Unlocking Modular Solana: The End of L2 Tradeoffs? | Anil Kumar

Hey everyone, we all know blockchain is changing the world, but who's actually driving it? I'm excited to invite you to the 6th edition of Meridian, a Web3 conference hosted by the Stellar Development Foundation in London from October 15th through 17th. Get early bird tickets at meridian.stellar.org with code BLOCKWORKSPOD. Special thanks to Stellar for sponsoring us. Now let's get into today's episode. Everyone knows there's a massive shortage of top senior marketers in crypto, and teams are constantly struggling to get their go-to-market done right. That's why we're excited to have partnered up with Renaud Partners, the premier go-to-market advisory firm for early-stage crypto teams. Get in touch with their team by heading on over to RenaudPartners.com or giving the co-founder Jeff a follow on X. Check out the links in the description below. This episode is brought to you by Avalanche. Avalanche is a leading network of blockchains giving some of the world's largest financial institutions new use cases and business models while leveraging completely customizable infrastructure. Check them out. It's core.app/discover. You can explore Avalanche's subnet ecosystem at core.app/discover. I was playing around earlier today. Really cool. What's up, everyone? Before we jump into today's episode, I'm excited to share Empire's first ever security partner. Harpy is the best tool to prevent your wallet from theft in real time. Harpy is not just a security solution. They are a peace of mind solution. But don't just take our word for it. Harpy is the only wallet security solution that protected 100% of its users from attacks like the Ledger one in Q4, which was an off-chain signature attack. To learn more about Harpy, click the link in the show notes or visit harpy.io/empire. Empire. Welcome back to Empire. Um, we got a lot of really good feedback on the episode, uh, with SkyTrade. Um, Santi and I are trying to bring some newer founders onto the, onto Empire, as opposed to just the— as much as we love, you know, Ave and, uh, Leschner and Stani and Hayden and, and that crew and kind of the OG founders— trying to bring some new people into the fold. And so we got a lot of really good feedback on the last episode, so we wanted to bring Anil from Roam onto Empire. Anil, welcome to the show, man.

Oh, thank you. Thank you for having me. It's a pleasure to be with you too. Looking forward to—

Yeah, yeah. Excited to have you. Okay, so for background for this episode, guys, we did an episode with Josh Bowen from Astria and Ben Fish from Espresso back in May of 2023. And the title was How Rollups Will Decentralize Their Sequencer. Roam has come onto the scene. They announced some funding recently. Santi and I are both investors. We've had kind of a sneak peek at what they've been building. It's really, really cool. I would say my one sentence summary is Rome and Anil and the team are basically helping L2s access some of the best parts of Solana. So just want to have Anil on. So maybe Anil, I think a good place to start because a lot of people might not be familiar with you and Rome. Can you just give us your background on like how you came to build this and I guess really importantly, like why you're building this?

Yes. So I was involved in crypto like from long time. I come from a background, JP Morgan banking background, but I was always interested in crypto. When Bitcoin first came out, I had set up mining server in my basement, though like everybody else, I didn't keep track of it. Then continued engagement, but I got very seriously engaged in crypto in 2017, '18, '19, where I was investor building some application just for my own use. Some application were more native in crypto than others. Some were more Web2 and Web3 combined. And then I got into looking into Vitalik's roadmap for rollup, how Ethereum is going to be scaled via rollup technology. And that caught my attention and then started looking around and started poking around what kind of problem. I mean, some of these problems were very obvious. Defragmentation, fragmentation of liquidity, fragmentation of composability, how much security and how you're going to secure the L2s in terms of what it executes off-chain and how it goes on-chain. So these were like some of these problems were very obvious and then some of them were not obvious. Like why would L2 actually use these that we are seeing today? And got engaged in this and started working on looking at the solution, what might be the solution. And it was very technical in nature. I come from a technical background, though I have had multiple hats throughout my life, from technical to product manager to a number of others, investors and stuff. So I kind of understood the kind of a road for the whole integration. And then I also understood the problems that what meaningful way these problems are, how meaningful these problems are for different participants. And so I worked with my co-founder Satvik, he, and came up with a number of solutions and we bounced it off a number of folks, Josh, Nick White, other people, Tarun, and then we realized what we really wanted to do was use Solana, which provided all these different properties, decentralization, speed, throughput, security of the network. And along with that, it solved a problem that nobody is addressing, which is interoperability between different chains. Yes, we can have Layer 0, we have Omnichain that are solving that at a distant level where you are able to do composing of transaction with different chain, but not at a very, uh, joined to the hip nature like atomic transaction. So it provided us an opportunity to at least solve this problem from Ethereum and Solana integration point of view, that these can be joined at a hip, uh, via the sequencer. And that's basically was a major impetus for us to basically use and create Roam to solve that problem.

Is it fair to say that, okay, so you had Ethereum, Ethereum started to scale, then you had this kind of like modular narrative that's popped up and Celestia and all the whole kind of modular ecosystem. It feels like you guys are trying to kind of modularize Solana. Is that the right way to describe it?

That is correct. So over a period of time, we realized what actually we were doing. So we started out as a sequencer. Shared sequencer and create more interoperability while solving the sequencer problem. What we realized was that Solana is much more than just a sequencer for this role. What we can provide based upon Solana's properties, the, some of the best properties, which is its consensus layer as a set of services for L2s. And it is not now limited to Ethereum L2s. Any L2s can use these services. And what we are building is Shared Sequencer is one of the services. Other services are that we are bringing out along with the Shared Sequencer are DA. So you can use Solana as a DA. We are also bringing out a fast finality layer where you have commitment via the AVS between confirmed Solana transaction and the finalized Solana transaction. So L2s can get much faster commitment of a state at the speed of Solana blocks. We are also bringing ZK coprocessor and number of other services that would be helpful for L2s to scale, which would provide like the aggregation layer. So they can get interoperability not at the atomic level between cross-rollup and with Solana, but also with the chains that are outside of this ecosystem. Like you have Arbitrum, Optimism, zkSync, any other chain that lives outside, you can get interoperability with this layer by sort of these services that we are creating. So now our narrative is more of a modular Solana services for L2. That is what we are creating.

Okay. Potentially a dumb question here, but I think just maybe a good thing to do a reminder for both me personally and maybe for the audience. But why does— like, tell me about this idea of a sequencer and what a sequencer even does and why it's important to decentralize it. And I know Santi will roll his eyes because that's such a 101-level question. But maybe just let's go there for a second.

Give me too much credit.

I think—

yeah, I think the basic function of a sequencer is at its core is one single function. If it provides that, you can call it a sequencer. Which is ordering of transaction. If it can order the transaction, it does not have to do anything. It does not have to provide commitment of this. It can, it does not have to execute these transaction. So all it has to do is order the transaction. And after that, it is up to the individual rollup to execute those transaction, do the state transition, anything that need to be done. Sequencer theoretically is not responsible for that. But If you are going to be more than just ordering transaction, because ordering transaction is like you have no other skin in the game. All you are doing is ordering and you can take the MEV, you can do a lot of other things. So you have a lot of power without doing too much. All you are doing is ordering. So sequencer need to have a little bit more than just ordering. So what we do is we provide, so in Roam, we are more of an executing sequencer. So we are not only ordering the transaction, we are actually executing the transaction and providing that commitment of ordering and execution to the end user and the rollups. So that is why it is an important function. Today we see that most of the sequencer are centralized sequencer. And I think there are a number of reasons for that. L2 does not want to give up because there is a lot of MEV that gets taken out by the sequencer, by ordering of these sequencers. And once you give up your sequencing to third party or shared sequencer or decentralized sequencer, that means you have no control of your MEV and somebody else, some third party is extracting that value. As an L2, you would like to keep control of your MEV because you are providing services to your user. and you should theoretically have control of these MEV. So, so these are some of the problems, and it is very important in terms of importance of why decentralized sequencer versus centralized sequencer. Centralized sequencer has censorship risk. It can go out at any time. You have no— we have seen these instances happening in past with some of the centralized sequencer and So we need to address this problem. Today, if the value of the chain is low, I think people are okay, we can go along with this. If we are going to attract a substantial amount of value to the chain, people want to know tomorrow I am not going to be left out of, I will not be censored, I will still continue to withdraw my money, I still would be able to process my transaction. The L2 would survive the sequencer, tomorrow government is not going to come after somebody and take over that sequencer. So you need to have a decentralized sequencer, not just from your own, that the sequencer itself might go rogue, but also that third party like government entity can come in and seize the operations. It is much harder for government to take control of the decentralized system compared to any centralized where a single person is responsible. So that is why it is important to have a decentralized system that is ordering your transaction rather than a centralized system.

What's the intuition, uh, to using Solana for the performance of the sequencing where you can— and why might other teams that Jan alluded to say, hey, It kind of, you don't kind of need, you don't need to go to Solana. You can just do it as we're doing it in Ethereum and like, why bother with the extra complexity?

Right. So if we look at any sequencer so far in the market, I find that it has to create a substantial amount of security that orders the transaction, that there is some force, there is some value behind this. So all the transactions need to be secured. With certain amount of security. And that would take a lot of time for a chain to accrue that value to be able to get up to a place where it can say, if it is taking a transaction of $1 billion, I'm just taking hypothetical, and it is secured with the, let's say, $100 million value of the chain security in terms of ordering a transaction. That seems like you should equivocate and order it wrongly so that you can extract that $1 billion. So you want to be able to do a transaction a high-value transaction, you would only feel confident when it is secured with a substantially higher amount of stake. So that is one reason, uh, you can bootstrap a chain and come up with the hundreds of billion dollars worth of value to secure that sequencer. So sequencer at the end of the day is an L1. So you would have to come— so that is one of the— along with that, you need throughput, you need speed, and you need well-tested system. It takes a time. It takes quite a bit of time for an L1 to mature. It has to go through a lot of issues, bug fixes. Solana has seen a lot of these issues and it has fixed those issues. And it is not like it will not see some issues. It will see some issues, but it is much more mature than any other L1 that you can bring up now as a shared sequencer or a sequencer. So that is the intuition behind this, that it has existing stake, well-tested, sufficiently decentralized. So why not use it? It has an EVM. We have built our own Roam EVM. So you can run different smart contracts, you can run different, and you can run it at a very high speed. So that basically all lead towards that, okay, all roads lead towards this, that, okay, we need to we can use Solana as a state machine and we can modularize this consensus layer and provide this shared sequencer as a service on Solana. Yeah.

What are the— yeah, as Solana becomes more and more popular, you know, you had like OR for instance that, you know, created a bit of an issue couple months back in terms of identifying something that you're stress testing the network, like, you know, failed transactions was a thing. How, um, how do you think about, uh, a lot of what we talk about on the podcast is, you know, the decisions that teams might make of where to build, uh, from like a bare, like, L1 kind of standpoint. Uh, no surprise, you know, we talk about Solana a lot on the podcast. We also talk about Ethereum, L1s, L2s, L3s, and so on and so forth. What are some of the things from your vantage point that inspired you to consider Solana versus maybe other high throughput chains out there?

Right. So we considered a couple of them. SUI was one. First was like, okay, find the chains that are high throughput, one of the best in terms of state machine. SUI, Aptos, they come close. In some ways they are better, in some ways Solana is better. When it comes down to, when we look at the security of the network, Solana has much more security than any of these chains in terms of the amount staked, in terms of validators, number of validators that are participating for ordering of transaction, execution of transaction. 1,500 for Solana, it's substantially order of magnitude more than other chains. So it has a sufficiently decentralized system, maybe not as decentralized as Bitcoin and Ethereum, but it is still a sufficiently decentralized system than any other chain that exists with the throughput. And plus the roadmap ahead of Solana is much more exciting in terms of FireDancer coming in, in terms of different improvements that are coming in. And so that basically lead to that we can scale along with Solana. Once we have the sequencer, you have a great amount of throughput today and it would continue to increase as the time go by. So you are not going to be restricted in terms of future, how the state machine is evolving and be able to address the needs of the L2s.

Anil, I have another, what I'm sure is a really dumb question. Again, I'm going to ask a lot of those this episode. Why would you want to hand off, if you're like an L2, why would you hand off the sequencing to a third party, whether that's a Roam or Espresso or Astra or whoever it may be? Like, why are you handing off kind of the crown jewel moneymaker to someone else?

Absolutely. Yeah, exactly. So I mean, I think that is the crux of the problem for the L2s. And rollups. They will not. Simple answer is they will not, right? Why should they? Unless their users or system forces them to do that, they will not do that. At the end of the day, they are losing substantial amount of capital for something their user may or may not be complaining about. So that is why we think we have to create this in a capitalistic way. The system has to be capitalistic way. So when we, when somebody uses Roam as a sequencer, they do not give out MEV. They are in control of their MEV. We do not extract any value. We cannot extract any value because the system Roam is designed that way. You are in full control of your MEV. So that is why we think we need to solve this problem. The sequencer need to solve this problem. Yes, I know there are in industry, there are other sequencer who are like, okay, we would share the MEV between us and rollup. But this becomes very semantic and very operational driven. Like how do we calculate and stuff like, why not build it in such a way that you as a sequencer, you are providing service and that service is just for ordering transaction and everything else belong to L2. You provide additional set of services like we do in terms of cross-rollup because you have to incentivize L2s to use you without taking away anything from them. So we do not take away their MEV rights. They have the full right. They control that. But we provide additional services which help them in terms of cross-rollup atomic transaction. So they can do their users or ecosystem participant. There are more apps that can come in. By using cross-rollup atomic transaction. We provide composability with Solana. So you have access to Solana liquidity. When you go join Roam as a shared sequencer, you get immediately access to Solana liquidity, which is a big boost. So it is like bootstrapping a rollup to access Solana's liquidity as they start using Roam without giving up anything. They can start using all the services. And access to interoperability with Solana and other rollups.

What is the business model of Roam, Anil? Like, how do you guys make money?

Right. So one is transaction fee-based. So we would take percentage of transaction fee that it costs to execute your transaction on Solana. Second is premium services. So we do have premium services which allows you to access the state in a premium way. You can do— so we are working on certain other aspects of how you can execute transactions in a faster way on Solana. Signatureless, gasless, so different ways where you would have a ROAM token or you would pay premium pricing. For these services. And we would allow you, you would have some templates that you can run for your trading application. If you have a certain template between rollup transaction, Solana transaction, we would store these templates on Solana and you are able to just submit the input parameters. And these templates would automatically create transaction that you have previously stored and execute this transaction. So this would make your transaction even faster in terms of taking advantage of whatever discrepancies exist between different rollup states and arbitrage. So these are premium services. So these are some of the premium services we are going to bring on top of the ecosystem that we are building. And all these services would be accessible via Roam SDK. So Roam SDK is our main, one of the main product alongside Sequencer, Roam EVM, Fast Finality Layer, Aggregation Layer, zkCoprocessor. Roam SDK allows you to access rollup states, Solana state, compose atomic transaction, cross-rollup, cross-chain atomic transaction, access these premium services, and basically use and create application much more holistic way by accessing the states.

When you think about who Rome is competing against, what do you think they got right and what do you think they got wrong?

I think, So we compete in like different markets. So we are not just a shared sequencer anymore. So I would say we are a shared sequencer is just one of our services. We have a RoamEVM. So anybody who wants to deploy a dApp on Solana, but still have the Ethereum dApps on Solana, but still have the front end of any Ethereum client like Opigeth or Orbit, you can run those client and still access RoamEVM. You can deploy your home media. So that is one of our services. So DA, fast finality layer. So we compete in different spectrum. So in sequencer space, we are competing with, let's say, I mean, I think Espresso is more of a sequencer network and we would love to participate in their sequencer network, Astria, Radius. So these are like in terms of shared sequencer where we are competing. Then in terms of our fast finality, look at Nuffo. Those are the products that we're competing. In terms of Roam SDK, which is more of a composing transaction cross-chain and cross-rollup, that is more of an omnichain and layer zero kind of space where we are operating. So I think different components of Roam are competing with different projects. So I'll go with first taking up with the sequencer space, right? So I think in the sequencer space, I think maybe we were at an advantage because we were able to see what Rollup will or will not do, what the problem would be, as opposed to, let's say, Espresso, who started like 2 years ago. So we were able to learn from some of those things that we saw in the market space happening. One was like, okay, decentralization, security of the network, throughput, uh, exist— it takes time for a chain to mature and your consensus layer to mature. Because I see it is mostly about consensus when we are talking about sequencer. So Espresso's consensus layer based on the hotshot. So I think these are some of the things that we learned, like, okay, why not use bootstrap? So I think these were some of the things. In terms of like L2s, what L2s are willing to do or not willing to do. Like they are not willing to give MEV unless their hands are forced. So why force their hand? Let them keep the MEV and solve the problem of decentralized sequencer. So in hindsight, we were able to accommodate, move, and be able to adjust our sequencer in a way that solves these problems as we saw them. So I think in a sequencer space, I would say we were beneficial of seeing what was happening in the market and adjust ourselves and be able to highlight and accentuate these kinds of solutions to the L2s and the end user.

Well, maybe I was jumping too ahead because I know you have a very ambitious kind of roadmap, Anil, like not just sequencing, but like there's this broader kind of ability to modularize, bring on, as Janno said in the prelude, like the components that are really good of Solana into an environment that like Ethereum, I wanted to just maybe, and Janno, cut me off here if this is like going too far ahead, but I wanted to understand like your broader kind of vision for, okay, it starts with sequencing, you prove that out. What else? Like what, what, how can we, how do you envision kind of the architecture of blockchains kind of writ large of like evolving, picking and choosing, you know, some of the best components?

Yeah, I think I might be crazy in saying and thinking, thinking what I'm thinking. But then, then we are in this world that is a changing world. So, so the way I look at it, so this is if you look at it, Solana and Ethereum, right? Combined together, you can do atomic transaction with Solana and Ethereum rollups. And now you can see that Ethereum and Solana are kind of joined at the hip. Now, and how is that enabled? It is enabled by the set of services. Shared Sequencer is one of them. And then you have other services that, that the modular services. So that creates a join at a hip interoperability. What does it do? It creates a surface area for application to be built. Now people can take best of both world, Ethereum and Solana, and create application, single unified UX. Which is an atomic UX that you can create application. Now, you— we take this as one set. Now you take Ethereum and another chain, Ethereum and another chain, replicate the same thing that we are doing with Solana and Ethereum. Do that with all other chains. Now what you have is kind of a unified blockchain, at least with Ethereum. So that is kind of the way I see how you unify all the blockchains, maybe using Sequencer and creating these set of services and that create at least somewhat more interoperable world where you can do atomic transaction, trustless transaction, non-custodial transaction. So that is the kind of vision I have. Maybe I'm very stupid, but That is how I'm thinking about it. That, okay, we do this for Solana and maybe somebody else would do it for pick up another chain and create the same kind of solution for some other chain and Ethereum.

Hey everyone, Jason here. Just wanted to take a quick break to talk about something exciting in the world of blockchain. The Stellar Development Foundation is hosting their 6th annual Meridian Conference this October 15th through 17th in London. It's a 3-day event where the brightest minds in Web3 come together to discuss everything from tokenization to DeFi. If you're a developer, builder, policymaker, or business leader looking to make an impact in transforming global systems through blockchain, this is your chance to join the conversation. You'll get to network with the forward thinkers who are defining the future.

Of this space.

Head over to meridian.stellar.org and use the code BLOCKWORKSPOD for early bird pricing. Now let's get back to today's episode. Hey everyone, we all know that there's a massive shortage of top senior marketers in crypto. We've seen it over and over that teams are constantly struggling to get their go-to-market done right. That's why we're excited to have partnered up with Renault Partners, the premier go-to-market advisory firm for early-stage crypto teams. They help top founders tell their stories better and build great non-technical teams. They advise founders on all things marketing, including tight brand positioning, community building, social media, and building A+ internal teams. Don't just take it from us. They work with some of the top projects across Solana, Monad, Base, and many more. If you're a founder or a VC looking for support for your teams, I highly recommend connecting with them. Get in touch with their team by heading on over to renaud. Partners.com or giving the co-founder Jeff a follow on X. Check out the links in the description below. This episode is brought to you by Avalanche. Avalanche is a leading network of blockchains, giving some of the world's largest financial institutions the design space to explore new use cases and business models while leveraging completely customizable infrastructure. Everything that Avalanche has been building the last few years has culminated in this moment where today you have institutions and developers able to essentially have limitless possibilities with the chain. They've enabled developers to build cross-chain swaps or create gaming environments where one subnet can unlock features in another or can use USDC as a native gas token. Onyx, for example, by JPMorgan or Citi or DTCC's digital assets or the Australian New Zealand Banking Group. They're all building on Avalanche evergreen subnets. Really cool stuff. Big thanks to Avalanche for sponsoring this episode of Empire. Go explore Avalanche subnet ecosystem at core.app/discover. That's core, core.app/discover. Thanks again, Avalanche. For a lot of Empire listeners, your crypto is not just another number on a screen. It's part of your future. I know Santi and myself feel that way. Our security sponsor of this episode, Harpy, takes this responsibility seriously and is the only wallet security tool that shields users from both on-chain threats and sneaky off-chain signature attacks. If you've ever been in that situation where you're moving quickly, you approve something on-chain, you realize that the address might be a dubious address, or you're really hoping that you could take that back, Harpy has you covered. Harpy can redirect your assets to your self-custody vault, ensuring they remain completely under your control, safe and sound. With Harpy's always-on monitoring, you're not just detecting threats, you're actively blocking and recovering compromised assets from malicious transactions before they can even confirm on-chain. Harpy is the only wallet security solution that protected 100% of its users from attacks like the Ledger one in Q4, which was an off-chain signature attack. So if you're serious about protecting your crypto investments, it's time to make the switch. Secure your wallet for free at harpy.io/empire. That's Harpy, H-A-R-P-Y-E dot I-O forward slash empire. You want it to be even easier?

Just click the link in the show I want to revisit a concept that you said earlier, which is you kept saying you keep the MEV. Uh, listeners should go back and to that discussion that we brought on Kyle from Multicoin, and we were talking about like what is the most valuable component of blockchains. L1s is MEV. There's different types of MEV, right? Malicious, toxic. Hopefully you clear that out, stuff like deflow, stuff like that, you know. But still MEV is going to be a far greater revenue stream than fees themselves. So I guess that's why sequencing becomes so important, right? The ordering of the transactions. And maybe I'll just hand it over to you, like more so the question is like, why is the sequencing kind of that thing that you've decided in this modular kind of stack to go after? Is it because if you have because it's like central to MEV and that becomes like the most important thing? Or is that just a beachhead into other stuff?

Right. So yeah, I mean, so MEV is a major problem. So we used to have all kinds of different— so Ethereum has put in some rule like, okay, per block, that is the maximum MEV you can extract. So we are putting— I mean, I would say they're going too far that we are putting Band-Aid. Instead of solving the real problem. Like, how do we solve the real problem, right? If, and when, so what is a bad MEV and what is a good MEV, right? So let's say bad MEV is where you are taking a transaction and putting it in such a way and that it gets worst execution or bad execution, that you are layering it in a way that puts it at a disadvantage. So that is bad. You should not, nobody should be allowed to do that. If a transaction comes in, in a natural way, it comes in at a certain place, it should be honored at that place because that is how it ended up being there, right? So we all like need to be cognizant of the fact that we do not allow, system itself should not allow layering of transaction in a way that it puts it at a disadvantage. So like a sandwich attack and this kind of attack. So we all know about these. I mean, these are like old stuff, right? Now, what is a good MVP? Let's say arbitrage, right? Arbitrage, you can say, is a good MVP, like cross-chain or between different DEXs, because it creates a better price discovery. It brings the price in harmony across different users and everybody benefits from that. So arbitrage, I would say if a price on one DEX is different than another DEX and you want to put that transaction as a cross-atomic transaction and send it, it is a good MEV. I mean, it is beneficial for everybody.

Financial markets. So you're creating tighter spreads, more efficient markets.

Right. So that is how I look at it. If it benefits everybody, it is good. If it is disadvantaging a particular a transaction, then it is bad. So that is, and how, and that is exactly the way we are approaching how we have designed our sequencer, right? So you cannot take a user's transaction and layer it and put it at a disadvantage. It is impossible. Because how the system works, it is as the user transaction come in, it is picked up by RIA and it is sent to Solana. Where it is encrypted and it is executed by the rules of Solana. Nobody can see. Yeah.

And to your point, it comes in code, like it's just, no, you know, it just gets sent in whatever order. And your point is more so you're not necessarily solving MEV per se. You're just creating stronger guarantees that there's no one single party that can manipulate that order flow, that cannot manipulate the ordering of said transactions. How those transactions come in might have some MEV attached to them, to your point around solving some like an arbitrage type of trade or whatever.

Yeah. And that we consider as a good, because now you can take, okay, you see a price on one rollup different or one DEX different than the other. You can take these two transactions, put them together as a single atomic transaction and you can send to Solana. It is executed as a single transaction. Either both the transactions going to happen in a trustless manner or neither of them is going to happen. So that is a good, because it creates the price discovery across the system. So that is in our view is a good MEV and it should be encouraged and allowed and people should be able to make profit with this. Yeah.

How do you, maybe this is a very basic question. So the transaction comes in, say from an L2, comes in, gets routed to Solana, gets picked up. At that point you're competing on other stuff in Solana, right? And now there's, you know, tips and you have, you know, how do you think about that flow, especially that last part, the last mile, like once it enters Solana, then how it gets prioritized, if you will, or not. Like you must have some visibility into, I'm sure like, say that there's like a lot of activity in Solana, you know, and how do you work through that preemptively or mitigate kind of some sort of failure? I don't know if this is actually possible, but I don't know if the question makes sense.

No, it does make sense. So basically, so rules of Solana applies to all transactions. So if a user wants to send with a higher tip, their transaction, they can. So if they are using, let's say, Roam SDK, and they want to send a transaction with a higher tip so that that gets executed, yes, you can do that. Then it comes to, in terms of hotspots, right? Because each L2 is its own smart contract on Solana. So the hotspots are all distributed. It is not one single Roam contract where it would create the hotspot. If an L2— so the L2 genuinely may have a lot of activity and it may slow down the processing. So same rule would apply that even if the L2 was operating by itself, it— and there were a lot of transactions, but it is not being clouded or it is not being disadvantaged by some other smart contract. So it is applying within— it is playing within its own rules without being impacted by others because it is its own smart contract. And the hotspot remains within that smart contract, that EVM, rather than it is impacting other EVMs.

I hate to put you on the spot, but I'm going to say it anyways. Feel free to deflect the answer because it's hard. But say we had Ben or someone else from another shared sequencer. How would you think these things stack up? Like, you go out there, you go talk to an L2, and you say, listen, 'You ought to use our solution because,' yeah, what are the key selling points? Performance, security, you're leveraging a Lindy chain, you know, a Solana versus trying to bootstrap your own validator like the others. What do you think matters most for these teams, these like L2s, like Arbitrum or a, you know, whatever, pick your L2 of poison.

To me, it seems like they should be, I mean, it should matter to them that the system is decentralized. It has high security, immutable. So that, and it is that they're not, I think they should not feel that they have to give up anything to gain these properties. Yeah. And then running cost, and, and, and that should be comparable. I think if they have to pay substantial amount of money to get this access, I think they would probably equivocate whether they go with this or not. So I think that cost has to remain low, and, and we feel Solana is balance of all these things to provide low-cost, immutable, decentralized, high-security ordering of transaction. Yeah. So I would say like, this is what you would get. You will not give up MEV. You would gain interoperability with Solana and other rollups. And you get far more services from the ecosystem, fast finality layer, aggregation, and you still have the settlement on Ethereum. So, and if you wanted to access, so we are going to announce, maybe this is alpha, we are partnering with Salesia to provide the layer, those rollups that want to use Salesia for DA. So that announcement is going to come out.

That's more on the cost. Yeah. Just— yeah, that's really interesting. Across those dimensions, what do you think resonates the most for teams making that decision? Is it, you know, is it the security, the I would argue that like for others, if you're trying to, from the ground up, build a like decentralized layer of sequencers, like that's a, that's just less proven, right? Yeah. And the security is, you know, just lower, right? I mean, I guess there could be a minimum viable security that you could argue is enough and you get comfortable with that, but certainly not going to match Solana day one.

Absolutely.

Across Nakamoto consensus, stake, whatever you want to— security is a broad-based term that we use constantly in the industry, but we don't necessarily agree with, right? But Nakamoto consensus for all intents and purposes, right?

Yeah, I think that is absolutely correct. It takes a period of time from you start to get a chain to a level where Solana is today. Or even less than Solana. Let's say, okay, we don't really need that kind of security, that much security as Solana is providing, right? It still takes for chain to mature, to remove the issues and have enough conviction in the market to stake enough money in that chain so that it is securing that network, the security of the network. It is not on a day one or even one year later that you can create the kind of commitment. Because if your chain is only $10 billion worth, if your sequencer is only $10 billion worth, it is the value that is securing your sequencer network. Then it would not be possible to have $100 billion worth of other chain using that as a shared sequencer. I think you would have to, you have to have at the end of the day, look at this in terms of what amount of throughput is going through the sequencer and what is the security of that network, sequencer network that is securing that throughput. Yeah. Because if there is a chance for somebody to fraud, it would fraud. We have to go with that, that it is possible that if somebody has a control to fraud, then they would fraud if there is enough economic incentive is there.

Well, let's actually go there because I think, um, You know, when L2s first launched, like there was this criticism that the sequencer was centralized. And I think this goes to the heart of, I'm not so sure people really understand what a sequencer like can and can't do. 'Cause you mentioned fraud, right? Like the worst case, if a sequencer is centralized and, you know, acts maliciously, we can talk about what that actually means. Ultimately, if you have fraud proofs, you can always go back to the main chain and like recover your money, so to speak. You can correct me here if I'm wrong. Like you can, you can ultimately go back if you have fraud proofs, right? You can say, hey, look, it might take a while. There's a period in it, you know, that in, in that period you might get liquidated, a whole host of problems, right? Of liveness. But you can go back, right? And if you don't have fraud proofs, that's the main issue, right? Of an L2. If it's a, if you don't have fraud proofs, then, you know, the security of that L2 is like nonexistent. You're really trusting that, right? Because then that's problematic. So what is the worst case scenario here for someone that might say, look, Anil, I hear you, you're using Solana. It has far more security than probably you're going to need for a while and the performance and the cost is going to be there as far as we can see, especially with Firedancer. What, um, like, is security a, an oversell? Is it something that like, it's more so like how you partition the economic gains that are being captured by one entity into hundreds or whatever by decentralizing. But from a security standpoint, it doesn't really matter much. It's just like, who captures that fee? Who captures that, right? I'm trying to play devil's advocate. I'm curious if—

No, no, this is a very good question. Uh, so yes, with fraud-proof, uh, maybe somebody would pay you, uh, somebody should pay you if there was a, uh, you, you provide the fraud-proof. But I think the security is not just that, uh, that you get your money back, right? Security is also censorship. Somebody can block your transaction, right? At a sequencer level, if it is a centralized sequencer, they may not let your transaction go through. They may hold it for a period of time and then your transaction fail.

It's a holding hostage problem here.

Yeah. So, and if your transaction doesn't go through at that time, something bad was happening somewhere. Like say some world was blowing up somewhere. And you could not get your transaction through, that could be a substantial cost to you. You may be trying to get your ETH over to Ethereum and put it on some SACS and sell it and whatnot.

You might eventually be able to bring it back to Ethereum L1, but you're going to be liquidated and the value is zero.

Exactly.

For all intents and purposes, practically it's the same.

Exactly.

I see your point.

Yep.

Let's transition more into more fun stuff. So like you've been at it for a while. Like, you know, you, uh, you know, when I first met you, as soon as you said JP Morgan, it's like, all right, well, you know, We saw eye to eye then and there. I committed on the spot.

Yeah.

As someone, you know, you've been building systems for a long time, maybe just more transition the conversation into like, what is your general interpretation of crypto? Like, I feel like a lot of people might be listening, hearing this video and be like, identify with you. You know, they've been at Cisco, they've been like building systems for a long time, and they may or may not be on the fence about crypto. They may not like it. They may have really strong opinions about this is just a database and there's no need to kind of work through all of this. But like, what really drives you to build in the space? What motivates you to come in and do what you do when you can be doing so many other things? And then the second part of the question is like, what has surprised you the most about building in crypto now that you kind of took the leap forward? You've been at it for a while, right?

So intrinsically, I mean, I'm I'm attracted towards decentralization, free market, free, not being controlled by, like you can do just like anybody else what they can do, right? So I think crypto is kind of a sense of that. It is a distributed database of records, right? And anybody can participate, anybody can build. It is a free enterprise, not controlled by, theoretically at least, I mean, there are some controls here and there when we look at some things, but it is not controlled by anybody can participate. Like I can come in and I can start building my enterprise on this blockchain, on this ledger. So I think that kind of like, and it provides like, it is a frame. So if we look at it, right, I've been in the corporate world for a long time. right? Where your tunnel vision, you are looking at one thing and at the end of the day, you feel constrained. You cannot use all your talent, your energy into the way you see things, but crypto allows that. So it's kind of freeing. It is a freeing environment. It is like a freeing of a human spirit. right? Where you can come in, start doing things without being constrained by anything, right? And you can, you can, you can create it and you would see result. You create good, you get good. You create bad, you get bad. I mean, it is reciprocal. It is not that, uh, uh, anybody can just, uh, succeed by doing bad thing, right? Though there can be some instances, right? So the way I look at it, so this is kind of a freeing of a human spirit. That is the kind of vibes I get. When I am in crypto. When I was outside crypto, I saw myself constrained. I saw myself like put in a box. So I think there are more applications, more things that are going to come in and it is going to be more freeing for more people. What the challenges are for our industry? It is very technical. It is very technical, even for a guy like me who has two master's degrees in engineering. I find it sometimes like, okay, I need to sit down and think about it for weeks before I understand certain things. Right. So how do we make it, how the knowledge is dissipated so that other people, other folks can join in this journey? right? Make it simple. Get more educational content out. Get more simplified educational content out. So I think that kind of things allow people to like, today we have like my 11-year-old, well, when he was 7. So he started building games using components, blocks, different things that he could create games and then play those games. So like in Roblox and stuff. So we need to make like some of these things in crypto where people can build these application just like that. Right. So people can come in, play around and those kinds of things. So I think that is the kind of thing I see that, okay, where we have challenges that it requires almost a PhD for you to understand sometime some of the core concept. So there need to be more people creating easily understandable contents. And allow people, I think the access to the blockchain and everything else is there. I think that if we can address, I think that would go quite a long way in terms of getting more people into crypto. Like UX, we talk about then this education, how people can simply come in and build application. So I think that is where I see. And I think more people experiences, more freeing they are. I think more they will talk about it.

Yeah. Yeah, definitely. I very much agree, even though I'm not a builder developer myself, that I sort of would push back on this common, more recent criticism that there is too much infra and not enough applications, because I still think that there is different parts of the kind of infra cake that still still need to be built, far more connectivity and interoperability. That middleware layer, I think, is very much unexplored, very much still nascent. And it sort of makes sense. Like, we didn't— we weren't talking about modularity a year ago. We now— that sort of— that unbundling of blockchain functions has— requires more investment and innovation at that layer, right? And exploring all these different combinations, permutations. We might, you know, rebundle into certain standards and certain best practices. But, you know, as far as I can tell, the tooling is improving.

Not—

not— still not there. Um, you know, more languages for support, more, you know, exploration at the, you know, you know, the modularity layer in particular is something that and interoperability layers is I think the areas that, that plus like optimizing fees and how you price blockchain space is like still requires and will probably require many more years of experimentation. And so, yeah, like I think that's just my two cents on what you said.

Yeah. I mean, infrastructure, I mean, the criticism is not well-founded of like too much infrastructure Because infrastructure is needed. Once infrastructure is there, then there is a lot of innovation, a lot of ability for creative thinking. When the cost of doing things is low, that's when people start to do things like, okay, when the roads are there, when the different infrastructure is built, when it is cheap, people can experiment and create things. And it is absolutely, I mean, I think needed and more infrastructure. I totally agree on the interoperability part where the infrastructure, the plumbing, all those things that need to be there. So that is where we are doing our bits, our two cents in this piece where like, okay, we provide some atomic composability with Solana and Crowswala. So I mean, that is just our half a cent in there.

Absolutely. Well, folks, I think it's been a fascinating discussion. I'm really excited to to see what emerges from, as Janno said, really leveraging some of the best components of Solana into a much-needed kind of area of infra in the L2 landscape, which is a sequencer. And so it's something that has been talked about since forever, as soon as L1 started to exist. And so seeing how these two kind of ecosystems merge together as you're doing it with Rome is pretty fascinating. So just to wrap it up, Timeline-wise, when can we expect to see some— I mean, you leaked some alpha, but like what's next on the horizon? And then where can people learn more about the project and get involved if they want to?

Yeah. Yeah. So we are gearing up with our DevNet release so it can come out soon. Hopefully in August itself, we would release our DevNet.

And generally we are in August.

Yes. And then hopefully we can get our mainnet out early next year. Okay. Like January, February, when I say early Q1 '25 and testnet from now to then. So we were actually ready to do our devnet, but we decided that we needed a better EVM. So we created Rome EVM in the process. And so our DevNet is going to be based on our own EVM, Rome EVM.

Fantastic. Well, very exciting, Anil. Best way to get a hold of you, the project developments, Discord, Twitter, Telegram?

Right now it is Twitter. So we just got a community because right now we were, so far we were heads down building. It was all engineers building, building, building. So we just got a community manager hired and we are setting up our Discord. So that would be out there. So right now it's just Twitter.

All right.

Yeah.

Awesome. We'll link that in the show notes. Anil, awesome having you on. I've really enjoyed our private discussions. We want to bring that more to the public space and hope people find this topic interesting and one that may not be as appreciated. But if you talk to builders, if you talk to— you're in these L2 communities, the sequencer is central and it's a very important component that is very much being worked on. So great to see you guys come on and share your vision. So we'll definitely be following closely as testnet and hopefully, you know, mainnet launches soon.

Yeah. Thank you for having me. It was a pleasure talking to you both.

Thanks, Anil. Hey everyone, big thanks for watching today's episode. Wanted to just quickly remind you about the Stellar Development Foundation's 6th Annual Meridian Conference this October 15th through 17th in London. This is an incredible opportunity to network with the forward thinkers who are defining the future of this space. Head over to meridian.stellar.org and use the code BLOCKWORKSPOD for early bird pricing.