Bankless - Devcon #8 - EigenLayer with Sreeram Kannan

Episode Date: October 25, 2022

Welcome to Devcon 6. The first Devcon for Bankless, the conference was a ton of fun and an amazing cultural experience in Bogota, Colombia. We’re speaking with Sreeram Kannan about EigenLayer, a platform to leverage Ethereum security through the innovative method of restaking. ------ 📣 Push | Try the Communication Protocol of Web3 https://bankless.cc/Push  ------ 🚀 SUBSCRIBE TO NEWSLETTER:          https://newsletter.banklesshq.com/  🎙️ SUBSCRIBE TO PODCAST:                 http://podcast.banklesshq.com/  ------ BANKLESS SPONSOR TOOLS:  👾 SEQUENCE | ALL-IN-ONE PLATFORM https://bankless.cc/Sequence  ❎ ACROSS | BRIDGE TO LAYER 2 https://bankless.cc/Across  🦁 BRAVE | THE BROWSER NATIVE WALLET https://bankless.cc/Brave  🔐 LEDGER | NANO HARDWARE WALLETS https://bankless.cc/Ledger  ⚡️FUEL | THE MODULAR EXECUTION LAYER https://bankless.cc/Fuelpod  ------ Resources: Sreeram on Twitter: https://twitter.com/sreeramkannan  EigenLayer: https://www.eigenlayer.com/  ----- Not financial or tax advice. This channel is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. This video is not tax advice. Talk to your accountant. Do your own research. Disclosure. From time-to-time I may add links in this newsletter to products I use. I may receive commission if you make a purchase through one of these links. Additionally, the Bankless writers hold crypto assets. See our investment disclosures here: https://www.bankless.com/disclosures 

Bankless Nation, welcome back to the DevCon 6 experience. We have a couple Alpha Leagues, some debuts of some episodes about particular projects in the Ethereum ecosystem, some that I think are getting particularly strong attention out of other community leaders in the space. This one is Eigenlayer. I'm talking with Serim from Eigenlayer, the co-founder of eigenlayer. And eigenlayer is a unique project that I think if it is successful, it's extremely ambitious, but if it is successful, really changes the game for what it means. to be an eth-staker and to provide security for many, many, many other things beside Ethereum

using eth-staking. Kind of crazy. YouTube watchers will actually notice that the production quality of this video goes up bigly. And that is because we happen to run into a scheduling conflict with the Infinite Garden documentary. So we just used their equipment. And so they were very generous to let us use their equipment, their video cameras, and their audio. So this production quality, probably be the best production quality video that we have on

bankless. So I hope you enjoy this exploration into eigenlayer and all of its potential consequences right after we talk to some of these fantastic sponsors that make the show possible. TrueFi is leading defy into the future of on-chain, uncollateralized loans, opening up defy to the $8 trillion global credit market. Whether you want higher yields on your lending or you're a fund manager who just wants access to global liquidity and the cost savings of defy, Trufi is here for you. But if you're going to use Trufai, use it through the brave, browser, the user-first browser for the Web3 age. The Brave browser keeps your digital footprint small, keeping you in the driver's seat while also being a powerful battle station for Web3,

letting you access your crypto through its native wallet, view your NFTs, and keep up to date with your Web3 communities. Another thing you can use through Brave is, of course, Arbitrum. And you all, of course, have heard about Arbitrum, but the Arbitrum ecosystem is really heating up. With a recent launch of Arbitrum Nova, Arbitrum has entered the world of multi-chain layer twos. and with a recent acquisition of Prismatic Labs, Arbitrum Firepower is bigger than ever. Arbitrum Nitro shipped last month and has made Arbitram faster and cheaper than ever before. So make sure that you can experience what Arbitrum has to offer before it's too late. But maybe you're a developer who hates the constraints of the EVM.

So check out the Fuel VM from the Fuel Network, which has opened up the world of parallel transaction execution, breaking fuel network free from EVM baggage. With fuel, you can leverage the... Rust Tooling ecosystem to build stronger apps all while keeping Ethereum level security. Also pushing the frontier of Web3 development is the Sequence wallet from Horizon. Sequence is the all-in-one developer platform you need to build Web3 games and applications. For users, Sequence is a smart wallet and perhaps the most intuitive wallet in all of Web3. But for developers, it's a plug-and-play platform for all Web3 apps and games. Check out Sequence, which is already

powering some of the best Web3 games out there. And lastly, DeSoe, is the decentralized social platform built from the ground up to disrupt the social media industry. Disrupting social media takes a lot of data, and DeSot's infinite state applications can finally store and index large amounts of content and data fully on-chain. With multiple crypto-native monetization mechanisms for both developers and creators, DeSot wants to usher in a new relationship with our social applications. Check them out at DeSo.com. Bankless Nation, I'm here with Sriram Kanan from Eganlayer, and we are at DevCon,

DevCon number six, which is also my first DevCon. So I'm particularly excited about it. Sirium, are you having fun? Yeah, this has been awesome. Thank you so much for having me on this bankless podcast. Yeah, your project, EganLayer. EganLayer? EganLayer.

I can come on my radar recently from a number of people that I consider very informed about this space. And so I wanted to get you into the show, and thank you for the Infinite Garden Doctrani Team for letting us use all of their equipment. And to explore EigenLayer what is up to and why so many people are excited about it. So let's start at the very beginning.

What is Eigenlayer? eigenlayer is a mechanism that we have designed, which allows the Ethereum Trust Network to be flexibly used to build general distributed systems. So the way we think about it is just a little bit of background about myself. I'm also an associate professor at the University of Washington, Seattle, where I run the UW Blockchain Research Lab. Go Huskies. Go huskies. Over the many years, we have actually worked on consensus protocols, new distributed systems, scalability. One thing we saw, which is amazing in the Ethereum ecosystem,

is the amount of permissionless innovation that's available at the smart contract layer. Anybody can come and deploy a new contract, and they don't need to be trusted. They're borrowing decentralized trust from Ethereum. In fact, when people talk about things like modular blockchain, I think of Ethereum as the first modular blockchain. It modularized trust so that DAPs can consume it, which led to the whole pseudonymous economy.

You don't need as a DAP creator to be trusted. The blockchain platform is trusted. But this freedom of permissionless innovation does not extend all the way down into the stack. If you want to build a bridge, you want to build an Oracle, you want to build a data availability layer, you need to go and find your own distributed network to validate it. And building a new distributed network to validate your own particular thing is a whole set of other things. And it is, in some sense, there is a large decentralized network with massive economic security and decentralization already existing, the Ethereum network. And why not use that so that anybody can actually build new distributed systems on top?

So that's what eigenlayers' core mechanism is. And the way we achieve it is by supplying the trust from the stakers, stakers stake to commit to producing Ethereum blocks, but they can also re-stake, reuse the same stake, to do other things. For example, I can make a credible commitment that I'm storing data. I can make a credible commitment that I'm providing Oracle inputs or a certain execution, functionality, and I'm putting my stake at risk, the yeat that is staked in the core platform at risk to do these other things. So it is an extension protocol, which extends the scope of what Ethereum can do so that there is permissionless innovation all the way down through the

stack. So the way that I understand this, and I'm going to try and regurgitate this to check my understanding, is that we have this pool of Ethereum stakers, and they are doing a very good job securing Ethereum, but that doesn't scale out beyond Ethereum. There are other things out there that we need to secure. There are cross-chain bridges. There are oracles that, you know, things like chain link, but other oracles, any sort of input, anything that you want to be secure, actually. And so EigenLayer is commoditizing, ETH staking to make it more general purpose. Is that a fair way to describe it? That is a great way to describe it. If you go back to the core value proposition of blockchain, it is decentralized trust, right?

And you have this massive pool of decentralized trust here that can be supplied through the Ethereum Protocol. And we made it more flexible with the Layer 2 era, where you can build Layer 2 roll-ups, which can consume this decentralized trust by writing proofs, either validity or fraud proofs. But we want to expand this even more. We want to supply the Ethereum trust from staking,

which is the root of trust, to applications far beyond the core conception. If you look at any other, you know, a blockchain protocol, they will have other advantages. For example, maybe the finality time is smaller. Maybe there is some other advantage or MEV management is better. And what you can do is now start internalizing all of this innovation and match it with the largest pool of decentralized trust that we have.

You can build a new chain on eigenlayer. You can build a new consensus protocol. You can build data availability. You can build bridges. You can build MEV management for Ethereum L1 all on eigenlare. And this isn't something totally new. new. We've seen other flavors like this in crypto's history. I mean, we were talking about this and you were explaining this to me. You used merge mining, which is something that I used to once know very well,

but it's escaped me now because it was just like a different era. It's proof of work. Can you explain like merge mining what the general principle is? Yes. And how eigenlayer is now like a kind of a new evolution and why it is new. Absolutely. So merge mining is this principle of can I take the massive approve of work underneath Bitcoin and then supply it to new networks that want to bootstrap. If you go back to kind of like already Satoshi made comments about we can actually do things like this. But the thing with merge mining is the economic incentives for merge mining are not well aligned. Why? If, for example, even if we get 100% of the Bitcoin miners to run another chain, if they turned up attacking that chain, they don't lose any time. They don't lose any

They definitely don't lose Bitcoin. They also don't lose their Bitcoin mining equipment. So all they have at stake is very little when they're actually merge mining to provide services to some other new service. So merge mining, a bunch of Bitcoin miners with a ton of hash power, they already have the security layer that's doing a great job securing Bitcoin. We can secure another blockchain by merge mining, tapping into this pool of security that already exists so that this new blockchain can also become secure.

But what you're saying is that this doesn't actually work too well because, because it doesn't put anything at stake for these miners, because what they are really risking is really on the Bitcoin side. If they are aligned to the Bitcoin network, they're just going to burn their electricity. That's a loss for them. But that doesn't really extend to a minority blockchain. And so it's an interesting idea,

but in execution doesn't really hold up, right? That is exactly what we're saying. And when you're merge mining a different chain, if you did anything bad on that chain, it definitely does not affect the Bitcoin to USD price, which is the major sensitivity for a Bitcoin. And so there is really nothing at stake in that merge mining. So easy come, easy go.

Sure. So you're doing something for free and you're not really getting anything from that. Right. And so you don't really have anything at risk, right? Absolutely. So how is Eigenlayer an evolution upon this? What's new doing? Absolutely.

So we just saw the merge and we're super happy. Ethereum did a great job, right? And with the merge, what we realized is proof of stake trust is much more flexible. And this goes to the root of why proof of stake is better is the ability to slash somebody when they're misbehaving. Sure. So imagine there's a $100 billion pool, which is underwriting Ethereum staking. Suppose you get only 5% of this to underwrite security on your protocol.

What they do is they're actually taking their ETH stake worth that $5 billion into your protocol. What that means is now if they misbehave in your protocol and there is a concrete slashing contract, for example, if you double sign a block in your additional protocol, they will just get slashed. their eth is getting slashed. This aligns perfectly the incentives. So from the stakers viewpoint, they are taking additional risk, the slashing risk,

in return for additional yield, which is by validating this other chain, they may either get the new native token of that chain or the fees paid in that chain. There is an economy around it. And it aligns the positive and the negative incentives, which is really what's taking dead, is to expand the scope of incentive design

from purely positive incentives, to positive and negative incentives, which is why the trust is more flexibly borrowed from Ethereum after proof of stake rather than before proof of stake. Okay. So once again, I'm going to try and repeat this back just to make sure I understand and hopefully the listeners are following along. Merge mining didn't work because there wasn't really anything at risk here. If you misbehaved on one chain, you didn't really lose anything on your Bitcoin chain. And so what this is different with proof of stake is this idea of slashing where what I think what eigenlayer is is an additional commitment upon normal Ethereum proof.

of stake, saying, hey, I'm going to behave with proof of stake Ethereum because that's how the rules are, but I'm also going to behave with this other network, and I'm going to use my ether collateral as the collateral, as the stake. And so you are opting into additional slashing potential if you are lying on a different chain, a different consensus mechanism, a different thing, an Oracle network, whatever. And so you as a staker are opting into more responsibility and promising to not lie to an additional chain. And it's the slashing of proof of stake that is really the fundamental different. between merge mining of the Bitcoin world, proof of work world, and the Ethereum proof of stake world. Is all right? Absolutely. The fact that slashing is programmable and Ethereum is a general purpose programmable chain in which right now we're seeing things like how do you do fraud proofs for general purpose execution engines. So the slashing layer now has become extremely powerful, both in terms of the trust that can be shared, but also in terms of the level of programmability in the slashing that we can build. So these two things converge into bring something like Igonlayer into reality.

And just the raw incentive is that maybe I just want to sign up for every single service possible. I'll do an Oracle. I'll do another proof-and-stake blockchain. And I'm just getting more and more rewards for everything I sign up for if I can manage that as like a node operator. Yeah. So the two things you're doing are you're committing additional computational responsibilities,

but also committing additional slashing risks. And you calculate the marginal value of these two. And if it is greater than the rewards that you're making, out of that system, yes, please opt in. So one thing we put a lot of thought into eigenlayer is how do we make it lightweight? How do we build services on top of eigenlayer, which do not tax the computational requirements of the each takers. We are very much in alignment with the Ethereum ethos of we need to have home stakers, solo stakers,

as well as, you know, other liquid staking pools so that a different category of people can participate in the staking ecosystem. And so we've made our services. to be run even by things like homestakers, at least some category of services, which can be yield-bearing. Even homesdakers can opt in and obtain these yields. Yeah, and I think one of the reasons why this project

has really captured my attention lately, especially as I've learned about it more, is it fits some of the vision that we have for the crypto world. And since we are, once again, using the Infinite Gardens equipment, I'll use the Infinite Garden metaphor, where if we have more trust, then we can grow more things. And if we have more security across all these different layers of crypto, first we can probably have fewer exploits, which is good. And we can create a garden where more and more trust comes about.

And what I mean by this is there are many new use cases that can now tap into Ethereum's, like, universe class level of trust and security because of proof of stake. And now we can extend these things to new things. So what are some of these new things? What are these things out in the world of crypto that need more trust that don't have it? Yeah. So if you look at one of the first of the things. of the immediate things, we are actually building ourselves. We want this to be a platform where anybody can come and build, but you have to bootstrap the platform.

So we're building the first service on top of eigenlayer. We call it eigenDA. Igen, DA is a data availability service run by each stakers, which can provide massive amounts of data availability throughput, so that the roll-up era can come to fruition much earlier. So our vision is that there is going to be a massive number of roll-ups, which all can use the security from Ethereum. On top of it, if we can supply massive data availability,

now these Rolubs write commitments, use Ethereum for censorship resistance, use Ethereum for settlement, but use Ethereum trust through eigenlayer to do massive data throughputs. We're already talking about things like 15 megabytes per second throughput in our data availability layer. One example of what we think you can build on eigener.

But there is a massive scope of other things that are coming up, and we expect a lot of others to participate in it. Another category of products that can be built on top of eigenlayer is MEV management. Suppose you're a block proposer and you're participating in, say, MEV Boost. So one thing you're unable to do today is you cannot make credible commitments outside of what is the scope of Ethereum. So the way MEV Boost works is it shoe-horns itself into the one-slashing condition that you have,

which is if you double sign a header, you can be penalized. So the builder sends you a header that you sign off and you have no agency to add anything new on top of it. Instead, if you were restaked, you could opt into eigenlayer and you could say, hey, I will include everything that the builder sends me. But there is always some extra space in the block because of 1559, because if all blocks are full all the time, then the price explodes. So there's always extra space. You can exert your proposer agency and include new transactions on top of the builder's block, thus contributing. to increasing the censorship resistance of Ethereum. One example of what you can do on eigen-Lair.

You can start doing many other MEP management things. For example, you know, you talk a lot about D-Fi. And one of the things that you can start doing is things like block proposals commit to doing collateral liquidations or refueling or taking atomic arbitrage and pass it back to the uniswap and sushi-swap pools. You can start building very interesting services on top of eigen-layer. Because the block proposals are staked,

they have authority over what can be returned into the block. They can even commit to things like threshold encryption. And when you send me like an encrypted transaction, I sign off that I'm going to include this, and I have to include the decryptured transaction for this. So there are a variety of things in the MEV space. But beyond that, you can think of bridges, you can run light-clan bridges off-chain,

you can think of running new consensus protocols. One of the things that we don't have an Ethereum is fast finality yet. And other protocols competing saying that, hey, I'm going to give you one second finality. Somebody can come and build a kind of a consensus protocol on eigenlayer on which roelps can settle at like one second latency because you have a massive economic security underwriting the finality of this chain.

So these are examples of what you can build on Igel. And I think the gist here is a lot of these applications, whether it's another data availability layer, defy apps, M.E.V, all of these things, if you add more trust to these things, they become more expressive. They can do more things. because we've commoditized trust in a new way and allowed that trust to work its way into parts of, you know, the Ethereum Protocol at the Layer 1, at the Defi Apps and Bridges and Oracles, if we have trust, we can have expressivity in our applications. And if I really want to take this into like some cerebral stuff, when there is a lot of trust, there's a lot of culture. And so with trust, we can actually become the best versions of ourselves. And that's also true for like MEV. That's also true for Defi Apps. They become with more trust, they can become the best versions of themselves. So that's what gives me really, really excited.

about this thing. I want to go back to the data availability layer really quick. With dank sharding, I think we get something like 1.3 megabytes a second of throughput at the Ethereum base layer, which is massively increased beyond like the kilobytes of data a second that we currently have. But 1.3 megabytes a second still is like slow. How much data throughput could we get with a eigen layer, data availability layer, and what can be unlocked with that? So firstly, we are building on the shoulders of giants. We are actually just taking Deng Sharding's core cryptographic architecture and just building and engineering around it.

And the degrees of freedom you have when you're innovating at an agile layer built on top of a protocol is very different from the degrees of freedom you have at the core protocol. There are various sets of constraints as well as various sets of guarantees which are stronger in the core protocol.

So what we can do is we can start taking up these core innovations and build engineering around it. For example, our estimates say that you can get gigabytes per second of like throughput on the same Ethereum data layer by basically using the ideas from dark sharing as well as building an engineering layer around it. The fact that this layer does not have to run a whole consensus protocol and is not intermingled with it and has degrees of freedom like nodes participating in it can have different fee dynamics, different exposure of IP addresses, all these different engineering freedoms, we can exploit all of it

in actually building very high-performance data availability layers, where each node does not need to have one gigabytes per second. That's not what we're implying. We do not want to go the Salon a route where each node needs to take a lot of bandwidth. The way we think about scaling is the net throughput of the system divided by the node requirements. And if we can maximize that, that's our...

And I think it's entirely possible, and this is already behind the vision of DunkRod and Justin and Vitalik in designing things like Dunkshadding, is that data availability is the one thing we know how to provision scalably. And so data availability is provisioned scalably, and then people can write execution proofs through this system. So that's the same vision we share. Sure.

And I think something that bankless listeners should really pay attention to and take note of is that we have, with think sharding, with the beacon chain, we're going to have 1.3 megabytes a second of data throughput. Sweet, that's like a huge improvement. With this eigenlayer idea, we can get to 1.3 gigabytes a second. Is that the number? That is a few year ago. We are already at 15 megabytes per second today.

Right. Okay. But we will get to, you know, 100 to 1,000 X of this. Right. Yeah, but also, dank charting is also like two years out as well. Yeah. So many things are two years out in this industry. But the point I want to drive home is that there's not an additional security mechanism

producing that 1.3 gigabytes a second of data. This is not like Celestia with a new blockchain and new security mechanism. This is the Ethereum stakers that are helping secure this 1.3 gigabyte a second of data. And that means that instance. is something like really close to the metal of what Ethereum actually is. These are the Ethereum Stakers securing this Ethereum data. And so it's a little bit like Rocket Pool, for example, where like it is inside of the Ethereum app layer helping secure Ethereum staking as a service system. And so it's an app that is about the protocol. And so there's fewer external

dependencies making it something like really, really Ethereum native. And when we have like fewer external dependencies, that's probably how you can get 1.3 gigabytes of data a second. And I'll again, remind listeners that 1.3 gigabytes of data a second is that much data of expressivity of us doing cool things in crypto that is all trustless in this trustless world that we want to want to build out. Serene, this has been an awesome exploration into eigenlayer. Where are you in the developmental roadmap? What's coming up next and how might people want to get involved? Yep. So right now we are at a phase where we are on internal dev nets, where we are actually having all of these protocols run. We will be on a public test net in like the next three months, three to four months.

months and like another three to four months after, hopefully we'll hit Mainland. So we have this timeline and the way you can participate with us is if you're a staker, you can come and talk to us on how you can actually engage in additional productive yield. If you are a builder, there is a whole bunch of cool stuff you can build on top of our eigenlayer. If you want to build a roll-up and want to consume massive data bandwidth, you want to build new VMs that can actually be very, very expressive or use EVM in a very parallelized manner, all of these things you can come and build on top of eigentya. If you're a community member, we will start engagement forums where we can kind of brainstorm on ideas for how to take this forward. Beautiful. Well, Trem, thank you so much for

joining me today. Thank you so much, David. This is really awesome. Cheers.