Why Crypto Is Worth the Risk 💥 - Interview with Eigenlayer

WEBVTT

00:00:00.530 --> 00:00:07.040
Welcome to the Chain Stories podcast, the podcast that celebrates disruptors who defy convention.

00:00:07.490 --> 00:00:14.650
Here we dive into the bold stories of trailblazers who turned audacious ideas into billion dollar ventures.

00:00:18.483 --> 00:00:19.073
Hey everyone.

00:00:19.073 --> 00:00:21.513
Welcome to another ChainStories Podcast.

00:00:21.612 --> 00:00:27.323
Today we have a very welcome guest that's also part of the Blockchain Education Network alumni.

00:00:27.632 --> 00:00:29.533
Ishaan, it's a pleasure having you.

00:00:29.612 --> 00:00:30.253
How are you doing?

00:00:30.902 --> 00:00:32.222
Hey, doing great, Tony.

00:00:32.222 --> 00:00:32.753
How are you?

00:00:32.802 --> 00:00:33.552
Thanks for having me.

00:00:34.393 --> 00:00:35.883
I'm doing awesome.

00:00:36.412 --> 00:00:43.313
And we started this podcast actually before we started recording with your story, how you come across us.

00:00:43.412 --> 00:00:49.823
And that really surprised me because most guests are have not been around crypto for a while.

00:00:50.332 --> 00:00:55.073
And the Blockchain Education Network has been around crypto for ages over decades.

00:00:55.618 --> 00:00:58.438
And let's start with a brief intro about yourself.

00:00:58.518 --> 00:01:03.338
And how did you came across BEN while you were in college?

00:01:04.108 --> 00:01:05.078
Yeah, absolutely.

00:01:05.108 --> 00:01:06.257
So I'm Ishaan.

00:01:06.278 --> 00:01:13.847
I'm a protocol researcher at EigenLayer, where I focus on novel AVSs and AVS consumers.

00:01:14.218 --> 00:01:18.578
But I started in crypto during COVID around 2021.

00:01:19.227 --> 00:01:25.188
And during that time, I was building this startup, and I had no idea what I was doing.

00:01:25.188 --> 00:01:32.007
So I was looking for help wherever I could get and ran into Erick from, the Blockchain Education Network.

00:01:32.308 --> 00:01:37.697
And he was just such a massive help as an advisor and making connections.

00:01:37.820 --> 00:01:39.271
It was great to work with him.

00:01:39.551 --> 00:01:46.281
Even though the startup didn't end up panning out, it was, I met a lot of people who I still consider friends, through him.

00:01:46.691 --> 00:01:51.740
And, yeah, glad to be here where I can help some other students in my own position.

00:01:52.090 --> 00:01:53.891
And Ishaan, I think it's pretty interesting.

00:01:53.900 --> 00:01:55.200
You went to Ann Arbor.

00:01:55.281 --> 00:02:03.081
That was the university that BEN was originally created, where Jeremy Garner was attending at the time when he co founded the organization.

00:02:03.400 --> 00:02:07.990
Can you tell us a little bit about your experience while a student at Ann Arbor?

00:02:08.341 --> 00:02:11.651
Was there a Bitcoin club or cryptocurrency club?

00:02:11.950 --> 00:02:13.901
Were you working on your startup while in school?

00:02:13.911 --> 00:02:17.181
Did you thought about dropping out or did you complete your studies?

00:02:17.681 --> 00:02:18.061
Yeah.

00:02:18.061 --> 00:02:18.461
Yeah.

00:02:18.491 --> 00:02:21.121
So I was, I did go to university in Michigan.

00:02:21.121 --> 00:02:26.721
I actually didn't know that BEN was founded there, but absolutely loved my time there.

00:02:27.080 --> 00:02:33.920
It was very fun because, they're very focused on both the theoretical as well as the application side of computer science.

00:02:34.431 --> 00:02:39.390
And there's also a pretty early, but burgeoning blockchain community, right?

00:02:39.441 --> 00:02:56.961
When I started leading the blockchain at Michigan club, like we ended up doing consulting services with many different major companies, both inside and outside of crypto, like Celo and TRM Labs on the native side, but also companies like PayPal and Microsoft.

00:02:57.371 --> 00:03:04.700
And we also had a major governance operation and it was cool to see that grow doing governance for Uniswap compound Ave.

00:03:05.161 --> 00:03:14.161
So definitely during some times you're like, okay, I'm working on all of this like crypto stuff and people are making a lot of money.

00:03:14.401 --> 00:03:16.431
I could be building stuff full time.

00:03:16.730 --> 00:03:32.156
And so there's definitely periods where I considered dropping out, but honestly, I think there's a lot of value to like going to school and learning the full extent of what a computer science degree has to offer.

00:03:32.515 --> 00:03:34.975
And I frankly just had a lot of fun.

00:03:35.556 --> 00:03:40.366
What was your idea when you're doing computer science at Michigan?

00:03:40.895 --> 00:03:42.816
Initially, did you knew about crypto at all?

00:03:42.816 --> 00:03:49.186
Or you're like, okay, I'm going to go graduate and work in a Fortune 100, Fortune 500 company.

00:03:49.756 --> 00:03:51.355
How did crypto come into your life?

00:03:52.036 --> 00:03:57.716
Before coming in, I thought, oh, the best thing you can do after college is work at Google, right?

00:03:57.985 --> 00:04:01.225
And so that was the plan for me, just going in.

00:04:01.496 --> 00:04:07.975
And then I stumbled upon cryptocurrency during COVID like a lot of people did, right?

00:04:07.975 --> 00:04:13.786
Because I went to university, I used to only eat at these local restaurants.

00:04:13.825 --> 00:04:17.935
And they loved these restaurants, but they were all going out of business.

00:04:18.375 --> 00:04:40.865
And so the question is how can I help these businesses and found out that okay I can create a startup to connect these small businesses with retail investors to crowdsource equity funding and effectively these retail investors can get a piece of the pie of their local area and these small businesses get some cash to bootstrap.

00:04:41.365 --> 00:04:48.495
Found out that okay, credit card payments are too prohibitive, like you can't be spending three and a half percent plus ten cents.

00:04:48.836 --> 00:04:58.000
So what if we could use stable coins and then I found out about gas fees and so Ethereum was back then way too expensive.

00:04:58.000 --> 00:05:09.980
So shifted to Avalanche and once we figured out, okay, if we use stable coins on like low cost fast blockchains then this actually works out very well.

00:05:10.141 --> 00:05:11.230
They started thinking.

00:05:11.401 --> 00:05:23.295
Hey, this new blockchains are so cool because they enable a whole sort of new applications that are only enabled by very low cost of moving money.

00:05:23.596 --> 00:05:26.725
And that was the start of how I discovered crypto.

00:05:26.725 --> 00:05:30.175
And now I've found out that it can be so much more.

00:05:30.646 --> 00:05:31.005
Yeah.

00:05:31.026 --> 00:05:49.870
And you touched upon a couple of interesting points and now with other experience, I wanted to ask you, do you think it's possible today to have a blockchain or a system that scales to the everyday consumer, to the everyday person without gas fees or without these massive fees that people usually incur?

00:05:50.370 --> 00:06:04.091
I think that horizontal scaling is extremely powerful, because for every new machine that you, for every new machine that you add, you're enabling new block space.

00:06:04.321 --> 00:06:11.776
And by machine in this case, it likely refers to roll ups, and you can have centralized or decentralized rollups, but let's not get into that.

00:06:12.216 --> 00:06:29.245
So assuming you have a scalable layer one, which frankly, Ethereum doesn't scale too well right now on its own, but you can have a scalable layer one and then build layer twos on top of that, that are posting ZK proofs.

00:06:29.596 --> 00:06:36.865
Then I think that's truly all you need and you need to make sure that as long as all of your bottlenecks are covered.

00:06:36.865 --> 00:06:43.576
So you need to have hyperscalable data availability, which you can do via horizontal scaling as well.

00:06:43.646 --> 00:06:49.745
Paradigms like EigenDA have really leaned into this rather than opting a consensus based approach.

00:06:50.065 --> 00:06:54.725
You can bring your execution down and make it extremely succinct.

00:06:54.980 --> 00:07:01.211
Via zero knowledge proofs or just I know Solana has ZK compression, which is also super exciting.

00:07:01.651 --> 00:07:07.341
And you have extremely fast, effectively native latency for execution.

00:07:07.711 --> 00:07:09.771
Like MegaEth is trying to do.

00:07:09.771 --> 00:07:13.891
And same thing with Monad and Solana and I think Sui as well.

00:07:14.701 --> 00:07:32.165
So if there's a lot of founders listening to our podcast, If you're starting your own project in crypto, would you build in a specific blockchain or do you think the future goes from Ethereum and then those L2s or you'd go on and build on BASE, AVAX, or any specific chain out there?

00:07:32.665 --> 00:07:33.016
Yeah.

00:07:33.516 --> 00:07:33.815
Yeah.

00:07:33.815 --> 00:07:34.036
Yeah.

00:07:34.036 --> 00:07:37.605
I definitely think you can build your own blockchain.

00:07:37.925 --> 00:07:45.935
And there's definitely new design patterns that are interesting to explore, but really, if you're a founder now, you should be thinking.

00:07:46.466 --> 00:07:55.555
What are the most pressing problems that I can be solving, today and also two, three, five years in the future from now, right?

00:07:56.096 --> 00:08:04.146
And I frankly think that the push towards scalability is a problem that's like being worked on.

00:08:04.146 --> 00:08:08.466
And at least if I were a founder, I would not be working on this.

00:08:08.485 --> 00:08:13.346
Instead, I'd be focused on some building, some new applications.

00:08:13.586 --> 00:08:18.165
Some things that I'm really excited about are under collateralized lending.

00:08:18.185 --> 00:08:25.766
I think that's a huge new paradigm unlock that, is necessary to actually bring finance on chain because.

00:08:26.136 --> 00:08:32.895
Every loan in the real life is pretty much under collateralized, including the biggest one of all, which is mortgages.

00:08:33.365 --> 00:08:42.745
I think about how I could incorporate ZKTLS to create new novel, experiences, for consumers.

00:08:43.135 --> 00:08:50.395
I'm not sure if you saw, there's this one prediction market called tomorrow.news.

00:08:50.826 --> 00:08:57.296
And what they're doing is they're saying, okay, what if you could predict the New York times headline for tomorrow?

00:08:57.666 --> 00:09:02.846
And they use ZKTLS to verify what the New York times headline is.

00:09:03.025 --> 00:09:17.586
So they're pulling that data from off chain and verifying it via zero knowledge proof, as well as scoring how close of the responses are using an AI model like Chat GPT 4o.

00:09:18.225 --> 00:09:23.186
And again they're pulling the inference results via ZKTLS.

00:09:23.760 --> 00:09:25.380
And I think that's so cool.

00:09:25.471 --> 00:09:32.721
What are some of these more novel experiences that you can be building that are applicable to a wide range of people?

00:09:33.160 --> 00:09:36.311
So I'd probably be focused on one of those two areas.

00:09:37.020 --> 00:09:52.360
Last technology I think that enables different trade offs or yeah, that breaks previous trade offs that used to exist is you've previously had to choose between verifiability and privacy.

00:09:52.990 --> 00:09:58.020
And I think trusted execution environments or trusted hardware really breaks that.

00:09:58.020 --> 00:10:00.130
So what are some things you can do there?

00:10:00.421 --> 00:10:16.671
Whether that means, being able to trust private computation or being able to see what if you can trust people with different pieces of information, like a block builders and what new experiences can come out of that.

00:10:17.171 --> 00:10:17.490
Yeah.

00:10:17.490 --> 00:10:28.985
So essentially are you saying that as a founder, I should get to building, and then all related about the L2s and the scalability, because that will come in as there's a lot of solutions in the market.

00:10:29.875 --> 00:10:37.755
It really depends on who you are as a founder, what your novel skill sets are, where you see your competitive advantages.

00:10:38.086 --> 00:10:45.000
For me as a founder, I think that these, more emerging areas of tech have less eyes on them.

00:10:45.000 --> 00:10:48.081
And there's a lot of potential in these three areas.

00:10:48.910 --> 00:10:49.721
Fascinating.

00:10:50.120 --> 00:10:56.221
Going back into your academic journey, you graduated in Arbor and what was your path there?

00:10:56.230 --> 00:10:58.880
Did you get immediately your first job in blockchain?

00:10:59.110 --> 00:11:07.811
Did you not get, what advice would you have for those they're about to graduate and upcoming 2025 potentially looking into entering the space?

00:11:08.311 --> 00:11:11.510
I was working in crypto pretty much all throughout college.

00:11:11.510 --> 00:11:14.441
I worked at Avalabs, the team behind Avalanche.

00:11:14.801 --> 00:11:23.211
I jumped crypto at the smaller Cosmos AMM called Duality and more recently now at EigenLayer after college.

00:11:23.711 --> 00:11:58.640
And I think you have to ask yourself, what is your risk tolerance for sure that's a baseline level of question, but you should also consider how risky is working in crypto really and I think the answer is not that much and you should strongly consider it, because if you look at the scope of projected outcomes, let's say you join a startup and, you give it maybe one, two, four years to see if it works out by the time you graduate college, and then have been working for four years.

00:11:58.975 --> 00:12:00.796
You're still only just 26.

00:12:01.046 --> 00:12:03.816
So you're pretty young at that time.

00:12:03.995 --> 00:12:08.235
And you can still pivot to whatever you need to, in the worst case scenario.

00:12:08.245 --> 00:12:17.076
Worst, if a startup you join completely goes defunct, then you can still do a job in TradFi or Tech or whatever you want afterwards.

00:12:17.306 --> 00:12:20.716
As long as you're learning valuable skills during the process.

00:12:21.235 --> 00:12:29.931
But, the asymmetric upside is just unmatched, both financially, but more importantly from what you're learning.

00:12:30.350 --> 00:12:36.211
Because when you're in crypto, people don't care about your previous credentials.

00:12:36.211 --> 00:12:37.961
They care about what you can do for them.

00:12:38.390 --> 00:12:45.640
It's a very merit based industry by design because we have no idea who Satoshi is, right?

00:12:45.640 --> 00:12:51.811
He could have just been a 12, 13 year old theoretically, with no credentials whatsoever.

00:12:52.130 --> 00:12:58.541
So because of that, we care about what you can do, not about what your pedigree says you can do.

00:12:59.010 --> 00:13:06.971
And what that enables is for you to take wild risks and do many iterations and be constantly learning.

00:13:07.270 --> 00:13:13.431
I'll say that at EigenLayer I've learned so much more than at any other previous job I've had.

00:13:13.461 --> 00:13:16.870
I've also explored working in TradFi.

00:13:16.870 --> 00:13:17.480
I was at J.P.

00:13:17.480 --> 00:13:21.441
Morgan for a bit, and did internships at larger companies.

00:13:21.441 --> 00:13:53.744
And in a month at EigenLayer I was learning more about how people think, what motivates builders in different ecosystems and thoughts about technical architectures of okay, let me know if this is getting too deep into the weeds, but for AVS, you're like, okay, what novel use cases does this, EigenLayer AVS enable by the fact that you have verifiable off chain software that can be run by, in any distributed system context that you see fit.

00:13:53.754 --> 00:14:03.918
So whether that means you're making a request maybe to generate a zero knowledge proof and you only send it to one node rather than to everyone.

00:14:04.229 --> 00:14:09.599
And then you have, you randomly select two or three nodes to verify that ZK proof.

00:14:10.009 --> 00:14:16.359
You can just be so much more efficient than if you push everything into this blockchain format.

00:14:16.818 --> 00:14:22.614
Because blockchains by nature of being finite state machines are pretty restrictive.

00:14:23.214 --> 00:14:27.323
And I think AVS has just opened up that design space significantly.

00:14:27.774 --> 00:14:28.394
And RAMP.

00:14:28.474 --> 00:14:35.188
But I just want to say that consider crypto because the upside is unmatched and you can learn a lot.

00:14:35.839 --> 00:14:39.629
And you mentioned two terms I think is important to dive deep for those who don't know.

00:14:39.629 --> 00:14:41.078
What is AVS and RAMP?

00:14:41.578 --> 00:14:46.989
So an EigenLayer AVS is an actively validated service.

00:14:47.009 --> 00:14:52.399
And what this just means is that a blockchain is a decentralized network.

00:14:52.428 --> 00:14:59.724
You have different validators who are validating the transactions that are occurring on this distributed ledger.

00:15:00.423 --> 00:15:04.943
But a blockchain has a lot of different restrictions on top.

00:15:05.303 --> 00:15:08.094
It can't natively take off chain data.

00:15:08.683 --> 00:15:22.614
There's a gas limit that's made by design, mostly because every node in the network has to run every computation and finally, you can't make different distributed systems designs.

00:15:23.033 --> 00:15:35.583
And so what an actively validated service does is it inherits trust from Ethereum, via the EigenLayer protocol and uses this trust to bootstrap other decentralized networks.

00:15:35.984 --> 00:15:45.874
And what this could look like is it could be a zero knowledge co processor, which is just generating ZK proofs on demand.

00:15:46.153 --> 00:15:52.024
Or it could be an auction network that can be used to reduce, MEV and AMMs.

00:15:52.524 --> 00:16:13.604
Or, it could be like a decentralized policy network that is looking at off chain data, to see whether different addresses have been OFAC sanctioned and whether they can interact with your protocol safely and then be integrated as a hook that your smart contracts can call.

00:16:14.224 --> 00:16:24.793
And so there's just so many different design spaces when you say: What if we look at decentralization or verification beyond what a blockchain has?

00:16:24.823 --> 00:16:33.553
And that's the core thesis of EigenLayer is just unleashing the floodgates of open innovation beyond blockchains, but still crypto.

00:16:34.053 --> 00:16:40.734
And throughout your career, it sounds like you had the experience to work at AVAX and then Cosmos, and now you're in the Ethereum ecosystem.

00:16:41.464 --> 00:16:41.984
Yeah.

00:16:42.634 --> 00:16:51.668
Do you think each ecosystem is specifically designed for something like I read a lot of times AVAX is for gaming, and Cosmos might be for interoperability.

00:16:52.149 --> 00:16:54.958
What is your opinion since you worked in each one of these?

00:16:55.639 --> 00:17:01.219
I have a bit of a controversial take that it's really just tech at the end of the day.

00:17:01.249 --> 00:17:10.499
And the reason people say Avalanche is for gaming is because like a lot of their BD focus was on gaming and also DeFi.

00:17:10.838 --> 00:17:20.558
On Cosmos, it is true that they've built this, wonderful protocol called IPC, which is great for interop between these single slot finality chains.

00:17:20.858 --> 00:17:22.969
But really what is interop?

00:17:23.058 --> 00:17:28.519
It's meant to connect you from point A to point B, but you don't care about the connections.

00:17:28.519 --> 00:17:31.568
You care about what each thing can do for you.

00:17:31.969 --> 00:17:43.239
So when I'd say how people should go about choosing what ecosystem to go into, I frankly wouldn't choose it based off of the ecosystem.

00:17:43.479 --> 00:17:52.808
I'd say, what are some interesting problems that need to be solved and what tech stack bet is best equipped to help me serve this problem?

00:17:53.519 --> 00:18:05.463
I love EigenLayer and I think that EigenLayer did something very similar, they saw we want to bootstrap new decentralized networks, by inheriting trust from other protocols.

00:18:05.824 --> 00:18:09.743
So you say, okay, what's the largest source of decentralized trust?

00:18:10.203 --> 00:18:12.513
And really there are three options.

00:18:12.513 --> 00:18:14.673
There's, Bitcoin, Ethereum, and Solana.

00:18:15.094 --> 00:18:21.733
I'd say Bitcoin is fairly restrictive in what you can build on top, just because they don't have smart contracts.

00:18:21.743 --> 00:18:27.068
So that just made it significantly harder and what the tech stack didn't suit our needs.

00:18:27.439 --> 00:18:45.459
Between Ethereum and Solana, Ethereum has a much larger source of decentralized trust, both by the market cap and also if you look at Ethereum's focus on including solo stakers and like solo validators.

00:18:45.959 --> 00:18:53.548
It just made sense and that's why EigenLayer is on Ethereum not because of Ethereum alignment.

00:18:54.038 --> 00:19:02.169
We are aligned to Ethereum, but that's because Ethereum and EigenLayer, they go hand in hand.

00:19:02.179 --> 00:19:03.308
It makes sense.

00:19:03.528 --> 00:19:04.838
And it's inevitable.

00:19:05.769 --> 00:19:06.239
Makes sense.

00:19:06.249 --> 00:19:10.419
Do you think that was part of the reason of EigenLayer being so successful?

00:19:10.638 --> 00:19:16.219
And for those listening, if you're not familiar with EigenLayer, it's one of the largest DeFi protocols.

00:19:16.433 --> 00:19:22.993
They surpassed, correct me if I'm wrong,$20 billion in TVL, since the launch in June, 2023.

00:19:23.493 --> 00:19:26.538
It was working in synchrony with Ethereum.

00:19:26.538 --> 00:19:35.348
I think a few reasons why it's become so successful, and I'll say we still haven't found like 100 percent product market fit yet.

00:19:35.378 --> 00:19:48.713
There's still a lot of work left to be done, but the reason behind most of EigenLayer's success so far is one, a relentless focus on what are the real problems that our users are facing?

00:19:49.013 --> 00:20:04.644
And we chose the tech stack as a result of, it's best suited to solve these problems rather than alignment, which can lead you to choose, less good tech stacks for your specific solution.

00:20:05.173 --> 00:20:10.423
And two, I'd say it's this idea that EigenLayer is inevitable.

00:20:10.473 --> 00:20:12.463
There's so much Ethereum stake.

00:20:12.794 --> 00:20:18.074
And right now all it's doing is validating the Ethereum network, which entails two things.

00:20:18.453 --> 00:20:25.284
One, looking at these transactions, making sure that they're all the correct state, like following correct state transition functions.

00:20:25.874 --> 00:20:32.273
And two is saying, okay, in the case of a fork, what is the longest chain?

00:20:32.273 --> 00:20:34.634
What is going to be the canonical fork?

00:20:35.634 --> 00:20:39.713
And this seems like a bit of an over solved problem.

00:20:39.723 --> 00:20:46.949
What if you can take the stake to do so much more, and really create this whole new economy.

00:20:47.249 --> 00:20:56.798
Because the internet is not just Amazon and not just Google, it's this whole swath of applications and microservices.

00:20:57.068 --> 00:21:01.269
And that's what EigenLayer tries to enable for the decentralized universe as well.

00:21:01.769 --> 00:21:08.483
And since we're touching on EigenLayer for those who don't understand, I didn't know what it does, what is in the purpose.

00:21:08.483 --> 00:21:13.824
For Grannies, since Christmas is approaching, and I think that could help a lot of people getting situated here.

00:21:14.493 --> 00:21:15.163
Definitely.

00:21:15.413 --> 00:21:27.534
I'd say EigenLayer is a developer platform that allows you to turn any software into a verifiable software and make credible commitments.

00:21:28.273 --> 00:21:31.294
What does this look like in practice?

00:21:31.483 --> 00:21:31.663
It says, okay.

00:21:31.713 --> 00:21:39.554
Let's you make any financial experiment and it decouples the person building it, the innovation from the trust behind it.

00:21:39.784 --> 00:21:45.374
You can look at the Uniswap V2 or V3 source code and you don't have to trust Hayden Adams.

00:21:45.374 --> 00:21:47.394
You can just say, hey, this makes sense.

00:21:47.828 --> 00:22:07.983
So what if we can do that for any software whether that's what if we can make Uber verifiable and they can make a credible commitment that says we're not going to take more than a 10 or more than a 20 percent fee on a transaction, like per ride or per transaction.

00:22:08.463 --> 00:22:18.513
You could say, you could make Facebook verifiable and say: Okay, we are going to be make sure that our user base is, like private.

00:22:18.683 --> 00:22:26.963
Or we want to gain your trust because you're giving us so much data and we're not going to unleash your data to anyone else.

00:22:27.163 --> 00:22:37.473
And so you can make that credible commitment with Facebook, in a more crypto native sense, you could say, Hey, we have this data availability problem, right?

00:22:37.523 --> 00:22:47.923
Rather than make a blockchain to solve the data availability problem where you need to make sure that all data is available under fault tolerant conditions.

00:22:48.413 --> 00:22:51.153
If you have a blockchain, then everyone needs to store everything.

00:22:51.433 --> 00:23:10.463
If you use an AVS like Eigenlayer, or through Eigenlayer, you can design a whole new decentralized system where you take advantage of horizontal scaling such that not every node needs to have contain all of the data and that unleashes whole new possibilities.

00:23:10.898 --> 00:23:11.519
Very interesting.

00:23:11.519 --> 00:23:17.048
So do you consider EigenLayer is not an L2, but an AVS as you call it in that terminology?

00:23:17.898 --> 00:23:22.338
I'd say EigenLayer enables AVSs through restaking.

00:23:22.868 --> 00:23:24.409
So it's not a DeFi protocol.

00:23:24.410 --> 00:23:26.259
It's not an L1, it's not an L2.

00:23:26.618 --> 00:23:30.189
It's a more decentralized component of infra.

00:23:30.588 --> 00:23:37.929
I like to think of it as the AWS for Web3 because it enables so many different components.

00:23:38.374 --> 00:23:44.294
Whether that's EC2 or S3 or for crypto, DA or oracles.

00:23:44.673 --> 00:23:47.284
And all of these things can be built on top of EigenLayer.

00:23:47.304 --> 00:23:51.034
And we're seeing a lot of these protocols built right now.

00:23:51.074 --> 00:23:55.144
Like Layer0 is building a crypto economically secure bridge.

00:23:55.584 --> 00:24:06.095
Infura is building a decentralized, RPC network on top called DIN, and it's really cool to see how many different things are being built on top.

00:24:06.575 --> 00:24:16.525
And you think this is because EigenLayer is a place where people can trust and holds these protocols accountable without them sharing the information per se, like you gave on the example earlier?

00:24:17.025 --> 00:24:19.704
Yeah, that's one way of looking at it.

00:24:19.704 --> 00:24:24.625
You could say that EigenLayer, Eigen means your own, right?

00:24:24.825 --> 00:24:33.045
So it enables you to build your own layer, your own infra, whatever you want to build, you can build on top of, EigenLayer.

00:24:33.724 --> 00:24:45.568
And really it's this, behind the scenes infra that allows you to connect and tap into the decentralized trust of Ethereum or any other ERC 20.

00:24:45.710 --> 00:24:52.039
Or if you want to use Eigen, then inner subjectivity, which I'm not sure if we have time to go into.

00:24:52.720 --> 00:24:53.509
Yeah, for sure.

00:24:53.779 --> 00:24:58.779
And earlier on, you mentioned that you're working in something very specifically called Trusted Execution Environments.

00:24:59.265 --> 00:25:05.984
How do you see the evolution and these shaping in terms of secure computation in these decentralized systems?

00:25:06.984 --> 00:25:09.075
So I'm a researcher at EigenLayer.

00:25:09.085 --> 00:25:19.484
So one of the things I've been researching is trusted execution environments or TEEs and how they can be used in the context of decentralized systems.

00:25:20.035 --> 00:25:28.509
And I think these are super powerful because, decentralized networks enable verifiability, but you don't have any privacy there.

00:25:28.799 --> 00:25:34.609
Some people like Aztec and Aleo are working on that, but it's very complex.

00:25:35.089 --> 00:25:48.410
Trusted execution environments, say that, okay, if you put some code in this box, it's going to run exactly as the code that you say is running is going to be the code that actually runs.

00:25:48.930 --> 00:25:57.539
And two, you can enable private, so you have verifiable execution, you can have private inputs and outputs, which is super powerful.

00:25:57.549 --> 00:26:04.009
So if you combine these two together, you can actually create new technology that breaks previous trade offs.

00:26:04.509 --> 00:26:05.809
Oh, that sounds fascinating.

00:26:06.190 --> 00:26:15.355
And what role do you see these playing in terms of fostering trust and privacy in a multi party computational system?

00:26:15.964 --> 00:26:19.005
I'll say, two things that I found interesting recently.

00:26:19.305 --> 00:26:27.234
One is Uniswap, they're building Unichain and they have something called Rollup Boost that they've built in collaboration with Flashbots.

00:26:27.644 --> 00:26:39.940
And what this does is, it says, what if we could give you the guarantees of a private RPC, but you don't need to send your data to a single block builder?

00:26:40.480 --> 00:26:50.420
And so they separate the sequencing and the block building layer, so that you can send your transactions and the transactions get sent to all block builders.

00:26:50.750 --> 00:27:00.115
But you can trust that they won't be able to use your transaction to attack you via MEV attacks like front running or sandwiching.

00:27:01.035 --> 00:27:04.795
But they will use this to order your transaction fairly.

00:27:05.075 --> 00:27:19.035
And this enables a whole suite of new things that are not necessarily incentive compatible, but they are commitments that if made credibly, that enables a broader design space.

00:27:19.424 --> 00:27:37.285
And so what Uniswap does is they have these flash blocks that are 250 millisecond, sub blocks within a, one second block, and it enables a better user experience through these faster pre confirmations for users and for traders.

00:27:37.954 --> 00:27:59.315
The second thing is something called SUAVE which again is related to this block building phenomenon and it's again by flashbots and they're saying what if we can share your transactions to every block builder via this decentralized network called SUAVE and then run auctions to see which block builder gets to win.

00:28:00.204 --> 00:28:07.414
And again, they won't be able to sandwich you or front run you because their algorithms are running in a TEE.

00:28:07.914 --> 00:28:15.115
What are some of the most promising applications of TEEs in DeFi and Web3 ecosystems that you've seen so far?

00:28:16.095 --> 00:28:22.105
Biggest one is definitely block building, for now, but I think privacy is probably going to be the biggest emergent.

00:28:23.075 --> 00:28:37.775
I fully expect to see in 2025 an L1 or L2 that is completely private, by making use of trusted execution environments and I'm excited to see how that goes.

00:28:38.095 --> 00:28:40.464
I think it can actually do quite well.

00:28:41.394 --> 00:28:47.144
Then the third thing I'll say is verifiable AI inference.

00:28:47.625 --> 00:29:16.815
And what we're seeing is that a lot of these AI chatbots like Eliza and AI 16Z want to run their AI models inside a TEE, because the TEE is attesting to the inference of the models, that there actually is an AI model that's controlling this Twitter account, or that's controlling this private key rather than some person behind the scenes who's making all these decisions.

00:29:17.315 --> 00:29:26.795
One last question, if we're at a time machine and you had a chance to go back to your Ann Arbor version of yourself, what would you tell to that Ishaan back in the day?

00:29:27.295 --> 00:29:30.416
Aside from buying it?

00:29:30.416 --> 00:29:31.656
Can I say buy Ethereum?

00:29:32.021 --> 00:29:34.011
Yeah.

00:29:34.392 --> 00:29:34.692
Yeah.

00:29:34.892 --> 00:29:40.021
Aside from trading, I'd say, actually, that's a good question.

00:29:40.521 --> 00:30:21.561
I'd say, I think I like to hop around into different topics, because it gives optionality and you get to learn more about broad suite of topics, but I'd say just don't be afraid to dive take the plunge and dive in work on a project for a year or two doesn't have to be your own if you don't want to start your own right now, but work on one project because there's so much you learn by going deep into an area that you don't just by going broad into many different areas and I think with EigenLayer I'm finally taking that advice to heart and really going deep into AVSs which has been a great time.

00:30:22.511 --> 00:30:23.352
Awesome, Ishaan.

00:30:23.372 --> 00:30:25.102
Thank you so much for being in our podcast.

00:30:25.652 --> 00:30:26.061
Yeah.

00:30:26.092 --> 00:30:26.951
Thank you, Tony.

00:30:27.082 --> 00:30:27.501
See you.

00:30:28.152 --> 00:30:33.301
Thanks for tuning in to the Chain Stories podcast, where disruptors become trailblazers.

00:30:33.582 --> 00:30:39.701
Don't forget to subscribe to hear more inspiring stories from those who are pushing boundaries in digital assets.

00:30:39.991 --> 00:30:45.162
Brought to you by Dropout Capital, where bold vision transforms into reality.

00:30:45.321 --> 00:30:49.672
Check out our social media links down below and visit us online at dropout.

00:30:49.961 --> 00:30:50.571
capital.

00:30:50.761 --> 00:30:56.392
And remember, the future belongs to those who dare to challenge the norm.