Endgame for Sei: AMA with Jeff, Co-Founder of SEI
What's next for Sei and SEI ecosystem to be excited about?
Sei has been on a spotlight recently. But why? What makes SEI L1 special when there are so many alt-L1s & L2s already?
So, I decided to ask the SEI team directly in order to understand their take on the L1 space, their perspective on monolithic vs modular blockchain development, and the endgame vision for SEI.
Other questions discussed:
Did Ethereum win the L1 games?
What’s so special about SEI V2 and parallelized EVM?
What's next for Sei and SEI ecosystem to be excited about?
Btw, I recently stopped sponsorships for my blog, and decided to share protocols I believe are going to give big airdrops. Yaka Finance DEX on Sei offers free airdrop for testnet users for completing “Milestones” such as swapping and providing LP. I would appreciate if you signed up with my referral link.
What’s the point of L1s, and Sei in particular, in the era of multiple L1s and a growing number of Layer 2s? Do you agree with Sanket (Polygon) and CEO of Matter labs that Ethereum has already won the L1 games?
Jay and I started Sei Labs in 2021 because at the time we were trying to build a DEX and evaluating each of the infrastructure options: Ethereum, Avalanche, Arbitrum, Solana, heck, even Terra!
It became very clear that it was impossible to build a decentralized exchange that could genuinely compete with the UX of Coinbase and Binance no matter what Layer 1 or Layer 2 you built it on.
This user problem is why we built Sei. Sei fundamentally aims to solve the user experience problem in crypto. If Sei is successful, then applications will be able to offer nearly the same UX as any Web2 competitor, and that is incredibly powerful.
Being the first to parallelize the EVM is crucially one part of the journey that Sei is taking to solve the user experience problem. Sei takes several of the major advancements Solana proved out and brings it to the thriving Ethereum ecosystem.
The journey includes several other major technical advantages such as optimistic parallelization and the Twin-turbo consensus that the Sei Labs team created.
To answer the question directly, Sei is the first to explore parallelizing the EVM which will serve as a potential scaling approach for the broader Ethereum ecosystem. We’re hopeful that the Ethereum community can one day leverage the open-source codebase of Sei to achieve new possibilities for scaling.
It’s easy to view the industry as zero-sum. However, the crypto industry is still incredibly nascent in the grand scheme of things, and we take a positive-sum viewpoint. Where Ethereum, Sei, and the Layer 2s can help the industry scale in the future.
You’ve recently announced Sei V2, the first parallelized EVM blockchain. But what makes the V2 special and what’s actually “parallelized blockchain” as I believe few actually understand it.
And why would dApp choose to build on Sei instead of other L1s and Ethereum L2s?
The V2 update is the biggest update so far to the current Sei blockchain. There are four major advancements Sei is introducing:
Backwards compatible EVM - Sei uses Geth under the hood to get full EVM bytecode compatibility. TLDR is that builders on Sei benefit from the huge universe of Ethereum tooling such as MetaMask, Foundry, Remix and Hardhat
Optimistic parallelization - currently, most parallelized chains like Sei v1 make parallelization *optional*, it requires developers to do extra work to use parallelization, which most do not. Sei v2 will automatically parallelize smart contracts so developers don’t have to do the work.
SeiDB. Sei will split up state commitment and state storage, to decrease block processing time, decrease state bloat, make it faster to run a new full node.
Sei V2 will allow EVM contracts to interact with existing Cosmwasm smart contracts, and vice versa. Allowing for the best builders on Cosmos to join forces with the best builders on Ethereum
Why would an application choose Sei?
Sei is the fastest Layer 1 in history: average of 390ms time to finality makes Sei multiples faster than even the highest performance Layer 1s and Layer 2s
Try the chain yourself! It is getting closer and closer to the near-instant UX we are accustomed to with the Web2 apps we use
Sei v2 takes the proven advantages of Sei and makes it accessible to a global base of EVM developers, using familiar tools such as Metamask.
Sei v2’s backward compatibility of EVM smart contracts allows developers to deploy audited smart contracts from EVM-compatible blockchains with no code changes
The Sei community is one of the most unique in the industry. There’s an incredibly tight knit group of community members forming on Sei who love trying out new apps and learning about new possibilities on Sei. I’d recommend checking them out yourself, get a feel of the vibes and the memes.
Modular versus monolithic designs represent two distinct approaches to scaling blockchains. Ethereum has taken the lead in modular scaling, with numerous Layer 2 solutions and an increasing number of Data Availability layer chains being launched. On the other hand, Solana and Fantom have opted for a monolithic design.
You focus on monolithic scaling. Could you please explain why you made this choice? Additionally, I would be interested to hear your perspective on these two approaches.
Sei focuses on monolithic design.
The upper bound for performance in monolithic systems is inherently higher. In essence, while modular designs offer flexibility and are excellent for experimentation and iteration, they often come with heavy added complexity at scale.
Looking into the future, especially five years down the line, I firmly believe that the most performant infrastructure will continue to be monolithic. This doesn't diminish the value of modular approaches, especially in the realm of innovation and specific use-cases. But for raw performance and streamlined operations, monolithic designs will lead the way.
This isn’t a brand new dynamic. Monolithic has continued to be the gold standard for performance and user experience in Web2 for decades. Modular is certainly helpful in the early stages when applications are just trying things out, but eventually in order to serve billions of users and offer the best user experience, monolithic makes the most sense.
What's next for Sei and the SEI ecosystem to be excited about? Could you share the dApps that are about to launch.
Looking ahead, we're gearing up for the Testnet for Sei v2, scheduled to launch in the next month or so. The transition to Mainnet, planned for the first half of 2024, will be subject to thorough governance and auditing processes to ensure maximum security and efficiency.
I encourage everyone to explore the Sei ecosystem, which is buzzing with activity right now. Special shoutouts to the current activity centers like Astroport, Silo Stake, Pallet, Webump, Colony, and Seiyans, which are leading the charge.
What’s the endgame plan & vision for Sei?
Sei V2 is already code-complete and live on devnet, so exceptional teams can come and try it out!
Performance testing has shown impressive results, with over 5,000 transactions per second (TPS) at 300 milliseconds time-to-finality, and a theoretical TPS of around 12,500. When you consider batched orders, this means we could handle approximately 28,000 operations per second.
Sei aims to combine the best of Ethereum with the best aspects of Solana.
The incredible tooling and universe of developers around Ethereum with the performance and speed advancements of Solana. This will offer the best environment for applications to thrive and serve the next billion users.
Thank Jeff, and everyone else reading the interview. Let me know who I should invite to the next interview in the comments :) And don’t forget to subscribe.