theCUBE + NYSE Wired: Crypto Trailblazers | Jay Jog, Sei Labs

Clips
More from theCUBE + NYSE Wired: Crypto Trailblazers

Jay Jog

Co-Founder

Sei Labs

play_circle_outline Revolutionizing Blockchain with Sei Labs: Overcoming Ethereum EVM's Limitations through Parallelization

play_circle_outline Sei V1 Goes Live on Main Net in 2023 with EVM Support After Successful Seed and Series A Funding Rounds

play_circle_outline Sei Giga roadmap focused on achieving high transaction throughput.

play_circle_outline Technical challenges in improving throughput include data availability, consensus, and execution.

Info

Jay Jog, Sei Labs

Jay Jog

Co-Founder Sei Labs

search

>> Hello, welcome back to theCube here in New York City. This is our Wall Street location with the NYC Wired community and theCube together. We've got Wall Street and Silicon Valley connecting two hubs, tech and finance together. I'm John Furrier, host of theCube. Jay Jog is here, co-founder of Sei Labs. Building into the decentralized finance world and infrastructure enabling apps that are going to come online. We're already starting to see in the mainstream liquidity bubble coming to the table here. And I won't say bubble in a bad way. I mean more finally the maturization is coming. Jay, thanks for coming into our NYC studio for the podcast here. Thanks for coming on.

Jay Jog

>> Yeah, thanks for having me on, John.

>> So I said bubble, but I meant it not in a negative way because the bubble of action is going to just burst onto the scene because we're seeing the exchanges coming, the US, the climate is phenomenal. And so you're starting to see, people, I won't say coming out of hiding, but they're finally emerging, certainly out into the United States. Ethereum was in California the past couple of weeks. Tech scenes popping. You guys are building really scalable technology at Layer 1. Explain what you guys do because I want to get into this. It's accelerating fast on the finance side right now. What do you guys do? Take a minute to explain what you guys do.

Jay Jog

>> Yeah, so we're building Sei, which is a layer one blockchain. And we're building the first Parallelized EVM. So what does this mean? When we got started, we realized one thing very quickly, which is that the Ethereum virtual machine, which is what basically every blockchain uses to process transactions, is here to stay. And what I mean by that is the default for any developer that is starting to write any kind of smart contracts is to build smart contracts on the EVM. And if you look at the actual market share for different virtual machines, around 70 to 80% of market share is with the EVM. Even though there have been several other types of virtual machines that people have tried to create. So examples here would be Solana's SVM. There's also been MoveVM created by Aptos and Sui. But none of these have really been able to get market share. So developers sticking with the EVM, and it's really tough to get them off. Because from a developer standpoint. It's not just a tech stack, it's also all the tooling that's created around that, all the mind share, all the developer activity. So developers, it's really tough to get them to move. So the realization that we have-

>> They're staying on Ethereum.

Jay Jog

>> They're staying with the EVM. Exactly.

>> Got it. All right.

Jay Jog

>> So developers want the EVM. So then we realized, okay, we should try to make the EVM better. A corollary here in web two would be JavaScript. JavaScript was created in the '90s. Definitely wasn't the best tech.

>> Of course. But we standardized.

Jay Jog

>> Exactly. There were a lot of competitors that came up along the way. None of them were really able to get market share. And what ended up happening is that people started creating things like TypeScript, React, jQuery, just things to make JavaScript better and more accessible. And we think the exact same thing is going to be happening with EVM. So we did a deep dive and tried to understand what is really missing from the EVM right now. And the biggest limitation right now with the EVM is a lack of throughput with the EVM right now.

>> A lack of what?

Jay Jog

>> Lack of throughput.

>> Got it. Okay.

Jay Jog

>> So you're not able to get in that many transactions per second. If you look at Ethereum L1, you're only able to get around 20 to 50 transactions per second. So if you think about that, if there's like 10,000 people that are trying to interact with some DeFi smart contract, only 20 of these transactions can be included each second. And what ends up happening is it becomes a bidding war, where you can pay higher gas fees in order to get your transaction included. And the end result is if people might pay a hundred dollars to get a transaction included. That just completely prices out most of the world from doing anything on Ethereum.

>> So the price goes up, to gas it up, so to speak.

Jay Jog

>> Exactly, yeah, gas becomes super expensive,

>> Pun intended.

Jay Jog

>> Yeah.

>> So talk about the importance of this, because you worked at Robinhood before that you had a lot of different tech positions at other companies like Facebook. They had a centralized model, I think, right? So they weren't DeFi. So DeFi is not centralized. So how does that impact transactions and what specifically are you guys doing to make that viable from a stability standpoint and get the throughput?

Jay Jog

>> Exactly. So we realized that we needed to scale the EVM. That is the biggest bottleneck that would help improve developer accessibility of the EVM. Because it makes it easier for developers to write smart contracts and also for normal users to come and use the chain. So we decided to paralyze the EVM. So what parallelization means is you're able to process multiple transactions at the same time. This is not a novel concept at all in the computer science world. It's been around for decades now.

>> AI thrives on it. You're talking about Nvidia, I was at GDC last week and everything has been parallelized. So it's like this is standard.

Jay Jog

>> Exactly.

>> Standard computer science.

Jay Jog

>> Standard computer science, and it's bizarre in a way.

>> What are you guys doing? How do you do that at L1? Because L1 is a very, it's a low level. It's the lowest level.

Jay Jog

>> It is the hardest problem in the entire blockchain ecosystem in terms of building the Layer 1. So what we ended up realizing is we needed to paralyze different transactions that are coming in. I'll avoid going into super deep in the weeds over there. We used this technique called optimistic parallelization, which allows multiple transactions to come in. We're able to process them at the same time if there's any conflicts and we rerun them. But basically through this approach, we're able to get substantially higher throughput. And a simple mental model here would be you're taking the EVM and you're making it really fast. So it's giving you the best of both Ethereum and Solana, where you get the developer mindshare tooling that you get with the Ethereum ecosystem and the extremely fast speed and accessibility you get with Solana. Now, this is completely different than the model that Robinhood and a lot of decentralized players take, where everything is just happening on one server. It's one company that controls everything. And I think fundamentally that is not the model that you want, trust based activities. Ideally finance to be happening, because ideally you don't want to be trusting anyone. And I worked at Robinhood and I saw the entire GameStop saga go down. And when that GameStop saga was happening, it was really bizarre to me in many ways. Because there was just so many things that are opaque about it. No one really knew what was going on. And I think decentralized financial realists would've helped quite a bit over there. Both in terms of-

>> Give an example, what would it have done?

Jay Jog

>> Yeah, so one of the biggest issues here was this idea of T plus two settlement, where Robin Hood needed to put up collateral because things took T plus two days to settle. This is something that normal people didn't understand at all. And if it had been clear through smart contracts that are interacting with each other on the blockchain that, first of all, you would've been able to see that there's these requirements. Secondly, with blockchain, you would have instant finality. So you don't even need to wait two days. You would have to wait in Sei's case 400 milliseconds.

>> There's no middleman.

Jay Jog

>> Exactly. Yeah. So you would've been able to avoid issues like that. The other part of it was just in terms of decision-making, even as an employee, I had no clue what Vlad and Baiju were doing behind the scenes. So I found out about things at the same time as the news basically announced them. And it just puts you in such a tough spot when you're employed like that, because put your reputation on the line and all of your friends are coming to you with questions. And you're the one that have to field them. So I think that using some kind of .

>> Basically the arbitrage of there was they took advantage of the fact that there was lag in the system and they just were operating in that in that gray area. Are we're simplifying it, basically?

Jay Jog

>> In my perspective, after going through everything, the biggest issue was they needed to put up collateral, but no one really knew that that entire process needed to happen. So there are stocks like GameStop that were just super volatile on price. So counterparties that Robin Hood was working with, they were basically saying, "You need to put up more money just because we don't want to take on more risks from your position size changing."

>> Yeah. Got it. They took advantage of us. It's a cultural point in time that will be written about. It's going to be-

Jay Jog

>> There's already so many movies and documentaries.

>> On one side you're like, "Oh shit, that sucks." And the other side you're like, "Damn, that's awesome." Again, this highlights the trends that's going on now that networks operate at scale and they can be very fast. So why have all this mechanism in between things.

Jay Jog

>> Exactly.

>> And that's the whole point of the theorem.

Jay Jog

>> Exactly.

>> Smart contracts direct to each other.

Jay Jog

>> Exactly.

>> We settle that.

Jay Jog

>> From my perspective right now, there's a ton of fixed happening in crypto. Two things that I think are invariably going to get substantially bigger in crypto are payments. So stable coins, being able to send money between different people. In the United States, people kind of take that for granted because we have relatively good financial rails. If you go to the second world countries, it's significantly more difficult to pay people. And also getting access to other currencies like USD is really, really difficult. So stable coins I think are going to take off and there's going to be a lot more adoption of crypto for payments. And then the other thing is exchange of assets. So trading is one of these things where you want people to be able to trustlessly do things. You don't want to trust the other party, the kind of party that you're dealing with. Because then there's a lot of things that can go wrong.

>> I was talking to someone here in New York and it's in the art world. And they're more on the elite side around collectors and selling to rich people who buy stuff to store their value, like they would buy Bitcoin or Ethereum. And the status is that actually people buy paintings still. And most of the museums don't have a lot of paintings. They're older, so they're worth more, but there's a digital culture coming on with digital art. Music is one. Gaming. In-game currency. So you're seeing the phenomenon of digital assets is absolutely here.

Jay Jog

>> And everything is being tokenized now.

>> And the pressure is coming from the capitalism side of the market, which is entrepreneurship. It's get out of the way, people. Because things like GameStop will happen in other forms because they're breaking down old protocols, other mechanisms that were built for the old rules of engagement where you needed a lawyer to paper the thing, or intellectual property rights. All these constructs were built over the notion that we did business a certain way.

Jay Jog

>> Exactly. I think that's going to be completely changing. I think people will consider tokenizing different things. I think art, for example, all these things that are just stored in vaults for investors to hold onto, there's no reason that will not be tokenized and not be traded aggressively on chain. Because I think it's just substantially a better experience for people.

>> Well, you're going to see a lot more art tech from us. We do mostly tech, now tech finance and money, art and societal type stuff, sports, stable coins, NFTs. Culturally that's a culture shift.

Jay Jog

>> Yeah, absolutely.

>> That's driving a lot of the action here. And how does that impact your guys' decisions on Hill? You guys saying we got to go faster, both in speed of transactions and just faster momentum. How are you guys operating things right now from a co-founder perspective? Is it full pedal to the metal, taking it slowly, making some key decisions. What are you optimizing for right now?

Jay Jog

>> So generally speaking, we have an extremely pragmatic Silicon Valley mindset at Sei Labs. And I think interestingly, this is very different than a lot of crypto companies. A lot of crypto companies are very idealistic. They're very in a way slow moving, because they don't really want to do things like quickly iterate fast, get user feedback, like the Silicon Valley mantra. We've taken a completely different approach. And I think that's been substantially better in terms of being able to validate whether people actually want what we're building. I think the most-

>> You're more product focused.

Jay Jog

>> Exactly, and I think that's the only ;.

>> Not Dow consensus focused.

Jay Jog

>> I think building a Dow is very good after you've achieved product market fit, because it helps you maintain something. But in terms of getting something started, you really need a team that is working really hard and really pragmatic in terms of decision making, in terms of trade-offs you're making. So I think you need to be fast. I think you need to be cheap. If it's too expensive for people to be trading, just no one can really access it .

>> Jay, let me ask you a question because this comes up a lot. By the way. I love that attitude. I think that's the right mindset. I think it's more entrepreneurial, less mission-based. I think you have mission-based and you have more entrepreneurial vigor, Silicon Valley for that. But New York's getting the vibe here. I'm seeing a lot of New York here. But you're starting to see a lot of people working on these hard problems. So as an entrepreneur, you guys have to think about scale. You mentioned you were at Robinhood and Facebook, the scale of what's happening now as global, so it's not build a SaaS app and put in the Apple Store. There's a system thinking that you need to have these kinds of product discussions. So it's product focused in the sense of the service solution, but it's infrastructure.

Jay Jog

>> Absolutely. Yeah.

>> So it's not banging away a prototype. "Hey, how's it going?" You're doing some serious hard stuff. What is the thinking? How do you guys look at the system and the scale aspect of it? Because other people are tackling large scale problems right now.

Jay Jog

>> Yeah, Sei is a multi-billion dollar blockchain. There's hundreds of millions of dollars of TBL on the chain as well. So we move as fast as we can, but .

>> How much on the chain right now?

Jay Jog

>> There's around 300 to 350 mil of TBL. So this is assets exogenous to say that are on the blockchain. So there's a ton of money that's there and we cannot afford to make any mistakes over here. Because it's just too much money that'll be lost if we make any mistakes. So we have to be extremely rigorous. And I think that there's a lot of different testing frameworks that we have, which I won't go into for now, but I think the other part of it is just the mentality that we have and the kind of folks we're hiring. We're basically getting the best and brightest from Silicon Valley to join our team and I think that has been one of the things that allowed us to both move quickly and also be able to preempt the scaling problem .

>> What kind of people are you hiring? What kind of level of people are they? Are they hardcore coders down to the network level? Are they protocols? Is it AIs? What's the makeup of the persona of the people on your team?

Jay Jog

>> The ideal kind of person that we found is someone that's extremely low level, so someone that's focused on infrastructure and there were places like Databricks, Robinhoods, those types of companies that have dealt with a ton of scaling problems before. And they've been the ideal kind of Web 2 persona to join. Right now we're in the middle of a bear market in crypto, so it's been easier for us to get really talented crypto native engineers to join the team as well.

>> Yeah, I liked how you mentioned Databricks because that's Cal. Stanford's got a ton of action. The Stanford Bitcoin Club is pretty solid, they got the blockchain club there. Cal's got great talent coming out of Cal. So Silicon Valley's got just those schools and a large ecosystem in the valley right now.

Jay Jog

>> Absolutely.

>> How about New York? What's the ecosystem like here?

Jay Jog

>> So right now New York is quickly becoming the crypto hub for the entire world. Undeniably, this is where everyone is coming to right now and I think the Trump administration has really fueled that to happen even faster.

>> You think New York's greater than Silicon Valley in terms of critical mass of talent?

Jay Jog

>> For crypto, absolutely right now.

>> Technical talent or business or both?

Jay Jog

>> 100% business talent is much better over here. I think there's starting to be more and more technical talent here as well. And I would make the bold claim that right now New York is actually better for crypto than SF is even from a technical standpoint.

>> That's awesome. What's some of the things in New York that are cool to do with the tribe and crypto? Are there meetups and certain things you guys attend? How do you guys do face-to-face activities? Obviously there's going to be an event coming up in May. What's going on here in New York?

Jay Jog

>> Everyone has offices over here. There's events happening basically every week, multiple times. There's a ton of conferences. Even last week there was the Digital Assets Summit that was hosted by BlockWork. So a ton of events happening over here. Super easy to meet people in person. And I think right now is the greatest time to be building in crypto. Because there's just so much energy and the regulatory clarity is also becoming a lot better.

>> It's interesting you mentioned New York. I've noticed, I was psyched that you said that, because I think a couple of weeks ago on the podcast we were talking about New York and a lot of the young guns, I call them the young under 30, basically talented, energy. They're doing hardcore enterprise financial services, blockchain because there's no consumer app anymore. Because if you're going to do anything consumer, you still got to go to AWS and that's enterprise by default. You've got all kinds of things you've got to provision or configure, which is great, because it's large scale. But here in New York, the one thing that people don't talk about is that you're one subway stop away from 10 customers. And so the customer base here is phenomenal. The meetups are great because it's mixed between industry, entrepreneurs, money, customers.

Jay Jog

>> Exactly.

>> You just walk down the street, you got insurance, you got finance. Every industry's here in New York. That's all going to be tokenized.

Jay Jog

>> Yeah. So I think in the next two to three years, I think the amount of crypto activity happening here is just going to continue scaling and New York is absolutely the place to be.

>> All right, let's get into some of the stats. You guys went on the chain. When did you guys go public? What are some of the stats?

Jay Jog

>> Sei V1 went live in August of 2023, and that's when we had the realization that we need to be supporting EVM.

>> On main net?

Jay Jog

>> On main net. That's correct.

>> Okay. All right.

Jay Jog

>> So taking a step back, we got started with this in 2021. We raised a $5 million seed round led by MultiCoin Capital in 2023. So this was around August we announced it. Multicoin is one of the top investors in the crypto space. They invested in Solana very, very early. So having them around the table was super helpful. Then we're able to raise a 30 mill series A in April of 2023 is when we announced it. I was off by year before. But yeah, we announced that and I think afterwards we're able to get a ton of traction going into the year.

>> So is that classic VC with preferred stock and equity?

Jay Jog

>> Exactly.

>> So it's classic startup formation.

Jay Jog

>> Yeah, yeah.

>> Okay, cool.

Jay Jog

>> So we raised a lot of money, had a ton of good investors around the table. We went live August of 2023 with Mainnet. Realized we need to be supporting the EVM. Then last year, so July of 2024 is when we went live with the parallelized EVM. And basically since then every single metric has been up into the right. It's been absolutely crazy to see. I think the one that matters the most is TVL. If people are putting money on the blockchain, that is the clearest metric that you want to really be focusing on. So at the start of 2024, we were at around five mill of TVL. Now we've crossed over 300, 350 mill. So that's a massive increase and it's just continuing to go up since here.

>> Yeah, talk about the smart contract, and as tokens become the key mechanism for instrumenting the value, whether it's network transactions or marketplaces. And then why that's important for developers and why Ethereum's winning? Because smart contracts, it's the application. You mentioned layer two and the roll-ups, that's on top of layer one. So you're down on the plumbing level. You're the transit if you will. You're making that transaction speed highly optimized. Which is needed. So check. What's that enabling up top?me What's happening above you?

Jay Jog

>> Yeah, so there's a lot of really interesting use cases you can think of that are supported, especially with higher throughput. So we're talking about payments before. Payments can only really make sense in ecosystems where it's really cheap to be making these transactions. So in Sei's case with Sei V2, which was the one that went live last year, we're able to get around 5,000 transactions per second. We're currently working on Sei Giga, which will allow us to get up to 250,000 transactions per second. That is a complete game changer because there's a lot of applications in Web 2 right now that just cannot be built in crypto using crypto rails. Examples over here would be Google. Google needs 100,000 TPS. No blockchain right now enables anything like Google to get built. Even things like NASDAQ, things like Visa, they need around 20,000 transactions per second, which no existing blockchain enables. So that's what we're really focused on right now. We want to enable those types of Web 2 applications to get built using crypto rails. And I anticipate it'll be used for things like finance, it'll be used for things like AI and several other use cases as well.

>> So Ethereum, Solana, what's the difference? A lot of people choosing those networks? You're starting to see use cases form. Are they competitive or are they just coopetition or was there a use case that fits better for each one?

Jay Jog

>> So yeah, historically Ethereum's been a chain where people did high value transactions and Solana's where a lot of the more degen activity that's lower value, they're more fun-

>> Like NFTs?

Jay Jog

>> NFTs, meme coins, all these things where people don't want to be putting in too much money, but they still want to be trading aggressively

>> It's like the public park.

Jay Jog

>> Exactly.

>> It's like, Hey, I don't want to pay gas. And all that stuff.

Jay Jog

>> Just I think fundamentally the Solana model is a better model because it allows people to actually go there and do things in a very cheap way. If people are priced out, then that ecosystem is just not going to be getting adoption.

>> And when you say priced out, what does that mean? Someone who's an individual small company?

Jay Jog

>> So if a normal person wants to trade on Ethereum, there's a lot of times where it'll cost them $50, $100 to make that trade happen. And that's just, no one's really going to do that. Especially if you leave countries like the United States, it's too expensive even for economically for a lot of companies .

>> So it's high value transactions. So it's a big transaction, I'm going to buy a car.

Jay Jog

>> Exactly.

>> Or do something like that level.

Jay Jog

>> Yeah. So what I think the direction we're going to move in is developers want to continue using the EVM. That's why you want to continue supporting the EVM. It makes it easy to take contracts that already exists, deploy them as well. But then you want to make it extremely fast and cheap and that's probably the biggest fundamental design I disagree with Solana on. Solana does not support the EVM as a first-class citizen and I think that's made it much harder for them to get to develop that option.

>> What's the goal from a timing standpoint to hit the new metrics for your throughput?

Jay Jog

>> We're currently heads down. That's the biggest thing that we're focusing on from the technical side, just the Sei Giga roadmap. I think the next big milestone is going to be the white paper. We've already made announcements around the performance we've been getting, so now we're going to explain the research around how we were able to get that. In terms of a test net, that'll be coming hopefully sometime later this year. But it's really, really difficult technical work.

>> Just scope the magnitude of the technical challenge for the folks watching. And that's also a good recruiting opportunity too because a lot of people want to work on hard problems. And so just scope the magnitude of the L1 challenge technically.

Jay Jog

>> Absolutely. So the biggest thing that we're focused on right now is improving throughput. So most other modern blockchains, they're able to get around this 5,000 transactions per second range. We're basically getting a 50X on that. So how do you get a 50X improvement against all these modern blockchains? Well, it's basically rebuilding everything from scratch. So one part of it is the data availability side. How do you make this data available to the entire network as quickly as possible? So how do you propagate it efficiently? That's a very low-level networking kind of problem. Second part of this is consensus, which is more of a game theory, crypto-economics type of problem, which is interesting in its own way. Then there's the execution side, which is more like, okay, what kind of database do you use? How do you quickly access date? How are you able to quickly parallelize these transactions? So there's a ton of really nitty-gritty engineering problems. And I think that's one of the reasons why we've had an easier time attracting talent than other types of companies, because a lot of strong people just want to work on interesting problems of theirs.

>> Yeah. You mentioned Databricks earlier. We'll wrap up the segment. But it made me think of the Hadoop days. So when we started our company 15 years ago, we covered the Hadoop, which was founded by Amr Awadallah and Michael Olson, the Cloudera, were our first office was, and Palo Alto. So we got to know that. That was the big data revolution. But Hadoop was so hard to use and manage. Apache Spark came out, which made Databricks. Databricks became successful because of Spark, and that just led them down a path. So slow, hard, easier with Spark. So EVM getting better. Is there a comparison there? It's a weird way to look at it.

Jay Jog

>> Yeah, I absolutely think there's a comparison there. I think the way that technology happens is it's generally not a complete revolution. It's not just something new completely comes and changes the game. I think it generally an evolution where you take something and it's a proof of concept, it's not built in a very good way. And then afterwards you keep iterating on it. And there eventually comes a point where you get something that is good enough that everyone is able to use, and then from there that becomes de facto. And I think that's what Sei is achieving to become.

>> Jay, thanks for coming on theCUBE. Give a quick plug for the company. What are you looking for? What's your goal? Obviously get the new test net up. What are you looking to do?

Jay Jog

>> Yeah, right now the biggest focus areas are Sei Giga from the technical side, which is building that chain that we're talking about. The second part of it is getting a killer application to come on. So there's a ton of founders that are building onto it right now. If you want to follow along with these applications, you could just follow Sei Network on Twitter, so S-E-I-N-E-T-W-O-R-K on Twitter and there'll be a lot more updates there.

>> Awesome. Thanks for coming on the podcast here. New York Stock Exchange, our new studio.

Jay Jog

>> Thanks for having me on then.

>> They're making trades behind us. They're yelling on the floor. I'm John Furrier. You're watching the crypto trailblazers, the pioneers, and also the folks plowing the trail again to a new generation of software and money coming together, bringing that startup focus to the world of crypto, because capitalism is coming in the front door fast. Jay, thanks for coming on theCUBE. And thanks for watching.

Jay Jog, Sei Labs

search

>> Hello, welcome back to theCube here in New York City. This is our Wall Street location with the NYC Wired community and theCube together. We've got Wall Street and Silicon Valley connecting two hubs, tech and finance together. I'm John Furrier, host of theCube. Jay Jog is here, co-founder of Sei Labs. Building into the decentralized finance world and infrastructure enabling apps that are going to come online. We're already starting to see in the mainstream liquidity bubble coming to the table here. And I won't say bubble in a bad way. I mean more finally the maturization is coming. Jay, thanks for coming into our NYC studio for the podcast here. Thanks for coming on.

Jay Jog

>> Yeah, thanks for having me on, John.

>> So I said bubble, but I meant it not in a negative way because the bubble of action is going to just burst onto the scene because we're seeing the exchanges coming, the US, the climate is phenomenal. And so you're starting to see, people, I won't say coming out of hiding, but they're finally emerging, certainly out into the United States. Ethereum was in California the past couple of weeks. Tech scenes popping. You guys are building really scalable technology at Layer 1. Explain what you guys do because I want to get into this. It's accelerating fast on the finance side right now. What do you guys do? Take a minute to explain what you guys do.

Jay Jog

>> Yeah, so we're building Sei, which is a layer one blockchain. And we're building the first Parallelized EVM. So what does this mean? When we got started, we realized one thing very quickly, which is that the Ethereum virtual machine, which is what basically every blockchain uses to process transactions, is here to stay. And what I mean by that is the default for any developer that is starting to write any kind of smart contracts is to build smart contracts on the EVM. And if you look at the actual market share for different virtual machines, around 70 to 80% of market share is with the EVM. Even though there have been several other types of virtual machines that people have tried to create. So examples here would be Solana's SVM. There's also been MoveVM created by Aptos and Sui. But none of these have really been able to get market share. So developers sticking with the EVM, and it's really tough to get them off. Because from a developer standpoint. It's not just a tech stack, it's also all the tooling that's created around that, all the mind share, all the developer activity. So developers, it's really tough to get them to move. So the realization that we have-

>> They're staying on Ethereum.

Jay Jog

>> They're staying with the EVM. Exactly.

>> Got it. All right.

Jay Jog

>> So developers want the EVM. So then we realized, okay, we should try to make the EVM better. A corollary here in web two would be JavaScript. JavaScript was created in the '90s. Definitely wasn't the best tech.

>> Of course. But we standardized.

Jay Jog

>> Exactly. There were a lot of competitors that came up along the way. None of them were really able to get market share. And what ended up happening is that people started creating things like TypeScript, React, jQuery, just things to make JavaScript better and more accessible. And we think the exact same thing is going to be happening with EVM. So we did a deep dive and tried to understand what is really missing from the EVM right now. And the biggest limitation right now with the EVM is a lack of throughput with the EVM right now.

>> A lack of what?

Jay Jog

>> Lack of throughput.

>> Got it. Okay.

Jay Jog

>> So you're not able to get in that many transactions per second. If you look at Ethereum L1, you're only able to get around 20 to 50 transactions per second. So if you think about that, if there's like 10,000 people that are trying to interact with some DeFi smart contract, only 20 of these transactions can be included each second. And what ends up happening is it becomes a bidding war, where you can pay higher gas fees in order to get your transaction included. And the end result is if people might pay a hundred dollars to get a transaction included. That just completely prices out most of the world from doing anything on Ethereum.

>> So the price goes up, to gas it up, so to speak.

Jay Jog

>> Exactly, yeah, gas becomes super expensive,

>> Pun intended.

Jay Jog

>> Yeah.

>> So talk about the importance of this, because you worked at Robinhood before that you had a lot of different tech positions at other companies like Facebook. They had a centralized model, I think, right? So they weren't DeFi. So DeFi is not centralized. So how does that impact transactions and what specifically are you guys doing to make that viable from a stability standpoint and get the throughput?

Jay Jog

>> Exactly. So we realized that we needed to scale the EVM. That is the biggest bottleneck that would help improve developer accessibility of the EVM. Because it makes it easier for developers to write smart contracts and also for normal users to come and use the chain. So we decided to paralyze the EVM. So what parallelization means is you're able to process multiple transactions at the same time. This is not a novel concept at all in the computer science world. It's been around for decades now.

>> AI thrives on it. You're talking about Nvidia, I was at GDC last week and everything has been parallelized. So it's like this is standard.

Jay Jog

>> Exactly.

>> Standard computer science.

Jay Jog

>> Standard computer science, and it's bizarre in a way.

>> What are you guys doing? How do you do that at L1? Because L1 is a very, it's a low level. It's the lowest level.

Jay Jog

>> It is the hardest problem in the entire blockchain ecosystem in terms of building the Layer 1. So what we ended up realizing is we needed to paralyze different transactions that are coming in. I'll avoid going into super deep in the weeds over there. We used this technique called optimistic parallelization, which allows multiple transactions to come in. We're able to process them at the same time if there's any conflicts and we rerun them. But basically through this approach, we're able to get substantially higher throughput. And a simple mental model here would be you're taking the EVM and you're making it really fast. So it's giving you the best of both Ethereum and Solana, where you get the developer mindshare tooling that you get with the Ethereum ecosystem and the extremely fast speed and accessibility you get with Solana. Now, this is completely different than the model that Robinhood and a lot of decentralized players take, where everything is just happening on one server. It's one company that controls everything. And I think fundamentally that is not the model that you want, trust based activities. Ideally finance to be happening, because ideally you don't want to be trusting anyone. And I worked at Robinhood and I saw the entire GameStop saga go down. And when that GameStop saga was happening, it was really bizarre to me in many ways. Because there was just so many things that are opaque about it. No one really knew what was going on. And I think decentralized financial realists would've helped quite a bit over there. Both in terms of-

>> Give an example, what would it have done?

Jay Jog

>> Yeah, so one of the biggest issues here was this idea of T plus two settlement, where Robin Hood needed to put up collateral because things took T plus two days to settle. This is something that normal people didn't understand at all. And if it had been clear through smart contracts that are interacting with each other on the blockchain that, first of all, you would've been able to see that there's these requirements. Secondly, with blockchain, you would have instant finality. So you don't even need to wait two days. You would have to wait in Sei's case 400 milliseconds.

>> There's no middleman.

Jay Jog

>> Exactly. Yeah. So you would've been able to avoid issues like that. The other part of it was just in terms of decision-making, even as an employee, I had no clue what Vlad and Baiju were doing behind the scenes. So I found out about things at the same time as the news basically announced them. And it just puts you in such a tough spot when you're employed like that, because put your reputation on the line and all of your friends are coming to you with questions. And you're the one that have to field them. So I think that using some kind of .

>> Basically the arbitrage of there was they took advantage of the fact that there was lag in the system and they just were operating in that in that gray area. Are we're simplifying it, basically?

Jay Jog

>> In my perspective, after going through everything, the biggest issue was they needed to put up collateral, but no one really knew that that entire process needed to happen. So there are stocks like GameStop that were just super volatile on price. So counterparties that Robin Hood was working with, they were basically saying, "You need to put up more money just because we don't want to take on more risks from your position size changing."

>> Yeah. Got it. They took advantage of us. It's a cultural point in time that will be written about. It's going to be-

Jay Jog

>> There's already so many movies and documentaries.

>> On one side you're like, "Oh shit, that sucks." And the other side you're like, "Damn, that's awesome." Again, this highlights the trends that's going on now that networks operate at scale and they can be very fast. So why have all this mechanism in between things.

Jay Jog

>> Exactly.

>> And that's the whole point of the theorem.

Jay Jog

>> Exactly.

>> Smart contracts direct to each other.

Jay Jog

>> Exactly.

>> We settle that.

Jay Jog

>> From my perspective right now, there's a ton of fixed happening in crypto. Two things that I think are invariably going to get substantially bigger in crypto are payments. So stable coins, being able to send money between different people. In the United States, people kind of take that for granted because we have relatively good financial rails. If you go to the second world countries, it's significantly more difficult to pay people. And also getting access to other currencies like USD is really, really difficult. So stable coins I think are going to take off and there's going to be a lot more adoption of crypto for payments. And then the other thing is exchange of assets. So trading is one of these things where you want people to be able to trustlessly do things. You don't want to trust the other party, the kind of party that you're dealing with. Because then there's a lot of things that can go wrong.

>> I was talking to someone here in New York and it's in the art world. And they're more on the elite side around collectors and selling to rich people who buy stuff to store their value, like they would buy Bitcoin or Ethereum. And the status is that actually people buy paintings still. And most of the museums don't have a lot of paintings. They're older, so they're worth more, but there's a digital culture coming on with digital art. Music is one. Gaming. In-game currency. So you're seeing the phenomenon of digital assets is absolutely here.

Jay Jog

>> And everything is being tokenized now.

>> And the pressure is coming from the capitalism side of the market, which is entrepreneurship. It's get out of the way, people. Because things like GameStop will happen in other forms because they're breaking down old protocols, other mechanisms that were built for the old rules of engagement where you needed a lawyer to paper the thing, or intellectual property rights. All these constructs were built over the notion that we did business a certain way.

Jay Jog

>> Exactly. I think that's going to be completely changing. I think people will consider tokenizing different things. I think art, for example, all these things that are just stored in vaults for investors to hold onto, there's no reason that will not be tokenized and not be traded aggressively on chain. Because I think it's just substantially a better experience for people.

>> Well, you're going to see a lot more art tech from us. We do mostly tech, now tech finance and money, art and societal type stuff, sports, stable coins, NFTs. Culturally that's a culture shift.

Jay Jog

>> Yeah, absolutely.

>> That's driving a lot of the action here. And how does that impact your guys' decisions on Hill? You guys saying we got to go faster, both in speed of transactions and just faster momentum. How are you guys operating things right now from a co-founder perspective? Is it full pedal to the metal, taking it slowly, making some key decisions. What are you optimizing for right now?

Jay Jog

>> So generally speaking, we have an extremely pragmatic Silicon Valley mindset at Sei Labs. And I think interestingly, this is very different than a lot of crypto companies. A lot of crypto companies are very idealistic. They're very in a way slow moving, because they don't really want to do things like quickly iterate fast, get user feedback, like the Silicon Valley mantra. We've taken a completely different approach. And I think that's been substantially better in terms of being able to validate whether people actually want what we're building. I think the most-

>> You're more product focused.

Jay Jog

>> Exactly, and I think that's the only ;.

>> Not Dow consensus focused.

Jay Jog

>> I think building a Dow is very good after you've achieved product market fit, because it helps you maintain something. But in terms of getting something started, you really need a team that is working really hard and really pragmatic in terms of decision making, in terms of trade-offs you're making. So I think you need to be fast. I think you need to be cheap. If it's too expensive for people to be trading, just no one can really access it .

>> Jay, let me ask you a question because this comes up a lot. By the way. I love that attitude. I think that's the right mindset. I think it's more entrepreneurial, less mission-based. I think you have mission-based and you have more entrepreneurial vigor, Silicon Valley for that. But New York's getting the vibe here. I'm seeing a lot of New York here. But you're starting to see a lot of people working on these hard problems. So as an entrepreneur, you guys have to think about scale. You mentioned you were at Robinhood and Facebook, the scale of what's happening now as global, so it's not build a SaaS app and put in the Apple Store. There's a system thinking that you need to have these kinds of product discussions. So it's product focused in the sense of the service solution, but it's infrastructure.

Jay Jog

>> Absolutely. Yeah.

>> So it's not banging away a prototype. "Hey, how's it going?" You're doing some serious hard stuff. What is the thinking? How do you guys look at the system and the scale aspect of it? Because other people are tackling large scale problems right now.

Jay Jog

>> Yeah, Sei is a multi-billion dollar blockchain. There's hundreds of millions of dollars of TBL on the chain as well. So we move as fast as we can, but .

>> How much on the chain right now?

Jay Jog

>> There's around 300 to 350 mil of TBL. So this is assets exogenous to say that are on the blockchain. So there's a ton of money that's there and we cannot afford to make any mistakes over here. Because it's just too much money that'll be lost if we make any mistakes. So we have to be extremely rigorous. And I think that there's a lot of different testing frameworks that we have, which I won't go into for now, but I think the other part of it is just the mentality that we have and the kind of folks we're hiring. We're basically getting the best and brightest from Silicon Valley to join our team and I think that has been one of the things that allowed us to both move quickly and also be able to preempt the scaling problem .

>> What kind of people are you hiring? What kind of level of people are they? Are they hardcore coders down to the network level? Are they protocols? Is it AIs? What's the makeup of the persona of the people on your team?

Jay Jog

>> The ideal kind of person that we found is someone that's extremely low level, so someone that's focused on infrastructure and there were places like Databricks, Robinhoods, those types of companies that have dealt with a ton of scaling problems before. And they've been the ideal kind of Web 2 persona to join. Right now we're in the middle of a bear market in crypto, so it's been easier for us to get really talented crypto native engineers to join the team as well.

>> Yeah, I liked how you mentioned Databricks because that's Cal. Stanford's got a ton of action. The Stanford Bitcoin Club is pretty solid, they got the blockchain club there. Cal's got great talent coming out of Cal. So Silicon Valley's got just those schools and a large ecosystem in the valley right now.

Jay Jog

>> Absolutely.

>> How about New York? What's the ecosystem like here?

Jay Jog

>> So right now New York is quickly becoming the crypto hub for the entire world. Undeniably, this is where everyone is coming to right now and I think the Trump administration has really fueled that to happen even faster.

>> You think New York's greater than Silicon Valley in terms of critical mass of talent?

Jay Jog

>> For crypto, absolutely right now.

>> Technical talent or business or both?

Jay Jog

>> 100% business talent is much better over here. I think there's starting to be more and more technical talent here as well. And I would make the bold claim that right now New York is actually better for crypto than SF is even from a technical standpoint.

>> That's awesome. What's some of the things in New York that are cool to do with the tribe and crypto? Are there meetups and certain things you guys attend? How do you guys do face-to-face activities? Obviously there's going to be an event coming up in May. What's going on here in New York?

Jay Jog

>> Everyone has offices over here. There's events happening basically every week, multiple times. There's a ton of conferences. Even last week there was the Digital Assets Summit that was hosted by BlockWork. So a ton of events happening over here. Super easy to meet people in person. And I think right now is the greatest time to be building in crypto. Because there's just so much energy and the regulatory clarity is also becoming a lot better.

>> It's interesting you mentioned New York. I've noticed, I was psyched that you said that, because I think a couple of weeks ago on the podcast we were talking about New York and a lot of the young guns, I call them the young under 30, basically talented, energy. They're doing hardcore enterprise financial services, blockchain because there's no consumer app anymore. Because if you're going to do anything consumer, you still got to go to AWS and that's enterprise by default. You've got all kinds of things you've got to provision or configure, which is great, because it's large scale. But here in New York, the one thing that people don't talk about is that you're one subway stop away from 10 customers. And so the customer base here is phenomenal. The meetups are great because it's mixed between industry, entrepreneurs, money, customers.

Jay Jog

>> Exactly.

>> You just walk down the street, you got insurance, you got finance. Every industry's here in New York. That's all going to be tokenized.

Jay Jog

>> Yeah. So I think in the next two to three years, I think the amount of crypto activity happening here is just going to continue scaling and New York is absolutely the place to be.

>> All right, let's get into some of the stats. You guys went on the chain. When did you guys go public? What are some of the stats?

Jay Jog

>> Sei V1 went live in August of 2023, and that's when we had the realization that we need to be supporting EVM.

>> On main net?

Jay Jog

>> On main net. That's correct.

>> Okay. All right.

Jay Jog

>> So taking a step back, we got started with this in 2021. We raised a $5 million seed round led by MultiCoin Capital in 2023. So this was around August we announced it. Multicoin is one of the top investors in the crypto space. They invested in Solana very, very early. So having them around the table was super helpful. Then we're able to raise a 30 mill series A in April of 2023 is when we announced it. I was off by year before. But yeah, we announced that and I think afterwards we're able to get a ton of traction going into the year.

>> So is that classic VC with preferred stock and equity?

Jay Jog

>> Exactly.

>> So it's classic startup formation.

Jay Jog

>> Yeah, yeah.

>> Okay, cool.

Jay Jog

>> So we raised a lot of money, had a ton of good investors around the table. We went live August of 2023 with Mainnet. Realized we need to be supporting the EVM. Then last year, so July of 2024 is when we went live with the parallelized EVM. And basically since then every single metric has been up into the right. It's been absolutely crazy to see. I think the one that matters the most is TVL. If people are putting money on the blockchain, that is the clearest metric that you want to really be focusing on. So at the start of 2024, we were at around five mill of TVL. Now we've crossed over 300, 350 mill. So that's a massive increase and it's just continuing to go up since here.

>> Yeah, talk about the smart contract, and as tokens become the key mechanism for instrumenting the value, whether it's network transactions or marketplaces. And then why that's important for developers and why Ethereum's winning? Because smart contracts, it's the application. You mentioned layer two and the roll-ups, that's on top of layer one. So you're down on the plumbing level. You're the transit if you will. You're making that transaction speed highly optimized. Which is needed. So check. What's that enabling up top?me What's happening above you?

Jay Jog

>> Yeah, so there's a lot of really interesting use cases you can think of that are supported, especially with higher throughput. So we're talking about payments before. Payments can only really make sense in ecosystems where it's really cheap to be making these transactions. So in Sei's case with Sei V2, which was the one that went live last year, we're able to get around 5,000 transactions per second. We're currently working on Sei Giga, which will allow us to get up to 250,000 transactions per second. That is a complete game changer because there's a lot of applications in Web 2 right now that just cannot be built in crypto using crypto rails. Examples over here would be Google. Google needs 100,000 TPS. No blockchain right now enables anything like Google to get built. Even things like NASDAQ, things like Visa, they need around 20,000 transactions per second, which no existing blockchain enables. So that's what we're really focused on right now. We want to enable those types of Web 2 applications to get built using crypto rails. And I anticipate it'll be used for things like finance, it'll be used for things like AI and several other use cases as well.

>> So Ethereum, Solana, what's the difference? A lot of people choosing those networks? You're starting to see use cases form. Are they competitive or are they just coopetition or was there a use case that fits better for each one?

Jay Jog

>> So yeah, historically Ethereum's been a chain where people did high value transactions and Solana's where a lot of the more degen activity that's lower value, they're more fun-

>> Like NFTs?

Jay Jog

>> NFTs, meme coins, all these things where people don't want to be putting in too much money, but they still want to be trading aggressively

>> It's like the public park.

Jay Jog

>> Exactly.

>> It's like, Hey, I don't want to pay gas. And all that stuff.

Jay Jog

>> Just I think fundamentally the Solana model is a better model because it allows people to actually go there and do things in a very cheap way. If people are priced out, then that ecosystem is just not going to be getting adoption.

>> And when you say priced out, what does that mean? Someone who's an individual small company?

Jay Jog

>> So if a normal person wants to trade on Ethereum, there's a lot of times where it'll cost them $50, $100 to make that trade happen. And that's just, no one's really going to do that. Especially if you leave countries like the United States, it's too expensive even for economically for a lot of companies .

>> So it's high value transactions. So it's a big transaction, I'm going to buy a car.

Jay Jog

>> Exactly.

>> Or do something like that level.

Jay Jog

>> Yeah. So what I think the direction we're going to move in is developers want to continue using the EVM. That's why you want to continue supporting the EVM. It makes it easy to take contracts that already exists, deploy them as well. But then you want to make it extremely fast and cheap and that's probably the biggest fundamental design I disagree with Solana on. Solana does not support the EVM as a first-class citizen and I think that's made it much harder for them to get to develop that option.

>> What's the goal from a timing standpoint to hit the new metrics for your throughput?

Jay Jog

>> We're currently heads down. That's the biggest thing that we're focusing on from the technical side, just the Sei Giga roadmap. I think the next big milestone is going to be the white paper. We've already made announcements around the performance we've been getting, so now we're going to explain the research around how we were able to get that. In terms of a test net, that'll be coming hopefully sometime later this year. But it's really, really difficult technical work.

>> Just scope the magnitude of the technical challenge for the folks watching. And that's also a good recruiting opportunity too because a lot of people want to work on hard problems. And so just scope the magnitude of the L1 challenge technically.

Jay Jog

>> Absolutely. So the biggest thing that we're focused on right now is improving throughput. So most other modern blockchains, they're able to get around this 5,000 transactions per second range. We're basically getting a 50X on that. So how do you get a 50X improvement against all these modern blockchains? Well, it's basically rebuilding everything from scratch. So one part of it is the data availability side. How do you make this data available to the entire network as quickly as possible? So how do you propagate it efficiently? That's a very low-level networking kind of problem. Second part of this is consensus, which is more of a game theory, crypto-economics type of problem, which is interesting in its own way. Then there's the execution side, which is more like, okay, what kind of database do you use? How do you quickly access date? How are you able to quickly parallelize these transactions? So there's a ton of really nitty-gritty engineering problems. And I think that's one of the reasons why we've had an easier time attracting talent than other types of companies, because a lot of strong people just want to work on interesting problems of theirs.

>> Yeah. You mentioned Databricks earlier. We'll wrap up the segment. But it made me think of the Hadoop days. So when we started our company 15 years ago, we covered the Hadoop, which was founded by Amr Awadallah and Michael Olson, the Cloudera, were our first office was, and Palo Alto. So we got to know that. That was the big data revolution. But Hadoop was so hard to use and manage. Apache Spark came out, which made Databricks. Databricks became successful because of Spark, and that just led them down a path. So slow, hard, easier with Spark. So EVM getting better. Is there a comparison there? It's a weird way to look at it.

Jay Jog

>> Yeah, I absolutely think there's a comparison there. I think the way that technology happens is it's generally not a complete revolution. It's not just something new completely comes and changes the game. I think it generally an evolution where you take something and it's a proof of concept, it's not built in a very good way. And then afterwards you keep iterating on it. And there eventually comes a point where you get something that is good enough that everyone is able to use, and then from there that becomes de facto. And I think that's what Sei is achieving to become.

>> Jay, thanks for coming on theCUBE. Give a quick plug for the company. What are you looking for? What's your goal? Obviously get the new test net up. What are you looking to do?

Jay Jog

>> Yeah, right now the biggest focus areas are Sei Giga from the technical side, which is building that chain that we're talking about. The second part of it is getting a killer application to come on. So there's a ton of founders that are building onto it right now. If you want to follow along with these applications, you could just follow Sei Network on Twitter, so S-E-I-N-E-T-W-O-R-K on Twitter and there'll be a lot more updates there.

>> Awesome. Thanks for coming on the podcast here. New York Stock Exchange, our new studio.

Jay Jog

>> Thanks for having me on then.

>> They're making trades behind us. They're yelling on the floor. I'm John Furrier. You're watching the crypto trailblazers, the pioneers, and also the folks plowing the trail again to a new generation of software and money coming together, bringing that startup focus to the world of crypto, because capitalism is coming in the front door fast. Jay, thanks for coming on theCUBE. And thanks for watching.