Podcast: The Round Trip
Compiled & Edited by: Yuliya, PANews
Having personally witnessed the GameStop short squeeze, and seeing his then-employer Robinhood "pull the plug" to forcibly end the retail frenzy, this "crash" moment in traditional finance directly spurred the creation of Sei—the world's fastest public chain aimed at replacing Nasdaq.
In the new Founder’s Talk series of *The Round Trip*, co-produced by PANews and Web3.com Ventures, Sei founder Jay Jog not only provides a hardcore breakdown of how parallel EVM achieves a 50x performance leap but also exclusively reveals the institutional funding secrets behind the hundred-billion TVL and the new track of AI Agent payments, offering a glimpse into the real landscape of a "decentralized Wall Street."
From Robinhood to Crypto: The Fracture of Traditional Finance and the Birth of Sei
PANews: Welcome Jay to *Round Trip*! How has Hong Kong been treating you so far? Perhaps we can start by chatting about your personal journey. How did you get to where you are today? And what projects have you been working on recently?
Jay Jog: Thanks for having me! This is my first time in Hong Kong since 2018, and the city is as amazing as I remember, if not more.
Regarding my background, I majored in Computer Science in college and joined the US brokerage platform Robinhood after graduation. It's most famous for the GameStop short squeeze incident. At that time, a large number of retail investors, driven by genuine belief or a certain "meme-like" price movement, bought stocks like GameStop, AMC, and a dozen others dubbed "meme stocks," causing prices to soar. Wall Street hedge funds, seeing this trend, tried to short these stocks. The mechanism of shorting involves borrowing shares to sell, then buying them back later to return. But as prices rose sharply, short sellers kept losing money and were forced to cover their positions by buying back shares, triggering a massive "short squeeze" that drove prices even higher. During that period in 2021, almost every retail investor made a lot of money.
But suddenly one day, the brokerage I worked for, Robinhood, directly shut down the buy function, meaning no one could continue buying those rising stocks, which basically ended that price movement. People across the US were shocked and extremely angry because it was a moment when the "little guy" finally triumphed over Wall Street, only to have Wall Street essentially turn off the buy side. As an internal employee, I felt terrible. Friends often treated me as the company itself, demanding, "Why can't I trade now?" "Why am I losing money?" And I couldn't do anything; I had no control over the situation.
This incident made me truly realize how "broken" the current financial system is. The root cause behind it is the T+2 settlement mechanism: Robinhood had to provide up to $3 billion in collateral to a third-party clearinghouse to allow users to continue trading, but the company didn't have that capital. This was the initial inspiration: the traditional financial system has structural problems. If you want to build a truly "internet-native" financial system, you need "internet-native" financial infrastructure. That's why I believe blockchain is the ideal carrier for all this, which is basically the initial motivation for us to create Sei.
Breaking the EVM Performance Bottleneck: The Rise of Parallel EVM and Ecosystem Explosion
PANews: With that motivation, how did you start building Sei?
Jay Jog: We started R&D in 2021, initially launching Sei V1 based on the Cosmos architecture, and went live with the first version of the mainnet in August 2023. Cosmos does have a large developer community, but we realized you also must support EVM smart contracts (typically written in Solidity and compiled into EVM bytecode). If you don't support EVM, it's very difficult to build a truly large, vibrant developer ecosystem.
So we started seriously considering supporting EVM while also researching its limitations. One of the most prominent issues is that the throughput supported by the Ethereum mainnet and various Rollups on it is very limited, around 50 transactions per second. For example, if you want to build an order book-based exchange like Nasdaq, you need to support about 20,000 transactions per second (TPS). This creates a complete disconnect: there's a huge gap between what can be done on-chain and the performance of the off-chain real world. We saw an opportunity to support that level of performance while still maintaining decentralization. That was the初衷 (initial intention) behind building parallel EVM.
PANews: After parallel EVM officially launched on the mainnet in July 2024, it sparked a huge wave of excitement. It's fair to say you pioneered this narrative, right?
Jay Jog: Yes, I believe we were indeed the first team to propose this narrative and the first to actually implement it on the mainnet. This also led to many application scenarios initially appearing only on our chain, followed by a significant increase in on-chain activity. To date, over 5 billion transactions have occurred on the mainnet, with about 100 million unique wallet addresses having conducted transactions, and over 1 million daily active users.
This level of activity further drove more TVL within the ecosystem; our TVL peak reached about $18 billion, a very substantial figure across the entire ecosystem.
The Underlying Logic of Institutional Entry: Traffic, Distribution, and Killer Apps
PANews: Such high numbers have indeed attracted institutional capital. When these institutional allocators choose between different blockchains, what do they value most? Is it traffic, brand, or the underlying technology?
Jay Jog: In the past year, we've had five large institutional funds (including BlackRock, Brevan Howard, Hamilton Lane, Apollo, and Laser) launch fund products on our chain. A few weeks ago, Ondo also launched USDY on our chain. We are now starting to see significant institutional adoption materializing.
Frankly, institutions don't care that much about the underlying technology itself. They care more about what kind of user base you have and whether the distribution channels are already well established. This is also why ecosystems like Ethereum and Solana naturally attract institutions more easily, as they already have the largest user scale and mature distribution networks.
From a broader perspective, the value of blockchain performance is mainly reflected in empowering developers. If you can support very high throughput, it opens up new design spaces for developers, allowing them to build entirely new types of applications. If you can create "killer apps" that simply cannot be realized in other ecosystems, it will naturally bring more users, which in turn makes institutions more interested in that ecosystem. In crypto, "excitement" usually comes from two things: either a new source of yield that allows users to make money, or an application scenario that is inherently very interesting and worth participating in. As long as one of these is satisfied, users will start transacting on-chain; this is the key to attracting and initiating the "flywheel effect."
Navigating Bull and Bear Markets: Maintaining Focus and Resilience in Adversity
PANews: I'd like to go back to your personal experience and delve deeper. Leaving a comfortable, high-paying, stable job at Robinhood to start a business in the crypto industry—during the journey of the past few years, were there any moments that made you think, "This isn't what I wanted"? Could you share those difficult moments that made you stronger?
Jay Jog: I feel like we've been sailing against the wind from the start. We started building the project in 2021, but our first fundraising happened almost right after the Terra collapse. At that time, Terra's roughly $50 billion market cap almost evaporated within a week, and just three weeks after that event, we, as a new team, went out to raise funds. You can imagine, most VCs were taking a wait-and-see attitude then.
That was probably our first real encounter with the reality of the crypto industry's fundraising environment. But when you start a project in a bear market, especially after a "doomsday" style crash, you become very frugal, pragmatic, and extremely careful with the resources at hand, keeping the team as lean as possible. This makes you very self-sufficient and clear on direction. If you start in a bull market, you face distractions from 10 different development directions; in a bear market, these distractions basically don't exist, allowing you to focus on one thing and execute it perfectly.
Overall, I'm very grateful to have embarked on this journey; the project has been much more successful than I initially imagined. Of course, there have been many low points. Whenever a bear market hits, morale across the industry drops, and confidence wanes. For example, at the time of recording this podcast, many influential people have left the industry in the past two weeks, and Bitcoin's price recently fell to a阶段性低点 (stage low) of around $59,000.
PANews: In recent weeks, we've seen many old players liquidating and exiting, even doubting whether crypto finance is scalable and whether everything we're building is meaningless. Faced with this sentiment, why are you still 100% optimistic about the future?
Jay Jog: The crypto industry is an extremely resilient industry. The current situation is very interesting: on one hand, the industry is making huge progress—for example, the US government is pushing legislation to provide legal status for stablecoins and establish a clear regulatory framework, institutions are launching stablecoins and fund products on-chain on a large scale; but on the other hand, Bitcoin's price is falling.
But one thing is very clear to me: as long as there is real growth and real adoption, the short-term fluctuations brought by bear markets will dissipate in the long run. As long as you maintain enough belief in your project, bear markets are often the best time to enter. From a builder's perspective, it's a great time to build truly substantial applications and achieve real product-market fit (PMF); from an investor's perspective, it's also a good time to make high-conviction investments and position in quality assets (like Bitcoin at a "discount"). So I remain very optimistic about the future of the entire crypto industry.
Building a "Decentralized Wall Street": Sei Giga, Permissionless Assets, and Internal Incubation
PANews: In your opinion, what else needs to be built to lay the foundation for the next bull market? What role does Sei play in this?
Jay Jog: Our vision is to build a "decentralized Wall Street." From our perspective, the bear market is an excellent opportunity to complete several key constructions. There are three core layers:
- The first core is the underlying L1 protocol itself. We are developing Sei Giga, which will help us achieve about a 50x performance improvement. This is a very夸张 (exaggerated/impressive) improvement compared to any other blockchain protocol on the market, making it possible to build systems like Nasdaq on-chain (which is very difficult to achieve with existing public chain architectures). There's a lot of very interesting technology behind this, like multi-concurrent block production mechanisms and related incentive design.
- The second layer is institutional adoption. We are already seeing assets like USDY and fund products going live. The next step is to make these assets as "permissionless" as possible. Currently, many assets are "permissioned"—after issuance, they cannot be freely traded on-chain. Once permissionless is achieved, these assets can truly be used in DeFi scenarios, such as in lending markets and other applications, which will be very exciting.
- The third point, which I most hope to see, is more killer applications landing in our ecosystem. Over the past few years, we've observed that there are basically two paths to building killer apps. The first is to contact as many entrepreneurs as possible and persuade them to join, but this often has limited effect because truly excellent founders usually choose top ecosystems that already have the largest user base (like Solana, Base, or Ethereum). This is a typical "chicken and egg" problem. The more effective way is the second: actively incubating projects and getting them to truly run within your own ecosystem. The traditional "standard playbook" (like hackathons, builder houses) has become less effective by 2026. We are already taking the internal incubation route, and I believe other ecosystems will gradually follow suit.
PANews: As a retail investor, I'd like to ask, among the projects incubated internally within your ecosystem, what are some unique features specific to Sei? After all, chains like Solana also emphasize high performance. Where's the difference?
Jay Jog: A key point is that once Sei Giga is live, all applications related to Central Limit Order Books (CLOB) will only be possible on our chain. If you want to build a system like Nasdaq, it requires about 20,000 TPS, which is basically impossible on existing public chains. And Nasdaq only accounts for about 10% of global securities trading volume. So, if you want securities trading to truly move on-chain, it's almost impossible on other chains currently, and Sei Giga will unlock this capability. I believe more financial applications, especially trading scenarios, will be the most promising direction.
Advice for Ordinary Investors: Establish Core Beliefs, Reject Blind Following
PANews: I believe you've experienced multiple bull and bear cycles. I personally believe we are in a super cycle, but market sentiment is really terrible. As an investor in this field, how would you advise ordinary investors to weather the bear market?
Jay Jog: This is an interesting question. I first entered the crypto market in 2017, experienced the frenzy of late 2017, and saw the market cool significantly in 2018, especially in 2019. Ironically, I took my Bitcoin to a poker gambling site at that time and lost it all, so I strongly advise everyone not to do such things. Later, I went through the 2020 and the bear market after the FTX collapse in 2023.
The most important advice I can give is: You must maintain a firm belief in your original purpose for participating in crypto. Many people don't have real conviction in crypto; they just come in because friends are doing it or the industry is hot.
You need to establish a core logic to understand why you yourself believe in crypto. Once you have this core understanding, whether you are a builder or an investor, all decisions during the bear market will be made more calmly. But if you just think crypto is "exciting," it often doesn't end well. So I strongly advise those going through the bear market to think deeply about why you truly believe in crypto.
Breaking the Myth of Technological Materialism: The Real Moat of an Ecosystem Lies in Users and Applications
PANews: By 2026, what misconceptions do you think the market has about Sei? How would you correct these views?
Jay Jog: The biggest misconception is that people think we are only focused on the technology itself, just疯狂地提升性能 (crazily improving performance).
We certainly value continuous technological optimization and improving chain performance. But we also understand that the value of technology has its limits. Beyond a certain stage, what really determines the success of an ecosystem is actual user growth, real traction, and killer applications, not TPS numbers or finality time. This is also why we focus on incubating projects, as that is the most efficient and direct way to enhance the overall strength of the ecosystem.
2026 Advice for Asian Developers: Deepen Financial Scenarios and Embrace AI Agent Payments
PANews: One last additional question: what advice would you give to Asian developers in 2026? What should they focus on when building applications?
Jay Jog: I think there are two broader trends currently:
- The first is that finance is gradually converging as the core application scenario for crypto. Of course, there are many other directions experimenting on-chain, like链游 (chain games/GameFi) and social, but we are starting to see that even in games, social, and various other crypto use cases, the part that is most effectively落地 (landed/implemented) is often the financial-related aspect最适合在链上执行 (most suitable for on-chain execution), while other modules can be built off-chain and then brought on-chain via state commitments or proof mechanisms. So I believe we should further deepen financial use cases and think about how to play a more important role within them.
- The second direction that will increasingly emerge is AI, especially with the rise of more AI Agents (like products such as Multibot), where users can run their own Agents and instruct them to perform tasks. Agents are a form of activity execution native to the internet, and for this form, using an internet-native currency is most reasonable. "Agent payments" will become a bigger trend. Protocols like x402 launched by Coinbase are being created, and we actually played an important role in helping them launch and become one of their first ecosystem promoters. So I believe "Agent payments" is definitely a new direction worth exploring, especially for entrepreneurs just starting out.
