go from any chain arbitrum optimism or main net into the fraxel main net Bridge your tokens and bring your FRA assets into the fractal ecosystem join today to get the best out of fraxel you are still early join today take part in our Expedition

into the singularity what's going on guys welcome back to the rollup today with my co-host Rob as always talking today about kind of building a safer rollup ecosystem with Devin from firewall um we have a ton to go over about talking about some of the kind of problems in the space how firewall is going to address some of these um and just kind of like generally building a better safer onchain ecosystem I think it's something that we can all kind of rally behind just got to figure out how we do it so excited for that today Deon I think it's your first pod so uh welcome I'm glad that it's that we're taking your podcast virginity man it's great awesome yeah really excited to be

here thanks for obviously taking the time and excited I mean we haven't really spoken too much about farall publicly yet um so this is a great opportunity and excited to start sharing what we our belief and vision is for the future yeah I mean we I think we got introduced to you from Smokey from bearing you guys are like friends from from something right yeah yeah Smokey is like one of our Angels um obviously great guy and love the bar bar Community um and of course you know clear synergies between what we're trying to build and what they're doing with proof of liquidity so um yes and he's also a fellow Canadian so you know we we viive on many levels dude what do you think about mert uh leaving

Canada I don't know if you saw that if you know who merch is he's like he's full Canada bear now honestly I can't blame him man I haven't I haven't even lived in can for years I just came back actually I don't blame him honestly can is a bit a a bit of an interesting country a lot of stopan things going on here especially if you believe in crypto ideologies but um I don't honestly I lived in Dubai the last few years right I don't I don't blame him like I get what he's saying you know but sometimes you know the fam is it's hard to leave them so got to come back uh home and um yeah I don't fault them put that way nice yeah I'm I'm all my family is Canadian as well but uh anyways I think

a great way to kick this off Rob is because we're gonna be talking about some of these kind of uh problems of onchain security and defi um I mean first of all Devin maybe you want to just give a really brief intro about yourself outside of like being a uh Canadian who's been living in Dubai yeah for sure uh you know backgrounds in comai um my focus like Ai and ml I got into crypto in 2017 um I fell in love with smart contracts that's why I entered this space I was already really into personal finance ftech whatever um and so smart contracts just made sense to me um I graduated in later 2018 and I ended up joining a stake. us

where I was a founding engineer first employee it's kind of all things like ethereum Pro a state consensus help build the early layer of def5 Primitives um and a fast forward you know stake was acquired by Kraken at the end of 2021 for it was like one of the biggest Acquisitions in Industry history um and still is to date um there I helped boost trp an MB team at Kraken um i' say like a year and a bit before I left Stak after about four and a half years um to pursue firewell fully with my co-founder Sam and Sam and I actually met at Stak he joined a few months after me in 2019 um and we've done a lot of fun things together and firewall just felt like the right thing to commit our our

next years to and um obviously a very important sort of vision and Mission that we're on what what were like the major breakthroughs that you had what when you were going through Stak and then and then getting acquired by Kraken and you work in me was there any light bulb moments that you had along the way yeah I think the the big like reason I even joined Stak initially was because I got defi it just made sense and you know I got like a great experience there like I built the first uh defi arator from like 2018 to 2021 um we had like Stak itself is very focused on like institutional adoption institutional clients essentially um and we were trying to be like the Gateway for trafi funds to you know you know hold crypto assets earn

yield we' stake their assets for them we take care of all that so I got like a couple good things from that one was like understand the challenges of an application Builder especially in defi like the challenges you face um and I realized that a lot of those underlying issues actually come from security Paradigm that smart contracts live in the second thing was yeah getting to work that closely on systems and like talking to these institutions for the you know last five six years and what they actually value and care about and how they would actually evolve into deploying capital on chain into defy what they need to see um and yeah I those like a lot of those things formed you know kind of the first principles that we founded firewall on and how we want to like actually scale smart

contract technology to be useful Beyond like our current Echo bubble today nice yeah it must have been interesting being a part of a company that uh got acquired Chang in manag it was a fun experience like um again like kraken's a great company of course it's like a very different like shift from like early stage startup to you know giant sort of company Kraken has a good job job balancing that um and you they were definitely receptive to like fun ideas like an any team and St on which is cool um did you have any like any like Equity as the first founding engineer or like how like I mean I don't know that's yeah yeah nice cool for sure I had I had equity and like it was my first it was actually my

first job you know after graduating so that was like your first exit kind of yeah literally yeah first job first exit you know I never actually when I joined Stak I didn't expect to stay there for you know as long as I did but it just made sense and I just fell in love with crypto more and more as I stayed there of course we were very successful business um and yeah I learned a lot honestly I mean I worked closely with the founders there uh they gave me the kind of the blueprint to to venture success and um obiously hoping to replicate that here but definitely a fun experience get Acquired and um a lot of cool things we've done I just remember like the Miami Retreat the after the acquisition it was just like crazy oh really yeah I think I would have chosen

like mikos or something or IA but Miami is good as well enough yeah they trying to keep it local okay TR to boost GDP yeah nice yeah I mean I definitely I will'll dive into firewall and and and and the why but but still like one more question on Stak like why do you think it was so successful and enough so to be acquired you know when you look at so many different startups out here and you know a lot of them raise don't ever launch anything you know they were successful enough in prod to get acquired why why do you think that was it honestly just comes from like caring about what actually brings value right I think there's a lot of exciting things

in crypto um and a big part of like the reason like things change every day in the ecosystem is because there's a lot of funding towards new innovative ideas that you know don't actually have any real pmf today but they might in the you know sometime in the future um Stakes really focused on more of like a traditional business model honestly when it came to hey like here's a simple problem you know people love love yield investors love yield especially back then in that era of like Zer it was basically you know Zer right the zero interest rate um rat from the Federal Reserve at that time people are dying for yield from trafi um so like being able to like model and sell like crypto staking yield as like this fixed income product just made a lot of sense and you

just where the money is right it's like okay yeah there's billions of dollars these guys hold they want to earn yield we can just make an easy business model make it super easy for these guys on the tentacle side we abstract everything for them you just collect some spread on that um and it's it's like a very fly whe business you know the more assets you bring into your your business um the more other sort of big clients and big Trad five funds will say hey like XYZ like panta using stake to stake all these assets they must be trust you know trustful and like we can like stake there as well it's just easy and it's just like that flywheel of liquidity um that's something I definitely learned is like how like the fly whe of liquidity is all that matters especially in you

think about like Define applications that are financialized um and how you create that is is obviously important so yeah we learned that a lot with network effects nowadays um and all of these rollup clusters have some form of network effects uh and it's it's almost like a like everybody wins you bring more people in flywheel upwards positive impact um so yeah that makes that makes a ton of sense and and so you guys were basically offering like portals for Trad five funds and then just like making it very easy for them to stake into protocols yeah exactly crypto funds trafi funds um like I remember one of the big things stake did from day one strategically at least was you know they had they brought on Venture from that

panta and like all these other big funds that held lot staked assets they became their first customers and um again that's just like how you build trust in a in a brand like that right um and so yeah yeah basically a portal for these funds to enter crypto essentially and see something that they can invest in really yeah very cool you know ultimately you ended up leaving that that uh and and going into firewall what were some of the issues that you saw and was like you know it doesn't really make sense for stage uh for state to you know kind of solve these issues um but you notice them and then and then they sort of uh influence your journey into the next venture yeah definitely like again

like I saw like there's a lot of Tailwinds even today you see on the market of like uh yeah trfi wants to enter the space there's a lot of regulatory adoption sort of tailin happening you know ETS Etc policies are shifting um that's all exciting stuff but the end of the day uh you look at the actual like what we have here today um people like complain there's a big meta right now about like obviously overinvestment into like infrastructure and um a lot of complaining about like that and uh everyone wants to focus on applications and like trust me I'm like right there with them like I want to see applications like uh be useful in crypto but you look at the fundamental issue it's like um if you try to explain how blockchain systems work you know today to someone who's outside of crypto

they'll think you're like crazy because it just doesn't make sense to them like oh really like you send a transaction or someone someone steals your money you can never get it back like it's it's irreversible there's like nothing that stops this from happening like there's no filtering mechanism um and that's kind of what I was like okay like it's nice that we're starting to see this adoption right on like the financialized side of of um like really decentralized side of like crypto but uh the decentralized side the onchain side is lacking a lot of that adoption because again of smart contract risk um and that's what made me realize hey yeah like we need to like shift the the Paradigm Shift required to like bring these guys onto onchain as well not just

like into centralized crypto um it needs to be built and that's an architecture piece right that's missing today and um that's really what inspired me is like hey there's no way that crypto is gonna and like smart contracts are going to like achieve the potential uh for adoption that they can unless this is fixed it's like a fundamental issue that still exists today so and I think one thing that's particularly interesting just to add on to that about the it's not just about the smart contract risk from what I've heard from different funds and like you know for example like uh Perpetual exchanges wanting to onboard different like liquid funds into their you know to be LPS and to trade and stuff and to Market make it's also about the fact that that these

regulatory ENT these regulated entities cannot interact on chain with uh non kyc AML addresses so like for example I'm pretty sure I'm rocking the ZK syn shirt today I'm pretty sure ZK sync did something or at least they have been working on it behind the scenes such that there's addresses that are not able to deploy funds into or to bridge into ZK sync in any capacity or at least through their canonical bridge that are registered with this list that that's blacklisted and so there's this software that every block it or like every second it Updates this list and if one of these blacklisted addresses sends funds to any other address or anything they they they

add that that uh wallet address and then it automatically gets blocked firewalled basically um and so that's kind of another another uh kind of situation happening where these some of these perpetuals exchanges are thinking about okay we should just launch our own chain like our own app chain bring this software in so the only people that are into this that can access our products are are basically non non blacklisted users and that gets them in like one step of the direction closer to like getting that adoption but they would still theoretically need to like have a a kyc account such that every user is kyc not just not just like uh innocent to proven guilty but

completely also uh identified and so that is also an interesting point that I think if you guys can figure out a way to to partner with some projects or some teams that are that you guys can provide this kind of smart contract stuff and we'll get into what you guys do this kind of protection and then also a team that does kind of this more like modering AML stuff not sure if you guys do that as well but yeah I I just wanted to to bring that in because I was having a conversation with builders in New York City last week for something and uh they were bringing it up trying to figure out how to get around that yeah no that's a good point um I think like I I don't know if it was coinbase with somebody who's trying to bring like their K from

their exchange like on chain to and like in a DK manner uh it's like something actually Sam my co-founder we talked about last summer is like this is a cool like you know side product but we ended up you know it's not our focus and yeah rather part with somebody who is focused on that fully uh and like focus on her own lane um I totally agree that's part of the issue and that kind of like ties into a lot of like what we can talk about later about like um the regulatory side of like this whole challenge that we face as an industry how do we actually make these this like 99% of the capital from like traditional markets feel comfortable because like again like crypto offer and smart contracts offer better rails it's more efficient they should want to use this and they do want

to use this stuff but there's just like these few road blocks to like work out still yeah I mean smart contract risk and Regulatory risk are two significant ones um and and the smart contract risk conversation has always been here and and it's resurfacing again it's this idea that yeah in this in this case it's happening on bitcoin quite a bit nowadays where people you know are holding their Bitcoin we have all these Bitcoin layer twos you know obviously we have R Bitcoin people don't want to take that smart contract risk they don't don't necessarily even want to operate with Native Bitcoin whether it's on the Bitcoin L1 or or an L2 they just want to keep it hold it store it uh huddle and and so you know that ever since ethereum

started like that that's been the conversation of of how we can mitigate these smart contract risks to get the most expressivity out of Finance use on chain rails but take away the the risk of of smart contract bugs and uh and also back doors and all of this stuff so um yeah I mean maybe this is a good time to get into like what what is firewall what you guys do and then and then you know how smart contract risk kind of has been has been a plague in the industry so far and how you guys are helping to mitigate that yeah yeah definitely um yeah so firewall's mission is to make smart contract technology safe um we're on this road map this is what helps

differentiate our solution and on this road map to zero exploits in the industry and today crypto networks as I mentioned right they operate like the early Internet it's essentially naked with no filtering or any protection if you send a message or transaction that meets some basic cryptographic checks it's accepted in the change database like there's nothing there that really filters or stops anything it's it's fully exposed right um so firewall is like this new critically missing piece of the web three stack it's not custodial um it's a trustless security layer to plug in enhance a smart contract Network execution consensus layer um that essentially just acts as a filtering mechanism for malicious actions um I think like today's cycle is

very focused on speed which is important but it's just not enough right um just speed is similar to like a race car no breakes it's useless in the greater context of utility um what I often say is like you know I it's like cool I can watch my $50 million get stolen in two milliseconds versus 200 milliseconds it's like you know it it doesn't actually add any value in that sense when we think about what is the goal of this industry and this technology um so at farall we uniquely believe the future of block space is speed time security um which is an architecture that we're pioneering for change the plugin and use uh and so i' say simply put firewall um we're we're building safe block space today block

space is unsafe for onchain financial applications and we believe that crypto's utility cannot scale beyond our current Echo chamber without fixing this issue and um the vision I'd say is in three years every smart contract network will be farall epic yeah I mean I I think you know no one is really no one's really gone gone around and like uh not like everyone knows that like there's been plenty of hacks I think it was4 billion dollars mostly in bridges and so there's also like chains that get hacked that go down liveness is an issue security is an issue multisig get get broken into sometimes it's physical hard that gets that gets hacked sometimes

it's smart contracts that have a software bug but like when and so I think everyone you know people in our our audience are like probably Vaguely Familiar with kind of like the way that smart contracts are set up nowadays certain backdoors multisig potential like uh limitations and security weaknesses what does what does a a a contract that is firewalled how does that differ from a contract that's not firewalled and why why does it make it safer yeah definitely we think about the evolution of Security Solutions in web 3 um yeah I it's informs like the answer to this uh you know a started like audits um then bug bounties kind of came

along uh then competitive Audits and uh obviously some forms of like smart contract insurance which in my opinion don't really work today um and then I'd say a couple years ago is when external monitoring and these other sort of detection tools became a thing i' say like more recently predeployment tooling such as formal verification became a little more popular obviously with like the birth of or like the evangelization of like the move ecosystem and so on um and these are all good and necessary but as you mentioned like not enough uh clearly as billions of dollars are exploited year-over-year from Smart contracts when it comes to firewall and like a firewall smart contract our solution is based on Dynamic analysis which yeah I can dive into like

more later but essentially we we effectively act as a runtime solution as part of the chain itself um and this shift in solution architecture offers new benefits um namely we can maintain speed we can also uh instrument essentially the client with the node uh the blockchain node with more information that lets our detection be smarter than if it was just you know built into the smart contract itself or uh if we're doing like predeployment checks the the big issue with security today is that that um most of it most of the solutions just they're faulty to human error right um and yeah like from the dawn of like smart contract like

software essentially there's always been bugs in in in software and um that's not going to change with smart contracts and of course the issue being that these things are very these software programs are very financialized they're completely exposed and the environment they run is just not friendly and so we're trying to make that environment a little bit more friendly by making it the environment self dynamically adaptive to you know what's happening um at runtime yeah yeah it's interesting it's um there's like we've seen quite a few different solutions for like stopping exploits before they happen or like responding after they happen and like preventing further kind of uh uh kind of

Downstream effects which I think like the downstream effects are like like there's these broader implications here for communities for protocols for the the entire kind of onchain user base you know if you get hacked like you're not really going to want to stay around you know if you're come on and you been promised this idea of crypto being like this extremely safe and secure Next Level Financial Ledger asset Ledger then you know getting exploited is not really a great a great thing I mean it's happened to everybody but I think like I think it there's also room here for and I think we've seen this where there's potential for more regulatory scrutiny we don't get this kind of get our [ __ ] together here right which could eventually maybe stifle some Innovation

you know just in the space and so I think that like getting this getting this problem solved right is going to be something that is like a a big Pro a big a big solution and like a big a big task that I think a lot of people are kind of working towards and teams will likely kind of collaborate to make it so and um yeah I mean I remember like I mean just the other day like for example something something silly like I sent Rob the wrong I sent Rob the usdc contract address to send a payment to and he was like bro like I almost just sent this like these types of things like that's more of like a ux thing but like I've I've clicked links and and had I had a hot wallet drained Robbie and I had some friends who had a lot of money drained I

we've all kind of had like some some problems with like the onchain ux but I think that like having that solution in in the more uh immediate term that's easily accessible for for rollup users is going to be something that uh that I think that will kind of take over or solve those problems that that I mentioned with regards to some of those implications of uh of this kind of state of block space as we know it now yeah just to add on to that because I do want to get into like how we can prevent that but just a little bit more about like why it's so important is is yeah because when some a new entry comes into the space as Andy just said like you know they they get disillusioned but I think to the most part like you know we we

collectively tell this story worry about not your keys not your coins the bank's money isn't in the bank it's loaned out to another bank your money isn't actually there come to the blockchain your money's actually here it's more safe it's more secure yeah and then they come on and it gets hacked immediately and it's just disingenuous it's like oh no like my money's not here anyway like it was safer in the bank or on a centralized exchange when I can hit forgot password so yeah H generally how do we prevent you know some of these things so that it is safer on chain yeah I mean here's a secret that uh nobody wants to S loud though you've generally referred to it but it's just like nobody with any semblance of risk management is deploying um you know in size to Smart

contracts for the exact reason you just said and um again just as I mentioned earlier like I don't even want people I don't even want to tell people to use some of these products because they're not safe um and yeah as you mentioned we all have friends and we've done ourselves like we've all lost like money in these things like in Defi and so on right and it's fine like a little bit of [ __ ] like [ __ ] money you're kind of like throwing around like having fun with right but you think about like actually trying to build a real alternative Financial system that's not like [ __ ] you [ __ ] around money basically that's real money that like people need and how can you how can you do that on the basis of like yeah losing all my funds in like

one transaction um yeah it's just not you know right and and that's like why you know we founded firewall of course and we kind of birthed our road map out of hey how do we get to zerox um and as you mentioned like um you know people are honestly scar [ __ ] lless of smart contract risks I've spoken to like countless individuals and funds who hold crypto assets would love to deploy assets on chain and maybe they still do deploy some yeah percentage of their those assets into smart contracts but the point is they'd rather take on centralized exchange custody risk and smart contract custody risk um and you know which is like the biggest group of people sorry to kind of interject that

group of people that are on centralized exchanges is what we need to come on chain like we need to get that segment on chain before we can even go think about like new segments and and I I was speaking to Kane the original founder of synthetics at um I think in Brussels invest beautiful and so like we were talking about like okay like he's building infinex or something right like like this really cool kind of abstracted experience for Defi and we we were just talking like it's not like defi against trafi right now it's like Defi and onchain against like binance but like yes binance brings them in for us but that segment of users we got to get them on chain and so let's talk about that

let's talk about this this like safe sequencer that you guys are building so obviously where the rollup I think a lot of people watching this probably some sort of Builders developers you know more or less like sophisticated retail people that care about the tech so you know if I'm a rollup Builder um you know why should I care about a safe sequencer model that you guys are proposing and kind of what what even is a safe sequencer and how does your kind of like AI background uh allow some of these machine learnings to kind of come into play to enable the safe sequencer to work yeah definitely um so start the safe sequencer kind what it is a little bit more um you know the safe sequencer

is a sequencer client uh it's integrated with AI and patent recognition to filter out smart contract exploits at block construction for a role been it's built to be extremely fast and stop M transactions while remaining trustless and not custodial um and a big part of our product kind of experience that we wanted to to make was like hey let's make this Plug and Play like it shouldn't cost or like put a lot of effort on the the roll up for the chain operator side to like adopt this solution um and so we started by modifying the op stack we forked the op stack um built our initial sequencer integration into that um upon the rest client and

um yeah from like U I don't even know how to put it simpler than this I mean to to roll to run the firewall like Solution on your rollup you just run like our sequencer instead of like the vanilla op Stacks sequencer we haven't uh we've only enhanced basically the client we haven't actually modified or made it incompatible any sense or any way uh um and that's kind of the yeah the TDR uh in terms of like why as a rup you'd want this um like to blandly say it like RS today are commoditized like you this is also part of like the modular thesis right like we expect these things to be commoditized that's how you creating cheap block space c creating fast transactions making that like common place is how you actually unlock a

better user experience to the end user um that's how you yeah unlock like the The Innovation that we're starting to see across like um this layer of rollups and the issue though that poses to like individual rollup you know chain operators is how do I differentiate from like the guy over there or the team over there and again like today like cheap transactions speed it's all like commoditized um firewall offers a way for you to differentiate and enhance your block space um to make it easier to attract liquidity essentially that's like what we're all about um and I think thing that's important to like call out here is like we we optimize and differentiate on an aspect that users

actually give a [ __ ] about uh which is yeah their asset safety um and so we think about like yeah kind of how we are entered into the sequencer um ecosystem and also there like different types of sequencers out there um that's like one thing I always stick to is like our um foothold that we've created is around security because we've talked to the users we know what they care about this is how you get more liquidity onto your rollup like make it safer it's that simple so yep and so is it like a a rapper like Okay so let's let's put this in Practical terms I'm arbitrum I'm running arbitrum one biggest robs in the world um how like how does do I use safe

sequencer to in a non-custodial and remain my key properties of basically like verifiability integrity liveness how do I how do I get the firewall safe sequencer like I'm thinking about it as like I have a sequencer it's a bunch of techol it's a bunch of tech in a kind of in a warehouse basically like hardware and then you wrap this software like code around it basically to to keep it but like how does that how does that like work to keep those like maintain those core properties of blockchain yeah definitely um yeah hly think it less of a rapper and more of like a direct like penetration in a sense of the client itself so like we're buried right in there like firewall um like we firewall

like the chain as and as you're envisioning it is like a wrapper around the chain itself on the technical side it's more like Hey we're right in there with you like we're integrated into your your client essentially um which unlocks a lot of cool stuff um for us on the the detection side but um yeah in terms of like how we like again like we are cryp like we're a crypto company we are crypto Founders we're crypto native um we believe in the ideologies of trustless um decentralization and so on right and everything we've built is is based on those principles um and so how we adhere to like a trustless product nature um I'd say it's effectively two steps in something that we are innovating on um you know the first step is essentially you know you want to

prove the detection capabilities of the AI models so you run like a back test essentially create this like provable reputation for the model saying hey this thing will block this this this these are the exploits will block this is what it won't block um and like make it very transparent to see like what what's the false positive rate what's the you know true negative rate um and the Second Step from that is essentially to prove that the firewall like the firewall sequencer the state sequencer is running those models uh in production and when you put like those two steps together you've created this provable computation basically system where users can transparently see hey this transaction was provely shown to be malicious and here's why

um and you guys like ZK for that right sorry yeah yeah exactly and it's kind of like this kind of starts relating to why we actually chose to build the fireal evm the safe sequencer around like a rust based stack in like off revm and W um because you know essentially when like well at least one of the more interesting things of being rust based is we get out of the box of compatibility with approvable computation tools and zkm from teams such as rero and cinct right so we basically integrated with those teams and yeah uh or those like tools and um that's how we're like that's like one of the Beautiful Things honestly when we see like this like this like Innovation on like the ZK layer um it's like yeah like some some of the stuff that we're

building today no one has done today in the detection security space because it is it wasn't even possible honestly or like the ability to do so was so hard before like we're not a team that's experts on the ZK nature of this stuff but they've made the tool so easy to use that we can create better Solutions like hey yeah we can prove Pro to a user that this is a algorithm that said this is malicious for this reason um and there's a trustless nature there it's not like Sam or I are like you know putting our hand on the button and saying hey that's we're we're gonna block that you know like it's impossible we don't have that control the system okay so so the firewall evm has it has its own set of like a there's some sort of like unique consensus mechanism here basically that

you guys have also built that that contributes to the verifiability of the safe sequencer yeah I'd say like when we think about our um road map so our go to market is a say squater to start it will be a single client's nature it won't be calized um but it will be trustless and it will be um neutral essentially because we will have the approval computation aspects to it um how we're kind of enhancing that the biggest kind of challenge I'd say or like trade-off that we make is like we're essentially introducing this like layer of censorship right it's like hey we're saying that this this thing here is delicious we're not going to let it be included in the state of this chain because it's trying to steal money um and that kind of burds like these

peripheral pieces of our ecosystem that we're we're developing past our our initial like launch product launch um and so again the nice thing is when you've made everything provable if you're saying you can prove what an exploit is you can inherently then prove what is considered actual censorship that um so if you're a user saying hey I got I got you know sensored but I wasn't attacking this chain here's all approval computation to prove that and like you can now get recompensated you can get basically your transaction included uh and that's like that secondary system that we are building essentially it's like that's where our Network kind of comes in like that's where consensus comes in essentially yes I see where your me background was stake comes into where like you're basically

building safe blocks over time yeah exactly and like because where these exploits happen is is basically on the Block building like like in like like somebody does transaction an exploitive or malicious act it's the block Builder and then like the proposer and and then the Builder in the ethereum land that basically makes this happen that confirms this on chain so you guys are just getting in between is it in between the build the proposer and the Builder where firewall comes in or like can you maybe explain on the Block building how firewall evm uh comes in there yeah definitely um so just the a transaction will come in like normal um sent to whatever

and at least on the sequence side right it's like honestly just one client right now for most rollups um that comes in it's like in the mle right whatever you want to call that um that is you know consumed by the execution environment of the evm it runs through that that's all normal right that's all part of our sequence or too the added step is that when we execute that transaction to like figure out hey is this valid or not we've enhanced like what we consider valid or not like if it triggers this it runs through this pipeline of detection tooling kind recognition AI models if it triggers like hey this flag saying hey this is actually malicious this is an attack instead of just including into the block now we just like discard it we jail we essentially jail it right or

quarantine that transaction fail does transaction just like fails then or like doesn't go through yeah yeah it basically gets isolated it fails and that's where like that second periphery system comes in to say hey like if the user wants to challenge that like quarantining of their transaction they can and that's where we built these tool we are building those tools out to let them challenge that and get remediation of like hey I was censored if you're censored then great the sequence are screwed up it gets slashed and like you get paid for that essentially okay okay so so the safe sequence Ser is as you were saying connected to the client's node um and then effectively the transaction goes through if it's valid it gets pushed through if any your models um kind of sent something's wrong

that gets basically sent down to the firewall kind of consensus layer and of which then there are proof of stake slashing enabled uh via the firewall chain which is a firewall evm chain specifically meant to uh in interact with various sequencers of various rollups that are inter that have the firewall code embedded and then basically the firewall chain is settling all of these different disputes or arbitrations or uh failed or malicious transactions or or just normal transactions onto the own consensus of the firewall evm like chain welcome to the ignition Mantle's newest campaign to

liberate the growth power of Bitcoin fbtc ignition has officially begun with a total token supply of over 2,000 fbtc with a tvl of over 150 million right now you can go on the fbtc website to see all of the featur campaigns saking opportunities and liquidity pools where you can utilize fbtc on chain today for some of the best yields available on l2s the ignition Sparkle campaign can be found at fbtc docomo dampa yeah sort of so like that's definitely a start like I'd say like um our explo consensus mechanism has like a few different uh values to it basically

one of them is definitely to describe right there which which is yeah to remediate uh the the the trade-off of like having you know better protection or detection essentially and blocking on your execution environment which is yeah some things are going to be censored uh by mistake right users will get blocked like we have optimized our models of course to be like extremely low false positive I'm there's going to be like some false positive still right um and so we we want to build that solution to remediate that but the bigger picture actually when it comes to our exploit consent ensus uh and this is not something that we're launching with our go to market with a say sequencer but it is on a road map and we have it in development today um it kind of comes from this idea you know blockchain

consensus mechanisms aren't or sorry they're they're meant to reflect social consensus we believe it's already social consensus that exploit should not be allowed for example nobody refers to the kyber transactions that netted $50 million in late 2023 right they say the kyber attack or the kyber exploit these words are synonymous with theft and theft is illegal so with the exploit consensus of firewall we're essentially updating blockchain consensus mechanisms to correctly reflect that fact um and so when you think about how do we do it you know today like fraud proof mechanisms are used to Define invalid State transitions in a blockchain which aren't allowed because they don't follow the rules for example you know 1 plus one does not equal three if such a

transaction or stay as proposed it's invalid it's very simple right it's all objective it's all math we're essentially expanding that definition of inv State transitions to include smart contract exploits so we essentially are using that mixture of mathematical and social adastation to classify transactions as exploits within the firewall evm with in the safe sequencer and we wrap that up into like this exploit proof which is where that ZK like provable sort of concept comes from and like we already we already see this like we see this as like the Natural Evolution of consensus and smart contract networks and I was going to say that um you already see this today this consensus right right uh basically after every exploit on crypto Twitter there's a clear consensus on like when and what

an exploit is it's all there we're essentially building the architecture and coordination layer to harness that consensus into a delivery mechanism so that blockchains you know smart contract applications and most importantly users can actually benefit from that recognition of theft um so that's like really a good description of where firewall's exploit consensus comes from and like what we're building there it's all birth from like the core pieces we're building today and we're launching with which is the safe sequencer and like that proval computation essentially yeah so you guys are kind of anticipating social consensus and saying like would the the Society of this app or this chain consider this a malicious transaction and so I mean a like

thinking back to some of the hacks that I've been just like seen and been around for like the Dow hack back in you know ethereum days was like ethereum social consensus said yes like Fork the chain and create ethereum classic and and you know put essentially like don't get rid of those coins on the chain and then we've seen that on a much more isolated Case by case basis with each of these apps even think about like blast and L lanaa like they literally did this on the rollup the rollup like level and yeah as you mentioning applications are doing this course all the time um so like the thing is like the issue we saw like Lan blast is like RS are forced to like they already feel responsible for like assets on their change right

especially when they are centralized so the issue today a lot of them face that the The Dilemma is like hey do I want to like optimize for credible neutrality do I not touch anything here or do I want to optimize for user safety like it's a really [ __ ] hard like I was gonna say like respectfully Devon like who are you to decide what is you know right and wrong on the on chain Forest right yeah for sure I think it's extremely justifiable if people are losing money the the industry is getting cracked down on by Regulators there are court cases going on people are upset like this is like this is clear a clear roadblock in the broader adoption of the space yeah 100% And like for us the whole point is like we solve that dilemma like we're

like this third party who's building an incredibly neutral Network that just plugs into you your rollup so that you don't have to worry about that you can get user safety and trouble neutrality and of course it's going to take the thing about us like the our like trade-offs from like day one is like we've decided to trade off like some decentralization component because we can add that in later and we will be adding that in later and it's part of our road map from day one of course but it's not going to be part of the product from day from day one um and we're optimizing our solution to First be like really good at detection and like stopping exploits because that's what users care about at the end of the day that's what liquidity cares about yeah um and uh yeah to your to your point uh

Andy like a big point point of explo consensus is like we want to involve the community to like literally plug directly into that we want to tap into the social layer right we want to automate social consensus to say yeah the these parties who are all like able to they they're all testing themselves say that's an exploit it's not just firewalls algorithms that are saying it it's like literally these different parties that are calized saying it too and that's going to be like part of our long longer term road map Force um but yeah to to start from day one it's like yeah our algorithms are saying this is malicious you as an a rollup has already accepted that our AI model is generally right which is why our approvable computation our models having reputation attached to

all that is actually really important from the product side but like today like no detection team does that because um well they don't building the same product as us but yeah I'm I'm thinking about like I I can't come up with an example off the top my head but the edge cases here would make me go silly like to to try to again like like I I think you know decipher you know who's to say like what's right and wrong in terms of or not is such a tough question to answer like that that someone's you know just playing the game better or or not because you do see a lot of those Twitter per personalities after the fact saying it's a theft or it's not but that's very reactive and happens after

the fact and so is there is there almost like a dispute delay period where maybe something gets flagged it gets sent to like kind of this contestable nature like uh debate type of Realm and then people can vote on it with some form of social consensus and then maybe there's a reorg right that allows that transaction to go through or for for some how how do you plan on handling some of these edge [Music] cases Cary is your home for app specific rollups at the Forefront of developing new tools and platforms that enhance the capabilities of blockchain Technology ctei enables applications that are

flexible and built to scale build on a full Linux environment and use traditional software libraries but in blockchain development go to cy. that's cy. to learn more about building your own app chain today yeah definitely um I think like honestly for maybe 80 to 90% of exploits you can actually probably pretty objectively classify them as exploits you don't even need to tap into social consensus those are the ones that are like it's just like math right it's like RAR Seas overflows under flows clear bugs like there there's a lot of things you can just objectively say that's an exploit um I'll also add and I haven't mention this yet but we don't we don't

just like block everything right we're focused on specific uh exploits and that's just like if the simple rule you can kind of use is if the smart contract is losing funds we will block it we'll protect you basically but we don't we don't cover like me we don't cover front running we don't cover sandwiching don't cover [ __ ] like that because it's like totally subjective right is that like an exploit is that part of the game theory like you know like we don't we don't touch that [ __ ] like okay so so it's not so much that like one user is able to take money from another user it's like look at the collective of the smart contract that has some leaky bucket hole that's when you guys are able to step in and say wait no that's actually like some trap door that wasn't supposed to

be there exactly and it's all based on like how traditional fraud you're talking about like my AI background like my basically like my background between like understanding blockchain system and infrastructure kind like the AI side is like really what's helped us not just me of course generally the team has helped us like understand like how what the future of like block space should be in this industry uh when it comes to speed time security and um yeah your point Robbie like um yeah that's how we've kind of uh been able to create an actual solvable problem right um like we don't just stop everything we we stop uh exploits because we've been able to model them off of like how trafi solve fraud uh which is anomaly detection right um and you can detect what anomaly

looks like if you're smart enough with your your algorithms uh for a lot of these exploits and we think about this this longer term solution like I keep saying zero exploits zero exploits of course our first product is not going to have zero exploits but we expect we're kind of internally at least saying and even to our customers hey we're g to adopt the same sort of mindset like open AI or any of these like AI teams are right like every version We ship and we're just continu to get better right it's going get better and better and better um and when it comes comes to zero exploits it's because like we need to create this like in credibly neutral social layer to like hand off those like contestable ones too the ones that aren't like objectively like mathematically like an exploit uh we need to be able to hand

them to like this incredbly neutral social ER to say hey that is an exploit or that isn't um and that is something that's honestly very a very long rabbit hole that I probably won't dive into here but it's off this concept that we've termed as programmable finality it's the idea that you can decouple execution of a transaction from finalization of a transaction this already exists today on rollups right obviously between like the time that the the data is post to the day layer and stuff and like there's already that time there and like we've created models which essentially detect like and you can even again plug in like social aspects to this uh that there's already like a natural settlement delay essentially between uh when a transaction is executed and a transaction is finalized um and so that is how you would essentially a little

Alpha League is how you would essentially create like zero exploit uh environments yeah and it it makes sense that you guys are being extra safe optimizing for for that to to start with and then working backwards to minimize the rate of false positives like you know that way you're able to again mitigate those risks maybe some people get you know transactions delayed but just out of out of the for the sake of safety and then you can kind of work backwards from that I I have an AI question for you but but first um I want like how do you plan on integrating this into the existing infrastructure like is this something that arbitrum can plug into its sequencer optimism can plug

into its sequencer you know and so on or or is this something that like new rollups can use the firewall stack and deploy their own rollup with the safe sequencer yeah it's um it's more the lad to the big the big I see the big trade-off so there's really two big tradeoffs one is false positives and what you just mentioned actually when I was talking about like us getting better and better it's actually we want to get better and better on Expo detection but we're optimizing for Speed and no false positive to start because you think about the solution itself from like a custom point of view this what we've heard from customers this is why we've done this but essentially our solution the safe sequencer is purely additive if you're just detecting some exploits you have no false positives and you're as

fast or faster so we've actually optimized our the safe sequencer in our V1 to have no false positives it's going to detect less exploits it still you know does a decent amount but it's not going to be 100% um because that's just pure additive versus if you added a bunch of false positives into the solution now you've actually introduced a new negative into the rollup right because you don't want again you don't want to block their users um and so that's like one thing we we made there but in terms of like how we yes that's like one trade-off right um the second that's not a tradeoff sorry but that was like just one point and that's how we address the false positives issue which is one of the major trade-offs you'd make with this sort of like architecture shift the second is kind of what you

just alluded to there it's a related to that which is you know you need to um essentially block off uh the ability for someone to be able to just submit a transaction into the chain forcibly right um and so today like on optimism or whatever I think on AR they still like a 72 hour delay or something if you try to like do an L1 transaction into the rollup but uh that's like we basically gate we gate that off so like you can't just like submit a transaction from like L1 into the rollup you have to go through the sequencer because if you don't gate gate it then like an exploor would just simply submit to L like through L1 and like still get included

right and still steal the money um and so how we again like that sounds bad from like you know just saying that but how we've addressed that is again relates to approval computation like if you can prove that you know your transactions on exploit like our nature of proofs then like you can then you can submit to L1 directly right so we're not censoring we're not censoring users right we're censoring attackers that's like the big kind of differentiator there um and so when it comes to like yeah can optimism plug this in like depends like depends how they want to you know adapt like their super chain sort of like uh laws and like the rule of law there or whatever um I'd say like they probably today like yeah they can't

because they probably don't have that Nuance in like their acceptance of like forced like transactions through L1 uh of course like our big our big thing that we're doing is like we were trying to like pioneer that so that actually becomes the adopted standard but it is more so for like new rups today like they just launched with like the op stack that's firewall essentially um and yeah it's just like Plug and Play like again all they need to do is like run the fireal sequencer um that's the only that's really different about the client and like the whole role base um we have like maybe I don't know we have like a very small like footprint essentially on the the technology stack itself it's still mostly like the ARB stack or mostly the op stack right um I have some Alpha question here when will we see a a

roll up with a safe sequencer live yeah yes this comes kind of like to our big milestones for this year um so we our our big milestone essentially is we're launching our public test net on Q4 this year um that's G to be a big push for us not just from like of course the tentacle side but in the side but also from the community side like we're starting to really um ramp up like our ecosystem development with the launch of the the safe sequencer test net uh and I'd imagine yeah q1 is when you see the first uh live roll up with a safe sequencer and um our big sort of thing there is again like we've talked a lot of rollups um and we've also realized that they all just want liquidity right

and that makes sense so you just want liquidity on your chain that's how you make your your apps like attractive to build on you it's like a flywheel right again and so when we think about how do we incentivize someone to run the safe sequencer it becomes simply will bring liquidity to your chain um and so to to that extent like we kind of see this way of like playing out uh we talked to like tra five funds and like just crypto funds too of course um this idea that it'll become like a fiduciary breach essentially to deploy onchain Capital to block SP that's not firewall like why would you do that you should be going to the safest area and of course it's going to take time to develop but it it makes sense because it's like true um and like when you look at like any

Capital Market in the world of course like the most liquid ones the most the most the one ones with the most assets the ones that are the safest like risk reward so we just simply reduce the risk and like the reward just happens to go up you know automatically by that balance the scale love it I got one last one last question I think this will take us home like we were we were chatting with Ilia from near the other day about uh kind of like his vision for AI and he he mentioned something that kind of gave me a light bul moment which is this idea of you know once we have ai once we have ai agents we can program smart contract logic to basically like deploy these AI agents and so this is kind of an ironic question but basically like if you have

an AI model that's detecting like this malicious behavior what's to stop that AI model from deploying a smart contract to deploy an agent to like have basically backo the AI model itself because it deployed the agent itself and take all the money from the smart contract sorry yeah so you're saying um an AI basically would be identifying like a hole in like our detection essentially and it yes yeah basically like the the AI model like deploys the AI agent because it knows what the model it can detect or not detect um that's interesting yeah like the AI finds the

loophole within the uh within the the model itself yeah yeah that um I I'd say like that's an interesting future uh I it's the future sir it's the future plenty of time figure it out like it's not happening anytime soon so we got our big thing is like um we don't actually expect to like forever compete on the AI side right especially in like the new meta that exists with the AI industry of course like there's trillion dollar companies that are open sourcing like Cutting Edge infrastructure and architecture for AI that could one day just completely wipe out like AR

detection abilities right um so like we want to basically create the architecture and the infrastructure to like plug in any AI right um uh and in that future OB see that exists like that I mean I'd imagine that there's also some like better AI that we could also use to defend that I think that like security generally in the future is just going to be like an arms race of like yeah you need to keep having the best AI model right of course and like that's what we invest in AI s um and yeah no real good good answer to the solution there but I think that we'll be able to figure it out I'm optimistic so nice cool man yeah I realized that was a tough one but I think that was a pretty good answer so dude appreciate you coming on um yeah I think it makes a ton

of sense uh to make these sequencers safer make these chains safer um and ultimately like just make it ready make the onchain world ready for the next wave of users so yeah keep up the good work yeah thanks guys thanks for having me on uh super super awesome and yeah as you said happy to let you guys take my podcast virginity here so hell yeah dude yeah yeah you know firewall we're excited we're a innovating in a new paradigm um it's going to take a lot of lot of time as well as you know uh ecosystem support to to get this sort of idea of like hey we should build in security into the the chain itself adopted I think but I'm very bullish on

the modular thesis and the ecosystem I our whole like road map is based around it um and so yeah excited to to see uh the next few months here and excited to launch our public test net later this year and get things rolling hell yeah brother have a great weekend thank you for coming on Cheers awesome thanks