DeFi By Design EP#109: One Click Rollup Deployment with Caldera

Transcript

welcome back episode 109 Rob D5 B is L what's going on guys got Matt here from Cala Rob GM GM yeah this is uh this is another podcast uh today episode 109 uh pretty wild we're on episode 109 because if you would have asked me two years ago three years ago if uh defi was still gonna be alive in two three years I don't know probably depend on the market conditions at the time uh I just said yet the I'm kidding we'll just edit that part out Matt Matt good to have you on yeah super great to be here

awesome well Matt we can we can uh get into it uh we like to have fun keep it conversational and Casual on the podcast here and then also talk about some high level Concepts that can provide our audience some Alpha so um let's let's talk about yourself first maybe that's a great place to start uh so before we dive into Caldera uh maybe you could tell us a bit about your background and how you got you got started in the space totally yeah so yeah first of all great to be here um yeah my my journey into the space uh it actually happened quite a while ago I'm I'm like I was like weirdly a crypto OG though I don't have like the net worth show for it so like when I was in Middle

School I was like really active on like online hacking forums and like script Kitty you know type of stuff a little bit of like you know just like General software development and like from somewhere you know on those forums like discovered Bitcoin in around like 2011 2012 I was I was you know like 12 at the time um it was like you know it was so different from now like I would I would be like begging my mom to like send money orders to like Russia to let me buy Bitcoin she obviously would not do that um but I eventually was able to like earn some Bitcoin through like online development work and that was like my first Fay into the space um ended up uh uh buying some Bitcoin when like the First ATM opened up in my hometown uh and then lost all that

Bitcoin in Mount gax that like my first full cycle crypto experience um post that uh you know kind of give up on crypto for little while stuck with the software development just doing a lot of like miscellaneous work throughout High School uh ended up at Stanford where I studied computer science uh with a focus on like distributed systems operating systems and a little bit of cryptography and of course when you're studying those three things you kind of end up back in the crypto space and so that's how uh you know that's what eventually led to me kind of like rejoining the space uh meeting up with some cool cool people here realizing a lot of the really cool things happening at the application layer realizing that scaling is going to be needed to allow those application to thrive and then that's what L to us

finding a scaling company in Cera nice yeah Rob did his degrees in math and philosophy and earlier today we had a a a presentation on batch proofs as far as ZK Prov go and um Rob was happy guy he was a happy man it felt good to be back in math class for a little bit I was like I was like picking up on Concepts that were um like probably beyond my understanding now and maybe I could understand when I was taking those classes so uh it's cool to see like the concepts you learn about at like such a like an analytical not just an analytical level but like you

you've taken plenty of high level math classes right so like you understand when someone asks like when are we gon to use this in real life that never really goes away throughout like your entire like math curriculum whether it's like eighth grade algebra or it's like like University combinatorics you're always kind of asking like when are we going to use this in real life so to see that that math in ZK proofs actually written out by someone doing it by hand was really cool like to design a Merkel tree and like set up some of these proofs was was really cool to see in action so you know we we kind of we take a few different perspectives when it

comes to looking at the crypto space like what is something that you're most interested in whether it's you know the problem that you're trying to solve the scalability problem with Caldera or generally like where where have you seen the most application of your your education in cryptography and distributed systems where have you seen that most directly applied yeah I think in terms of my educational background um definitely kind of what we're working on right now the scaling space it's it's kind of funny because I would say like rollups are in some sense like the web two ification of web 3 or like harmonizing the lessons we've learned in web 2 and applying them to web 3 I mean the whole point of a rollup is like you're taking

all this computation and storage and you're doing it you're moving that all offchain which means you're doing it in a web two way for efficiency purposes and then you're figuring out how under that Paradigm to like get the security benefits of web 3 and still transfer them on onto those new processes so um definitely like what we're doing right now feels like a good combination of like the stuff I worked on at my previous rules which were all like Web Two or web 2.5 uh and like what I studied in in college and and you know web 3 things in terms of like applications that I'm really excited about um I think we're finally seeing that scaling is no longer going to be the bottleneck for a lot of really great applications um thanks to the work that like we're doing and other folks in the

modular blockchain space are doing um so like earlier today I was actually just uh on a Twitter spaces with some folks building in like the onchain gaming sector that's that's a sector I'm super bullish on basically how do we put like entire video games all in smart contracts all on chain so that like they actually work in a completely decentralized manner where people are going to be able to like develop new clients and develop smart contracts on top of those games and we're going to see games become a lot more like uh like protocols versus like you know monolithic pieces of software um and so enabling those types of use cases is also something I'm super excited about yeah our buddy from uh ZK sync Alex our buddy the the co-founder of matter Labs he uh he was saying that basically he thinks ZK and just general

rollups he's a little you know very very ZK focused but I think optimistic as well they just enable new innovation and application and this I mean we've just been dreaming about this in 2020 um even in 2017 like just when there was not much onchain during those two kind of adoption Cycles or however you want to call them so yeah I I we align with that at the rollup and frankly we are super interested in like educating our community on um these terms like I have DEC centralized sequencer Maxi in my bio like in what world is that English Matt so let's talk about um the decentralization and the centralization of sequencers and kind of where we're at

so we had a podcast earlier with Matt Cutler from block native he told us a lot about like kind of you know different pros and cons to different designs in in ethereum I'll just leave it at that because that one's great you guys should all go watch that but we got the sequencers and basically we talked about the pros and cons of of um you know having a a centralized sequencer and a decentralized one and just what that means so um I know that was a lot but what kind of is the is the uh tradeoffs and the state of a rollup sequencer as we know it um and kind of where do you guys fit in there yeah so this is of course Very topical because

we just last week we announced the first optimistic rollup that is working with the decentralized sequencer network uh we we built that with espresso who's working on their own decentralized sequencer Network I think they're the best team in that space But there they're of course um a bunch of others um so on the topic of like rollups today every rollup today is centrally sequenced uh which means that when you as a user send a transaction uh to that rollup you send a transaction to like their server their server will order um all the transactions and that server will produce blocks and eventually submit those blocks to ethereum mainnet or wherever the rollup is to um and so this is like this generally Works um and it's pretty secure uh so like these

rollups they still if as long as they have like working proof systems whether that's an optimistic proof scheme or a ZK proof scheme like they can't steal funds they can't you know arbitrarily change the the state of the rollup they can't really act maliciously in most ways um and so that's why like we've been able to with all the centrally sequence rollups that we have right now we've been able to still secure billions of dollars of value that being said when you are centrally sequenced there are a few ways that that sequencer can act maliciously the first way is uh through censorship so the rollup gets to decide then what transactions are included in which blocks and The Ordering of those transactions that means that that sequencer can then uh theoretically prevent a certain user from submitting

transactions to the chain oftentimes a lot of the good rollup uh designs will have some way to bypass this but that will often take you know on the order of like a day or or so 24 hours 12 hours in order to force trans uh the inclusion of a transaction in the rollup so like arbitrum will do this I think optimism will do this so there's a way to get around complete censorship but the centrally sequenced rollups could theoretically delay your transactions and ruin your user experience if they wanted to censor you the second thing uh with centralized sequencers is that you as the user don't have any visibility into the rups me poool and how they're ordering transactions and in fact you can't there's not really a way for them to prove that they're ordering

transactions honestly because you don't know like when those transactions are come kind of actually coming into the sequencer via the network right might have different latencies Etc um so the sequencer you know the extent to which those teams can promise things are like you know marketing and saying you know oh we promis to order these transactions fairly but they themselves even if those teams want to and I think they do want to can't prove that they're ordering uh transactions fairly and so that means that there's a lot of me that those uh that those sequence could extract um if they wanted to and so like some teams like optimism actually had it in their original design they would extract those that me and return it to users and use it to fund public goods I think they've since uh deviated from that position um

but it is like a there's actually like a ton of me that those rollups can extract um and so that motivates the moving to a decentralized sequencer where instead of this mempool and all this ordering happening on a server in a centralized way people folks just like how they would in like ethereum or polygon proof of stake other you know Bitcoin other top networks like they would just submit their transactions to these decentralized sequencer networks and and then they're able to have more faith that that network is ordering that those transactions in a decentralized way no single entity has control over the ordering of all the transactions on the network um and then the the rollup node will then pull all that that ordering from the decentralized sequencer Network

rather than uh being relied upon to order those transactions themselves so you remove the you know uh the you the fear of like censorship and you also decrease the amount of M that the sequencer can can extract from the chain yeah that's what um that's what that's what Matt was saying rob it was all about the me because when you know it comes down to the economic uh viability of this you know situation in terms of security and profitability and kind of me is is very uh a dominant Factor yeah yeah totally because we we saw that play out like we we saw the me ecosystem flourish and

then all of a sudden people people raced in to be part of that competition um and share sharing that economic value and then we saw Solutions come up uh we saw me boost right where that was an attempt to decentralize me a bit with auctions I believe and now we're increasing ly seeing like more decentralization in those economic systems and this this brings me to The Narrative the other narrative that that we're seeing in in lots of Twitter bios which is modular systems because as we kind of like start to unpack the stacks of these rollups like ethereum made a transition from from proof of work to proof of

stake and part of that came the separation of proposers and Builders and then with with layer twos we're starting to see almost like th those come back together with the sequencers and the bundlers and now we're starting to unpack those again by decentralizing sequencers from from a centralized sequencer so so my question is why why is what why are why is decentralization so important in this world of modul modular rollups and modular chains why is decentralized sequencers how does that help to kind of like set up this decentralized network of chains and rollups so that we're not we're not

siphoning off economic value from one another but we're more so taking part in this Collective economic machine totally yeah that's a really great question um and I think it's really insightful to talk about like bundling and unbundling people talk about that in like the web to Startup space as well right like you either make money by bundling Services together or by unbundling services that were previously bundled uh and I think in in web 3 um we we do see like similar Trends um so like as you mentioned like we we did have a paradigm where a lot of where where like sequencing was decentralized we we still do right ethereum sequencing is decentralized then in the interest of efficiency and scaling we moved uh to rollups right now which are all centrally sequenced and

then now of course we're trying to go back the other way um and I think it's it's basically like trying to you know strike the balance where move in the direction of performance first and then you move in the performance decentralization second and then you end up at a you know a later equilibrium uh you know then that that's better on both fronts than than what you had before um specifically with regards to like the modular uh like Paradigm I think decentralization decentralized sequencers are really important because this allows you to now plug in decentralization where you would have needed to previously build an entirely new network um so like like you know you could build a decentralized network you something like the cosmos SDK now you

needed to but you know you need to recruit a bunch of validators to your chain you need to incentivize them somehow um you need to make sure you have a token to incentivize those validators which leads to inflation and maybe it limits your ability to design the protocol in the way that you want limits your like business uh you know timelines like you'll need to launch a token at the same time you launch the network there are all these like undesirable properties of doing something in like such a tightly coupled way now you can just into an existing uh decentralized sequencer Network like espresso and you know maintain all those benefits I think there's also a bit of a second uh secondary effect here that encourages decentralization not at the network level but at like the sort of like social or like Society level where

previously the only teams that could have the most decentralized systems were like ethereum right or like the top the top chains those are the most decentralized if you wanted to launch uh a a chain you know on Cosmos or something that Cosmos chain will never be as decentralized um or as legitimate as ethereum now with like the modular stack you're able to plug into that decentralized aspect of ethereum or you know in future something like espresso or or both um with our architecture and that means that like if you're this new chain uh it's a lot easier for you to convince other users and convince the community that you're aligned with those ecosystems and that you're sufficiently decentralized well previously that was really hard to do if you're trying to

bootstrap your own networ and prove to people that that you're decentralized and so we think for that reason it's going to become a lot easier to start successful new chains and so the the total space you can call it decentralization you can call it maybe fragmentation but we're going to end up in like a much more kind of multi-polar decentralized world than we were before just because we no longer have these like centralizing forces of trust that play yeah interesting yeah kind of it it exponentially increases uh each choice that you make in favor of of to just say like the good guys which is just like choosing permissionless and uh trustless every day and like having that ethos in your in your blood you know and just

kind of trying to build that even if it does even if it means less profitability unfortunately the it's definitely not a bad thing it's a great thing that bases make I think they made over10 million in Revenue last month from from friendch like like what their their second month live they made 10 mil what so like the possibilities of on the econ on the economic side are wild um for for for the sequencers but again it's like that choice of um of kind of the ethos but I don't want to get too deep into into decentralized sequencing uh without talking about shared sequencing as well which is kind of like the big key thing with espresso Rob's

got something though so hold on one sec I want to I want to pick you back on the on the base conversation because if we're if if we if we do like nudge systems in this in the direction of C of decentralization and base is currently bringing in $10 million a month of in Revenue per month with a centralized sequencer what what is the the economic decision that that they make to choose to keep the centralized or or go decentralized is it is it just a matter of ethos versus like Revenue that they're bringing in or is there a a scenario where they can continue to

bring in close to $10 million in Revenue each month but open up to a to a decentralized sequencer what what is the the trade-off of Bas specifically making that decision yeah so on on the technical level uh there's the these these decisions are actually very very separate like the fees that base chooses to charge to users um are completely separate from the fees that base needs to pay to ethereum or the you know the integration of some shared sequencer Network or distributed sequencer Network um that ultimately is kind of set at like the protocol level uh and base will still have control uh over that that protocol as the developers of it um and so base could decide you know tomorrow

to like increase uh the fees they could decide to decrease the fees um that on some level is like an arbitrary decision um so the only thing that like a decentralized sequencer Network like integrating with espresso or similar will will do is like it will force them to incur um increased costs because now they need to pay to have these transactions sequenced by that separate Network and typically you know submitting transactions to network has some cost associated with it in crypto although hopefully like the scalability properties of all these networks will keep those costs low um but it doesn't necessarily mean that that base needs to change uh how they charge users um uh which is you know another really

great thing about like the modular stack and this architecture that a lot of these decisions are decoupled um I will say like base the way that a lot of these like op stack chains um make money is that they don't deviate too far away from the standard set by optimism and and to some extent arbitrum which is to charge a little bit of like extra buffer uh you know to account for uh fee volatility um on eel one and because they add that buffer in expectation they to make money and so that's why you'll see those graphs on like Dune analytics of like all the the rollup profits that all these sequencers make um and so I think the thing that will really motivate the decisions of these folks who are running large rollups is the extent to which the community cares

about decentralization if it turns out that the community does care a lot about decentralization I think we're going to see new decentralized uh rollups uh really eat into the market share of these existing players and kind of force their hand um if that turns out not to be the case uh then I think we'll we'll see a little bit of inertia um from them moving um you know to a decentralized sequencer network but I think you know as things have typically played out in the web 3 space like typically there is some pressure to decentralize over time and and I think the same pattern will play out with rollups yeah to bring it to an analogy like you had like the eth killers versus eth obviously we hold eth like a lot of people do so perhaps bias but like these

ones that were more centralized in my opinion have ultimately a lot of them 90% plus of them have lost to ethereum who has kept its mind on like it's as if there's a this this organization of of ethereum has just kept its mind on that train of decentralization more so than I don't even want to name I don't even need to name one there's tons right and so to that same regard that decentralization provides pressure in the same sense that like when arbitrum came out I was more keen to to deposit significant sums of capital into arbitrum than I was to hold it on these other various l1s that were evm compatible but they had side chain um that side chain architecture low validator counts different kind of

various pros and cons you're just like dude at this point like obviously this arbitrum chain is centralized but it's better than everything else and to to that extent the same things is kind of going to happen here with the with the rollos because from a user perspective with good enough education makes more security sense to be in a place with there's where there's more decentralization um there's usually more cost associated with that or there sometimes has to be or there's innovative ways to get around that but that is usually true so I agree with that point and yeah I still want to tie in So to that to answer further on the base question the changing to decentralized sequencer what is the current shared sequencer landscape look like from your lens um and are teams

implementing them our team implementing you guys frankly what do you guys even do from a high level for everybody who we haven't got there too much yet but like can you tell everybody what you guys do as well but I know that shared sequencing is something that that you're GNA want to touch on yeah so Caldera we're the platform uh to launch modular blockchains and all the infrastructure those blockchains need so we're kind of like a One-Stop shop for folks who are thinking about deploying their own rollups whether those are like application specific rollups or ecosystem uh specific rollups they'll come to us they'll log onto our platform or they'll hop on a call with us they'll tell us what they need uh on their rollup and we'll help them launch a chain and so in

the web 3 space uh a lot of uh there are a lot of really really talented teams that don't have protocol developers or like distributed systems Engineers or developer operations folks and so it means like for like a lot of these teams that are looking to to launch their own dedicated chain they don't actually have a lot of the expertise in house uh to to do that they're really focused on like smart contract development and developing a really great app and you know front end user interface and so we helped abstract basically all the concerns that aren't those two things away from them so they can focus on uh building building their app uh without meing to get into the nitty-gritty of like protocol engineering um uh you know in deploying their own rollup um yeah that's sick that's such a sorry to

interrupt real quick it's just a great business idea we pivoted to the rollup like last not last minute but you know like six eight months ago similarly just kind of like understanding that the trend and then having like like the devops yeah it's sweet and you you guys then have a host of shared sequencers and sequencers that you allow these chains to come and or or these teams to come and build upon and then use yeah so we aim to be like the hub for all of this modular infrastructure as I'm sure you guys know you know as users of chains like when you're when you uh want to like onboard onto a chain just having like a chain existing there is not like the only thing you need

right you need developer tooling you need bridging you need common infrastructure right like the graph data providers nft marketplaces right dexes Etc and so we aggregate over basically all of the the top teams uh in the space we we help those uh the new chains that are deployment those integrate with those teams and so that you know all the stuff I mentioned kind of exists above the stack uh uh that that we operate on right graph index are Marketplace Etc but we also integrate with a lot of Partners kind of below the stack um or or like at the the level of the rollup stack and like those two things are mostly da and sequencing um so on the da front um you know right now most rollups that are built on ethereum they post

data directly to ethereum as with anything on ethereum that is super super expensive um definitely uh good for certain use cases and really doesn't work for a lot of other use cases especially High throughput ones like gaming social and so we allow the rups uh that are built with us to to launch with uh Celestia da for example eventually will integrate the other top providers like igen maybe Avil Etc um to like radically reduce costs on their train uh and then of course we will also offer a Litany of like shared sequencer options so right now we we have two options you can use a centralized sequencer model like the the same model that arbitrum optimism ZK saying everybody else is using or you can

integrate with the espresso sequencer um but eventually we want to offer Integrations with every credible shared sequencer Network and we really see ourselves as like the top of funnel for a lot of these folks who are building uh lower in the stack than we are to like gain access to like all the developers who are actually looking to deploy their own rols and you you hit on the the centralized and the decentralized uh sequencer models um so could you touch a bit more on the on the shared the shared sequencer mod model and then and then also like what what are the ethical tradeoffs and then the economic tradeoffs for choosing a shared sequencer yeah so the way I think about this is like there are two axes um on

which to like talk about sequencers there's centralized to decentralized and then like shared I guess to like not shared and so typically there's like a bit of a conflation between these two terms um but but they they do technically mean different things so a shared sequencer is uh it means you know at the base level that like the sequ you're using is also being used by other rollups and other chains and so something like decentral uh like like espresso is both decentralized and shared but you could uh have and some people are definitely working on this like centralized but shared sequencers where you're still Central you're still sequencing the chains in a centralized way but you're sequencing multiple different uh chains and that's can I can I interrupt with kind of like a comment but also a question like that that

Dynamic that architecture sounds the most economically sound the centralized and shared model sounds like I can centralize my my operations keep costs low and then I can also provide this for multiple different networks at once because it's shared and and again split costs even further is that is that kind of like a uh accurate uh mental Surplus too Rob these Shar cash a centralized no no that's okay soal no I mean yeah Matt yeah yeah um yeah I I think you're

you're correct um so yeah you you do get to split split the costs of like multiple chains all into one sequencer um another benefit you get is is interoperability where like now that when all when all these chains are are sequenced by the same sequencer you do get some interoperability benefits um it makes it easier to do atomic swaps across chains Etc um because you know one way to kind of model it out is like once you have a completely shared sequencer between multiple chains there's actually very little difference between that on a technical level and treating all of those distinct chains as kind of like one chain and so you you get like big benefits of like atomicity um between those chains uh as as well so yeah a lot of really nice properties and I think I think you're you're correct

that like if you're going at this from a pure economic standpoint like you get all the inout benefits um uh by doing it that way and you don't need to deal with the costs of running a decentralized network or plugging into a decentralized network as you would with decentralized sequencing and what do you do with the Surplus what are all these rollups doing with the surplus of like what is bass doing with this who knows what like what's is is that to be distributed to token holders to team to Dow like like and with Caldera I guess what's the what's the economic model for the Surplus left over after and for those who are a little bit uh new to like the the The Da Side of Things da is like some some of the most expensive um

operation as you're saying Matt so you you've got the user fees minus basically da costs and then you get money left over yeah I think the great thing about rollups is that it's up to each individual chain to decide what they do with that Surplus so there will be chains that just treat as profit there will be chains that use it to fund some grants program there will be chains that try to minimize it right now like very few rups are trying to minimize it but they basically can decide to right set a target of like zero dollars profit and then there there are going to be chains that try to uh find some novel mechanism to redistribute this funds uh These funds right or maybe they'll like stake it and use it as like an endowment for the chain to you know ensure that that

the chain can be continually operated and provide stronger guarantees for their users over long periods of time I think right now um there's very little Focus uh on this I think people have just started really talking about the Surplus within the last three to six months as you mentioned part of it I think motivated by friend Tech's major success on base um but because there's now a ton of uh interest in just like what gets done with these like massive piles of money I think we're we're going to see a lot of innovation more on like the business side and the economic side of just like how those funds end up being used and it would be just like one of you know many other ways that rollups differentiate from each other public goods man I'm sure a lot of them are going to use public goods totally

use it it's like it's like it's like a business it makes me smile because it's still a self-fulfilling prophecy of a business if you think about it in a sense however uh with Innovation more real real use cases more real users quote unquote and bigger communities I mean rollups are a business model in a sense and so like that that's a big step for crypto um and for like this entire industry to to think about things like oh we have a surplus and I think yeah it's going to be a cool differentiating factor to see public goods or who's more capitalistic and profit taking just kind of gets rich off creating a rollup or kind of who does like and then there's gonna I'm sure there's going to be interesting rollup stories of people doing bad things or good guess just

going to be interesting you can see it developing right now there's couple dozen rollups right now maybe maybe I think probably maybe less but probably I'd say a couple dozen dude in six to 12 months there's going to be like 1500 or a thousand they're going to be like what just happened so yeah this is going to be very interesting to see what what happens yeah that's all I got yeah I I totally agree um and uh I I think you had a really good point there as well uh with just like you know this being kind of like a new form of business model it's actually one of the reasons one of the biggest reasons we're seeing projects come to us and want to De deploy a rollup is like it's really rare in crypto to find like defensible

sustainable business models almost everyone relies on Pony nomics like inflation just like any something that just like can't be sustained over time the rollup business model is actually really really clean you it's effectively just like a usage based pricing model right you're taking a p you know a percentage of of transaction fees you're taking some like it's like a Cost Plus model right for the application um and so we're seeing you know I think a lot of applications take frch for example right like like they will probably in the future maybe move to their own chain because they'll be able to internalize a lot of that revenue and their users are going to want them to internalize that revenue and you know allow that to fund development um and so it's another you know it's another reason why I think we're going to see as you said like,

1500 tens of thousands millions of rollups in in the near future yeah I think that unlocks this this this new narrative we're seeing play out as well which is app chains um right which is like I I think you make a great point about like the consideration that frch is probably is probably thinking about where what if we do launch our own app chain why why should we or why should we not and kind of like it it would be fun to play out kind of that hypothetical situation of like you know we're in this boardroom Round Table conversation where the three SE Suite Executives of friend Tech and we're choosing whether or not to launch an app chain like do we do we launch it you know with our own sequencer do we Outsource it to Espresso and Caldera so that we operate at like a

at like a zero cost basis we're able to get high scale at relatively no cost and then just recoup the the friend Tech fees and redistribute those FR Tech fees right to our FR Tech users and to our friends like what you know what what are some of the considerations that uh you know higher up at friendch might be thinking about when they're when they're considering these these moves yeah the biggest trade-off that everyone needs to to think about when they're making uh when they're deciding whether to launch on a general purpose chain or to launch an app chain or an app roll up um is whether they get more out of the network effects of launching on some general purpose chain or like you know do they get more from those

Network effects or will they eventually profit more or or get more freedom from launching their own chain and of course you know in the life cycle of every successful project there will be a time when like that project crosses the threshold when you're early and you're new and you're completely unproven right like frch got a lot as a new project from launching on base there's a ton of liquidity there users already had funds on that chain like they uh you know it allowed them to grow super quickly to just have a captive audience of folks who are looking to try new applications on base but event like we've seen what that costed them right if they're able to replicate that user growth on their their own chain it would have you know would have been a little bit more painful maybe a little bit more friction

but it cost of them you know $10 million right and and th those costs are going to increase so the biggest uh thing that I think those teams teams need to think about every team it'll be a little bit different there are going to be uh protocols like you know Unis swap where like they get a ton of Ben benefit from being on a general purpose chain where like they interoperate with all these other uh protocols and are very tightly integrated um with with other applications um where where those Network effects are really strong there other things like games where those Network effects I think are pretty weak maybe the nfts will go onto openc or something but they don't need a ton of interoperability with the uh the rest of the the network and so there's there's that consider consideration and there's also like the profit side of things how

uh how much money can can you make by by launching your own chain and we've actually for a lot of the folks we talk to um when trying to sell them like we'll do this calculation uh with teams and a lot of teams like they don't realize how much money they could be making if they launch their their protocol as an app chain you think about all the amount of money folks are spending on optimism arbitrum mainnet ethereum like even if you're able to capture a small piece of that it it gets really substantial very quickly yeah when the rollup chain the rollup media education Good Vibes chain totally yeah you guys should should seriously think about it we are working on a rollup stack for now but that is under wraps got it let let's say we were going to to to you know

let's say we we said yeah like the rollup chain what you know kind of sitting sitting in between a lot of these intersecting Technologies like what what do you suggest for in this case the rollup or or any other project that's launching their own their own network and their own chain uh like what what do you yeah like what what do you suggest as far as like a first pass or like a just gener generally curious you know like as far as the considerations and then the the uh steps to launch your own chain yeah so we definitely uh recommend teams that are thinking about this to

talk to a provider like us um they you know we've done uh a ton of work in like evaluating different rollup Stacks uh in like figuring out the most efficient way the most cost effective way the most secure and reliable way to run those rollup Stacks um and we abstract away as I said like basically needing to like if you guys were were trying to you know launch your own chain you would need to hire like a devops team a protocol engineering team um you know potentially uh like a frontend team to work on things like we uh abstract away needing to hire like three different teams uh from the folks who are thinking about launching a roll up um so generally like in the abstract uh I think it makes sense to go with an infrastructure provider like us or or one of our

competitors I guess uh in order to like think through these decisions just because we've found that we we've taken like time to Market from you know what could potentially be years uh down to like weeks to months um in terms of like the actual like base layer rollup stack um we think in general like the existing rollup Stacks today uh they cover most use cases really well um and I think for a lot of the folks who are looking to uh to launch their their own chain they're not looking to make like protocol level uh Innovation they're looking to make like social or like business or or app level Innovation and so for those folks I would say go with like a tried andrue stack op stack the arbitrum stack uh potentially some of the new like ZK evm Stacks like like

ZK sync polygon CK evm like those are all relatively uh easy to come to an infrastructure provider like us and ask us to deploy and those are going to like cover you know 95% of use cases really really well um and then between those Stacks it comes down to like specific features uh which Community you feel more aligned with ZK versus optimistic kind of cost profile you want the chain to have there are like a lot of micro decisions but you know at a macro level just like it it always makes sense to go with the most reliable software so cool so cool no wonder there's going to be 10,000 rollups you've just you've just closed Robin ey on three minute conversation no I'm just around but but um yeah very very interesting thought

process to be having from a de perspective especially if you start to look at things from a yeah just from the economic business cost and uh liquidity composability needs perspective it's like do you really need to be in a general purpose execution environment like are you sure um but very very cool stuff um I know that you mentioned what's happening with uh the espresso team and I just wanted to briefly touch on that that update um one last time here so you guys are working on an optimistic uh decentralized shared sequencer set so fill us in on kind of like how that looks like you and espresso are

Partners in a sense or are are collaborators in this model of where they use you use their sequence sh sequencers in your stack of deliverables kind of what what is going on with espresso what's this announcement yeah like how does that kind of look yeah so espresso they're about red and butter is is is this uh shared sequencer Network so they've built out all the software for the network itself and that Network acts uh as the sequencer for transactions so users can submit transactions to that Network that Network were will order those transactions basically patch them up into blocks and that like that ordering will be considered the canonical

ordering of the rollup chain and so that's a network that you know completely handles that piece of the the modular puzzle um that being said like that is just a piece of the puzzle um and the biggest unlock for a network like that is to get integrated with the popular rollup Stacks I mentioned um so that user like I guess not even users necessarily but like developers and the people who are launching new rollups can easily plug into that Network and that's where we come in in that collaboration we helped espresso uh integrate that uh their uh their their infrastructure their shared sequencer Network into the op stack um so now we have a really tight integration with the op stack there uh and we also hosted uh a test

net for that op stack um on our infrastructure and so we're enabling it you know we're enabling users now to go check out that test net request funds from the faucet submit a transaction and that transaction will now go to the espressor sequence Serv um will be sequenced by them and then we come to the op stack node for for like final submission um and processing and so that collaboration uh we're handling a lot of like some of the software implementation and the infrastructure uh on the the rollup node and they're managing the software development and the infrastructure on the separate uh shared sequence or network that the node is plugging into and do they have anything live on mainnet Sir yeah they do not um they're currently in testet mode so early so we

are so early to the rollup Wars it feels like early defi days and it's just hitting me so hard on this pod right now oh man I this I I have a bunch more questions and I know we're like coming up on somewhat of time so I'm GNA save my my decentralized sequencer question for perhaps a conversation with espresso about how to actually like participate in those in those sequencings and and earn as a result um so I guess like the question I would like to ask is um like uh you you helped espresso integrate into the op stack so that when new rollups deploy using the op stack

they utilize this shared espresso sequencer so is there is there a difference between using like the C Caldera stack or like if I was to use the op stack and I would get espresso as Ed as as a part of that um um and let's say you know there's let let's say you have a future you have a future partnership with Celestia or igen layer and you help integrate those data availability Solutions into the op stack does does coming to the op stack given that I I have access to Espresso and and Celestia kind of give me access to everything I might I might need in terms of infrastructure or what else is out there that Caldera can can Loop into the

equation yeah so CA like we are mostly focused on hosting these existent Stacks that means we do a lot of protocol development on those Stacks that includes um uh like what you mentioned the Celestia integration the uh espresso integration we did that in collaboration with those teams um it also includes a lot of other things just nice improvements for like reliability for security um for making it easier to deploy these Stacks in the cloud I would say if there is to be anything like a Caldera stack um the what I would Define like most of our proprietary work being as like increasing the the customizability um of uh of uh like these rollup Stacks that exist so

previously it would be pretty difficult if you wanted to uh you know let's say do uh a network with like um all these like let's say custom pre-compiled and a custom gas token and using Celestia for da we have created interfaces that allow those customization points to be really quick where they're effectively just like toggles rather than you needing to do like a ton of work on the back end um plus uh you know just these additional like Integrations with other infrastructure providers um we we're not focused on building our own proprietary stack that is super Capital intensive takes a really long time and there are like all these amazing teams that have done this really well for us already so all of our work is mostly on the infrastructure and customizability side

and like we started this podcast you can you can do a lot of good business by bundling up these pre-existing good good infrastructure um you're either a bundler or an unbundle um and C therea bundles together some of the best web through infrastructure available totally 100% Matt I think we could go on for hours but um the good news is the def5 world will just provide us with another thing to talk about in like less than a week or two or three and then there's another big announcement and then there's more to to go on but to summarize here for all of listeners and um just everybody who's interested in rollups uh there's definitely pros and cons for uh

centralized and decentralized sequencers using Caldera to um jump start your own chain experience um definitely seems to be a uh you know a conversation willing to be had all right Matt and um just you know the technology with the shared sequencers espresso coming on uh live on mayet soon hopefully test it now the absolute cusp of this world that we're about to embark on in 2024 with rollups is it it this is this is going to be awesome and yeah I just want to say thanks Matt again for coming on and um yeah Rob any closing thoughts it's a it's been a pleasure

Matt I'll kick it over to you any closing thoughts or anything else you'd like to share yeah I mean I I'll I'll Echo Andy's enthusiasm it's it's palpable and we feel the exact same thing uh super early days uh but I think we're at like the the the point where we figured out how to do a lot of the table Stakes things like we know how to run rollups there are all these folks who built really great software in this space and now we're just waiting for the kambrian explosion to happen of all this new innovation and all these new chains um so cool yeah uh and yeah if I could sh Caldera a little bit um would encourage everyone to check out our website ca. XYZ or our Twitter Caldera XYZ um especially for teams or velers who are looking to like launch their own

chain we'd love to speak with them I think the folks who we've worked with have found us to at the very least be genuinely quite helpful in helping them think through stacks and think through these technical architecture decisions whether they end up launching a rollup or not so we we love to chat with Builders if any of them are listening please get in touch thanks for coming on Matt it's been a pleasure thanks so much guys see you thanks cool