Preston Van Loon, Co-Founder of Prysmatic Labs: The Plan to Scale Ethereum

Preston Van Loon is an Ethereum Protocol Developer and Co-Founder of Prysmatic Labs. In this conversation, we discuss testing frameworks, working at Google, blcokchain scalability, Ethereum 2.0, sharding, Ethereum Foundation Grants and what Preston and his team are building at Prysmatic Labs. ----- If you enjoyed this conversation, share it with your colleagues & friends, rate, review, and subscribe. This podcast is presented by BlockWorks Group. For exclusive content and events that provide insights into the crypto and blockchain space, visit them at: https://www.blockworksgroup.io

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:01Anthony Pompliano?

  What's up everyone, this is Anthony Pompliano. Most of you know me as Pomp. You're listening to Off the Chain, simply the best podcast in crypto. Let's kick this thing off.

• 0:13Anthony Pompliano?

  Preston Van Loon is an Ethereum Protocol developer and co-founder of Prysmatic Labs. In this conversation, we discuss testing frameworks, working at Google, blockchain scalability, Ethereum 2.0, sharding, Ethereum Foundation grants, and what Preston and his team are building at Prysmatic Labs. I really enjoyed this conversation, and I hope you do as well.

• 0:37Speaker 01

  Anthony Pompliano is a partner at Morgan Creek Digital. All opinions expressed by Pomp or his guests on this podcast are solely their opinions and do not reflect the opinions of Morgan Creek Digital or Morgan Creek Capital Management. You should not treat any opinion expressed by Pomp as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of his opinion. This podcast is for informational purposes only.

• 1:04Anthony Pompliano?

  Alright guys, bang bang, I'm here with Preston. I feel like I'm going to learn a ton today because frankly, he's got a bunch of experience in working on things that I don't really understand. So, thank you so much for taking the time to come over from Brooklyn and do this conversation.

• 1:17Speaker 02

  Yeah, thanks for having me.

• 1:18Anthony Pompliano?

  Absolutely. Alright, let's start with your background, kind of pre-crypto. What were you doing?

• 1:23Speaker 02

  Sure, so, um, let's start with, um, university. So, I went to Middle Tennessee State University. I originally went there for aerospace.

• 1:34Anthony Pompliano?

  Okay.

• 1:35Speaker 02

  And the whole time, I've been, you know, interested in computers, like messing around, come from like playing a lot of video games and like, I want to run my own server and add mods and stuff. So, I got as far as I got my pilot's license, a single legit pilot's license, and then I decided I wanted to switch over to computer science because it's just more interesting to me.

• 1:56Anthony Pompliano?

  Some people would compare like you wanted to go to the moon aerospace and then like crypto going to the moon as well, right?

• 2:02Speaker 02

  Yeah, that that might be, you know, I I just wanted to have a really big like positive impact on the world. And I think that I can maybe have a better chance to do that in computer science.

• 2:12Anthony Pompliano?

  So, okay.

• 2:13Speaker 02

  So I switched over, um, but I hate school, so eventually dropped out. Um, and then I just started working on like random stuff. So I was working on a e-commerce company and they were selling things on eBay and they need to automate a lot of stuff. And then I hopped over to another company that was doing like, uh, some like financing in the trucking industry. And they had all their software in house. And then I was working on like a point of sale company. Uh, and then Google plucked me out of there and I moved to New York. So I worked at Google for a few years.

• 2:44Anthony Pompliano?

  So you were doing all of pre-Google, you were in Tennessee, still doing all this?

• 2:49Speaker 02

  I was in Texas.

• 2:50Anthony Pompliano?

  In Texas.

• 2:50Speaker 02

  So I went from Tennessee to Texas, I'm from Texas. My family's all there, so I moved back, uh, back home and just started working.

• 2:57Anthony Pompliano?

  Got it, okay. And then you end up in, uh, New York, working at Google.

• 3:00Speaker 02

  That's right, yeah.

• 3:01Anthony Pompliano?

  And what was the draw to go to, to Google?

• 3:03Speaker 02

  Um, yeah, Google is so is really interesting company to work at. It's a lot of opportunity, a lot of things to do. Uh, and New York was a place I'd always been like kind of interested in going there. I probably never would have moved here if it wasn't for, for Google though. But I thought, um, Google has a lot of, they're solving a lot of hard problems and that's interesting to me. And I, I'm drawn to that. I want to, to solve the hardest problem I can and, and hopefully make a big impact there.

• 3:29Anthony Pompliano?

  Yep, and, and it feels like to me, you know, I worked at Facebook, you worked at Google. A lot of these companies, there's a natural draw to them because you kind of know that, hey, you're gonna work on hard problems. You're gonna work on them at scale, right? And you're gonna have a, a good chance of working with really smart people alongside you on those hard problems. And so it's kind of, you almost don't even need to know what you're gonna work on yet, right? There's already an allure to those companies. And then once you get in, you can kind of figure out what you want to do. Is that accurate?

• 3:56Speaker 02

  That is totally accurate. And the Google hiring process is that you're, you're, you're hired as a software engineer and you don't know, like one, at that point, you don't know what team you're gonna be on. But you're just so pumped because anything you're gonna land on is gonna be amazing.

• 4:07Anthony Pompliano?

  Yeah, it's funny, Facebook's product manager, uh, track is very similar. You, you know, you're going to be a product manager, but you don't necessarily know what team you're gonna work on. And then during the onboarding, you basically figure out what team you're gonna be on.

• 4:19Speaker 02

  Yeah.

• 4:19Anthony Pompliano?

  It's pretty cool. Um, all right, so what, uh, what'd you work on at Google?

• 4:22Speaker 02

  I worked in the ad space on a product called DoubleClick. Uh, it's recently renamed to Display Video three sixty as part of their advertising suite. But it's for more like, uh, ad agencies or like big, um, big ad companies. So it's bigger than, you know, AdWords is probably the something that you and I might use because we have low spend, but if, if I want to spend a million dollars on a campaign, I probably would want the support from a system like this. So I was a software engineer on that platform, mostly doing on the advertiser side, uh, doing web development for, for their web application mostly. And our team was pretty small, so we managed, um, everything from deployments to our own site reliability, keeping things in line.

• 5:06Speaker 02

  So I kind of got a really well-rounded, uh, set of experience there because we were doing everything on our team.

• 5:12Anthony Pompliano?

  Yeah, it, it is pretty funny how, uh, whether it's Facebook, Google, Amazon, you know, e-Twitter, et cetera, a lot of these companies, although they are large in number of employees, the individual teams are pretty small, right? And, and they really don't use, um, an incredible amount of resources to accomplish things. And it feels like, um, that's part of the magic sauce, if you will, right? Is even though you have the ability to work at scale, you still might only have ten or less people working on the project, right? And, and so you get the backing and the resources of this large organization, but the actual core team, it's like a startup almost, right?

• 5:49Speaker 02

  Yeah, so, so DoubleClick is, is a pretty big product. There's probably a hundred people working on it, but they divided into smaller pieces. So each team, um, it maybe is eight, eight or so people and say, oh, you own the front end of the application or part of the front end. So that was kind of what our team did.

• 6:06Anthony Pompliano?

  Got it. And, and like, it's pretty incredible to think of DoubleClick, which is probably, you know, billions of dollars a year going through it. A hundred people is actually not, is nothing compared to the, you know, tens of thousands that work at Google.

• 6:19Speaker 02

  Yeah, I mean, it's it's a they said, you know, keeps the lights on, right? Ads keep the lights on at Google, yeah.

• 6:26Anthony Pompliano?

  Absolutely, and it empowers everything else.

• 6:28Speaker 02

  Yeah.

• 6:29Anthony Pompliano?

  Right? All right, cool. And so one of the things that you've talked a lot about in the past that I find fascinating is this idea of just like testing, right? And experimentation.

• 6:37Speaker 02

  Right.

• 6:37Anthony Pompliano?

  Maybe talk a little bit about kind of where your fascination or, you know, focus on testing and experimentation comes from?

• 6:44Speaker 02

  Yeah. So we'll start with the testing part is that when I'm building something, I don't trust that just like clicking the button, that it does a thing, it's going to work. Like, maybe it worked then, but if it if I don't have good testing infrastructure in place, I don't I'm not assured that I'm not going to that I'm going to break it later. So on our team, we really advocate and push for and require that everyone's writing really good tests. So that's starting with like the the smallest level you can go, which are unit tests. So if I have something that multiplies two numbers, I'm going to write a suite of tests. This is these inputs, give me this output. And anytime someone new comes along, they're editing it, they're they're changing it up.

• 7:28Speaker 02

  If they accidentally break it, we'll know before they even check the code in. So that's that's super important.

• 7:34Speaker 02

  Going a step higher, we want to ensure that things in the actual system don't break. So doing like integration testing, which is to actually like stand up the whole cluster and run things and then check the in-state or, you know, various checkpoints of the state and just see that, oh, it's still advancing. So in our case, we're building a blockchain, we want to see that does rolling out this new release, will it sync with other peers or will it will it break entirely? So that's really important to me for testing because these kind of like, you can't, especially once it goes to main net, you can't. It's really hard to do upgrades and to roll it out to the entire network.

• 8:19Speaker 02

  So you have to get it right the first time, pretty much. So we're we're really, really big on providing this testing infrastructure and making sure that things are done right the first time before they're before they're sent out there.

• 8:32Anthony Pompliano?

  And so when you look at the testing piece, you know, one, one of the frameworks that I took from working at these large technology companies was this idea that first you have to design the correct test, right? The test has to actually be accurate. And what we mean by that is you're trying to solve a problem and there's multiple solutions. You come up with some sort of framework or test that you're going to put a hypothesis through. And then the second piece is you have to execute the test perfectly, right? And the idea is if you don't have perfect execution and the test doesn't work, you're then left wondering, well, was it because we executed poorly or was it because the test was invalid, right? But if you always execute perfectly, then you can very quickly tell either the

• 9:16Anthony Pompliano?

  test failed or it didn't, because you never have to worry about that execution. How do you think about that in a world where you're talking about testing in kind of the code is continuously checked every time there's changes made?

• 9:30Speaker 02

  Exactly right. And another thing to think about is what are you testing? So if you, if you're just testing, like basically what the method says, I mean, that's great. But well, you really want to test our behaviors. So behavior might be ensuring that a user can login, not just like that their password has passed all these validity checks, but, but really that it goes all the way through the system and that the behavior, we're testing behaviors and not just methods. So that's, that's really important to me and keeping it in continuous integration and automated. That's another important thing because when we're doing code reviews and things, the most annoying thing is to be, to get feedback that, hey, I ran your tests and they fail. If we can automate that, a robot tells you

• 10:14Speaker 02

  it's faster feedback cycle, we can get things done faster.

• 10:17Anthony Pompliano?

  Yeah, I've, I forget the name of the company, but I recently saw, recently, maybe like last six months, there was a company that basically creates some sort of technology that can identify all the potential vulnerabilities and basically tries to penetrate software, right? Not just blockchain related, but just in general. And it's almost like this automated testing program can drastically reduce the number of vulnerabilities, you know, on day one. And then also it can help identify the, the most critical areas that the team should go back and try to shore up. Is that happening in the blockchain space as well? Or is that more outside of blockchain right now?

• 10:55Speaker 02

  I I'm not too familiar with uh it inside of blockchain. I know it's really popular outside like automated tooling for security testing is huge. Another thing for security testing is having the test like really help for audits so they can actually see how the code is supposed to be used. In this isolated scope, so they can look at this. Look at the big picture, but the big picture might be a hundred thousand lines of code and they if they. They can't don't have time to read all of it and to understand one method, they look at the test. That's that's super important. In terms of like blockchain, um it's it's uh I think things like smart contracts are kind of hard to test right now.

• 11:34Speaker 02

  Um but there are like frameworks coming out that that that do help you do that and and that's getting a lot more mature. So I'm pretty happy to see that.

• 11:42Anthony Pompliano?

  Got it. And and so let's switch gears to talk about Ethereum, right? Obviously, um you're you're you're quite bullish and and think highly of uh of the platform. Maybe just give us a kind of an understanding of, you know, how do you view it today? Where are we? What's going well? What's not going well? What what do you think the challenges are?

• 11:58Speaker 02

  Uh, yeah, so.

• 12:00Anthony Pompliano?

  The most loaded question in the history of this podcast.

• 12:03Speaker 02

  I know. Yeah, that's that's a pretty loaded question. So I'm gonna I'm gonna go with uh with my bias answer is that that I want to talk about scalability. So Ethereum has had this great success where they've they've hit the. Peak of what the blockchain can do. So they've already like maxed out. Uh if you want to call it transactions per second being the the scalable measurement of of a blockchain. They did max that out already uh at the hype of uh twenty seventeen and other other points in time. So. That's a great that's kind of a great problem to have. You have this blockchain that's so successful that um it's really hit the capacity. So now we're wondering. Okay, well, how do we take it to the next level?

• 12:45Speaker 02

  And so the research team, uh our team, Prysmatic Labs, and a lot of other developers are building on it. Uh Ethereum two point O, which is just in in entirely new rewrite of of the the whole ecosystem. So. So things that aren't going well are that blockchain doesn't scale and we know that. So if if if we're about to enter another run like we saw in 2017, where there's lots of excitement, we're getting lots of users on board, kind of getting closer to that global adoption that we're all striving for, we're going to hit this peak again where where the blockchain starts to struggle, we're getting all the blocks are full of people, you know, paying

• 13:26Speaker 02

  high gas fees or waiting a long time for their transactions. So so that's still a problem. It's the biggest problem in my mind. The second biggest problem is usability UX and just general like knowledge of how things work. So one great thing that Ethereum community has been doing is they launched the new ethereum.org, which is a new redesign that has some controversial feedback. Some people like it, some people don't. But the point is the content is better. It's hard to argue that this is more substance to it than it did before. So I think that's a huge improvement and that it's on GitHub. So anybody can suggest changes.

• 14:06Anthony Pompliano?

  Yep.

• 14:07Speaker 02

  Or file a bug or whatever. So that's.

• 14:09Anthony Pompliano?

  This was much more controversial than I think I understood.

• 14:13Speaker 02

  Right.

• 14:13Anthony Pompliano?

  Right? Like, like, it's actually, people are, what are the detractors point, right? The people who don't like it, what are they complaining about?

• 14:20Speaker 02

  The look of it, really.

• 14:22Anthony Pompliano?

  Okay.

• 14:22Speaker 02

  It not, not anything else that I know. They just say, wow, this looks terrible. And they're just complaining. Some, some people have been helpful where they'll say, I hate this. And then they'll put their explicit feedback on the site. And some people even submit pull requests, say, I hate this. This is what you need to do to fix it. And I've done it for you. Just accept it. So.

• 14:41Anthony Pompliano?

  Got it.

• 14:42Speaker 02

  That's actually like way better. And it's a step in the right direction. So maybe the, I mean, they even claim that it's a work in progress. Like, they, they, nobody said this is the final thing. So I'm super happy that they're, that they're doing it all because the site before was from, I don't know, 2014, and no one's touched it since then. So big improvement, I think.

• 15:01Anthony Pompliano?

  Yeah.

• 15:02Speaker 02

  Yeah.

• 15:02Anthony Pompliano?

  Absolutely. And so let's get to Prysmatic Labs, right? And maybe talk a little bit about what you're actually building and why you're doing it.

• 15:11Speaker 02

  Uh, okay, so Prysmatic Labs, we're a team, independent team of blockchain developers who are interested in scaling Ethereum. So what that means is we're, we're developing an Ethereum 2.0 client. Ethereum 2.0 is the sharding and proof of stake rewrite of Ethereum. So everything is completely different. It's a huge undertaking and.

• 15:36Anthony Pompliano?

  So what you're essentially doing, right? And you correct me if I'm wrong, because this is the part where you're gonna do some teaching to me, is uh, there's Ethereum, we'll call it Ethereum 1.0, which is this smart contract platform that um, has a lot of advantages. It had a lot of breakthroughs in terms of how people thought about using a blockchain and some of the applications it could have. But it runs into the scaling issue. What you guys essentially are saying is let's change to proof of stake and let's make some other uh, improvements to the Ethereum model and you label that Ethereum 2.0 and that's what you're building.

• 16:09Speaker 02

  Exactly. So um, it's a, it's a new blockchain that is fundamentally still Ethereum.

• 16:16Anthony Pompliano?

  Okay.

• 16:16Speaker 02

  Um, but the, for, for some amount of time, there will be two chains running in parallel.

• 16:21Anthony Pompliano?

  Okay.

• 16:21Speaker 02

  They'll be connected. Ethereum 1 and 2 will have a one-way bridge, so you can migrate over early. If you want to be a staker, if you're just ready to move over, you, that will, that mechanism will be there. And then I think the eventual goal is that we'll break that separation and then um, the Ethereum 1 chain will live on in some form, but the, the, the main Ethereum lives on is Ethereum 2.0 and this new proof of stake and, and sharded blockchain.

• 16:51Anthony Pompliano?

  Got it. And so um, let's, let's come back to proof of work versus proof of stake in a second, but explain what sharding is and why that's important.

• 16:58Speaker 02

  Okay, so sharding in the traditional database model is we, we have basically one computer, right? And maybe this computer has uh, one terabyte of disk, but once we start to fill up data on that disk, it's getting full. We only have one slot in, maybe only have one slot in our computer. We get another computer and then split the data between the two. So now we've effectively doubled our capacity. Now, there's a little bit of overhead because these nice computers have to talk to each other, but we think that it's so infrequent and the trade-off is worth it because we just doubled our capacity maybe for this uh, small overhead of, of doing things. So.

• 17:37Anthony Pompliano?

  So really, it's just sharding is a big word for being able to break up large pieces of data into multiple parts. And you can put those multiple parts in a more efficient manner, right? In terms of storage of the data.

• 17:48Speaker 02

  Yeah, so it's it's it's a horizontally scaling model. So instead of having like a better computer, we just added more computers. And that's what this is instead of having like a faster, better blockchain. We're we're just cloning the Ethereum one chain and we'll have a thousand chains, which can talk to each other and split up. So we have a thousand times more data storage, um, capacity throughput because we can run things in parallel. So it's it's a big improvement.

• 18:15Anthony Pompliano?

  Got it. And and so in this sharding model where you're essentially just adding more blockchains, kind of parallel to the initial one, you obviously get, um, or at least the theory, right? Or even in practice, you're getting added capacity or performance.

• 18:31Speaker 02

  Yeah.

• 18:31Anthony Pompliano?

  What the people who disagree with a sharding approach, what is their complaint?

• 18:36Speaker 02

  Um, I I guess the biggest complaint is that, um, there's the additional overhead of these shards. So if what we're what we're envisioning is that in order to take the the most advantage of sharding is that you'll localize your application. So if you're a smart contract developer, you'll you'll deploy, you have five contracts that talk to each other, put them all on the same shard.

• 18:58Anthony Pompliano?

  Got it.

• 18:58Speaker 02

  Because that's that's where the optimization comes.

• 19:01Anthony Pompliano?

  That's the most efficient place, way to actually do this.

• 19:03Speaker 02

  Yeah, but the argument is that, uh, given a random distribution of users, they're going to be randomly split over all the shards. They're going to be talking to each other and it's not a real improvement. But we think that, you know, the behavior is going to be that people localize and that layer two will take advantage of handling the cross sharding stuff. That's what I think is that, yeah, there's going to be some overhead, but we can. Put some more of the work onto the, take some work off the chain and put it on layer two to facilitate these quick, you know, getting between shards if you need to do that and keeping it infrequent.

• 19:40Anthony Pompliano?

  Got it. Okay, so that's sharding.

• 19:42Speaker 02

  Yes.

• 19:42Anthony Pompliano?

  And then let's go back to proof of state, proof of work.

• 19:44Speaker 02

  Yep.

• 19:45Anthony Pompliano?

  I think we've talked quite a bit on the podcast about proof of work.

• 19:49Speaker 02

  Sure.

• 19:50Anthony Pompliano?

  Explain what your view of the advantages of proof of state, right? So rather than having the machines actually doing the validation, this proof of state concept, what's the benefits there?

• 20:00Speaker 02

  Right, so for the listeners to recall, the proof of work is the consensus model where

• 20:08Speaker 02

  miners will produce a block by guessing this random number that we know it takes a long time, it's difficult to do. So we can verify it quickly and know that they spent the work to do it. This is proof of work. Proof of state is a different model where you don't have to spend a lot of resources to do it in terms of computing. You put up some collateral to hold yourself accountable, right? And then you do your work where you're producing blocks or testing to data to ensure that it's there and available and correct. And if you, the mechanism to keep people from acting maliciously is that if if they're proposing blocks that have invalid state transitions, so maybe they're printing money for themselves

• 20:53Speaker 02

  or just making junk, trying to attack people or harm the system. Well, you can enforce that within the base layer chain, the beacon chain, and actually penalize them by taking their deposit away. And either burning it or probably the person who was able to prove it, get some percentage of it, something to incentivize people to hold other people accountable. But the main benefit there is that you can do it on consumer hardware. You don't need to have a server farm with ASIC rigs and all this spinning tons of energy. You actually just need to be online and be listening for the data.

• 21:36Speaker 02

  And when your time comes, you just sign the correct thing and then broadcast it to the network. And it's really cheap in terms of computing.

• 21:44Anthony Pompliano?

  Got it. And so the thought process here is that what you're building is essentially a proof of state-based sharding blockchain, right?

• 21:54Speaker 02

  That's right.

• 21:54Anthony Pompliano?

  And The belief, if I understand this correctly, is Ethereum won, good idea, hit scalability issues. By going proof of stake with sharding, you're actually able to still have the quote-unquote Ethereum dream built, but in a scalable way.

• 22:09Speaker 02

  Yep.

• 22:10Anthony Pompliano?

  That's okay.

• 22:10Speaker 02

  Yep.

• 22:11Anthony Pompliano?

  What do the, what I'll call hardcore Ethereum developers and that whole community, are they on board with this? Are they open to the idea? Do they hate you? I guess, what's kind of your relationship with what I think of as like the, you know, I'll call the core developers without meaning anything other than just the name?

• 22:29Speaker 02

  Sure, yeah, I, I, in general, I think people like what we're doing. I don't, I don't hear that people hate us. I'm sure someone out there thinks negatively of the effort.

• 22:39Anthony Pompliano?

  Somebody does, for sure.

• 22:40Speaker 02

  Someone's upset about it, but in general, at least in the Ethereum community, in our little bubble, everyone is super supportive. So, um, that, that's actually like a really great question because, um, like, maybe something we can talk about is like, how are we funding this?

• 22:55Anthony Pompliano?

  Yep.

• 22:56Speaker 02

  Um, so like Prysmatic Labs doesn't have a revenue model. Like, we're building this open source client where anybody can use it for free. And we're not going to sell it to you. Like, it's open. You can download it and use it, whatever. Uh, and so we started this last year in January. And how it came about is that, um, in December, I was on, you know, the chat rooms, like, asking developers, like, hey, is anybody working on sharding? Because I see that, um, Vitalik and others have like this paper written up and it's looking pretty complete. And, you know, why is nobody working on it? Uh, and then Raul Jordan was in the same rooms as me, basically asking the same questions within the same week or so.

• 23:40Speaker 02

  So we linked up on, on there and he said, hey, you know, I'm, I'm actually gonna be in New York City, uh, next week. Uh, maybe we should have a coffee and talk about it. So, so we met up, we went over everything and said, like, okay, let's, this seems like a good idea. Let's put a team together and just start, start building it because no one's working on it. We're suffering from scaling. Let's, let's do it. So we got a team of, uh, five or so people, um, put a name on it, Prysmatic Labs. Um, Raul actually thought of it on the way over to meet me. And I didn't know that until like six months later. He's like, oh, by the way, that's when I thought of it. It was like, I was walking from the train.

• 24:19Anthony Pompliano?

  I love that.

• 24:20Speaker 02

  That was pretty cool. So we started working on it, started building. And then the Ethereum Foundation announced they wanted to give out grants for for this like type of work. And we said, wow, you know, like the stars are aligning here. Let's apply for this grant and see what happens. And we actually ended up getting in the first wave of grants, a hundred thousand dollars. So that was super cool. We were, we didn't think that this was gonna go even that far. We just wanted to work on this because it's interesting. Uh, no one else was doing it yet. And it, and it was just like something we were passionate about.

• 24:53Anthony Pompliano?

  Mm hmm.

• 24:54Speaker 02

  Um, and then it started getting, you know, we, we just started working. We flew to Thailand to hang out with the research team. And there's like Ethereum 2.0 workshop for a couple of days. That was super awesome. We came back having to, um, start over almost, um, the work we'd done so far. And then in June of, uh, last year, there was another, um, sort of Ethereum 2 workshop where we decided to take Casper, which was the proof of state model that, that was going to go to Ethereum. Uh, and it was, it was pretty close to being ready. We said, let's just merge these two ideas because some of the research for proof of state can be reused.

• 25:39Speaker 02

  So we merged that and basically had to start over again. Um, in between that time, we had received more grants from, from Aragon, which is a governance, um, DAO framework. And we got from District Zero X, Spank Chain, uh, Wanchain, um, Ethereum Community Fund. Uh, and then in October, we got a big grant from the Ethereum Foundation again. And then we were working on things and we had, you know, some pretty good capital. Uh, but we, we, we were still all working our full-time jobs because it, it wasn't quite enough for us to, you know, leave Google and our other places we were working at.

• 26:20Speaker 02

  And so Vitalik, uh, does the Yolo grant, if you've seen on Twitter, where he just, um, the guy who never gives away Ether suddenly gives out a thousand Ether at a time. So he gave us a thousand Ether, which is super cool. He gave it to ChainSafe, another, another Ethereum 2 team. Uh, and yeah, so that was like super, super, super cool. So at that point, we said, okay, you know, in January of 2019, we're gonna go full, everyone's gonna go full time and we're just gonna heads down build it. We have at least a year's worth of runway at, for five people. And we want to see this Ethereum 2.0 thing done. So we started doing it. And then today,

• 27:00Speaker 02

  this morning, um, we've launched our phase zero test network. So it's super exciting, huge milestone. It's taken us basically two months of debugging. Like it's, it's been actually running for two months, but it just wasn't ready for, for the masses yet. But, but this morning it's out there. So it's super exciting.

• 27:18Anthony Pompliano?

  Absolutely. First of all, congratulations. It's, it's exciting to be talking to you today when you get that out. Uh, maybe talk a little bit about what some of the last two months, as you guys have been testing, like what some of the bugs are. Cause I don't think a lot of people understand from the developer's perspective, how hard this stuff is. Right? Like, like it's not just like, hey, I woke up, I wrote some code and I ship it and you know, hey, I changed the world. Like you spent two months debugging something that you built for months.

• 27:43Speaker 02

  Yeah. Yeah. Oh, we worked on it for a year and debugged it for two months. And part of the problem was that the, we have the specification, right? That, that we're all agreeing on. Um, and that's like what the behavior that should happen. And this is a constantly moving target. So in, in the entire year that we were building this thing, that, that specification had been rewritten from the ground up three times. And then on the last, you know, iteration, it had been changing a lot. And it's still changing until June, the end of June, um, being Q2, that's when they're going to freeze it finally. And that's going to be phase zero. So a lot of bugs come, come around because when we've been updating to the changes in the specification,

• 28:27Speaker 02

  maybe we updated one part, but forgot another, or there's like an off by one error. So people are just not behaving correctly. And it's all like things that you can't just unit test. Like this is one of those things we test with the whole system. And another is like, you have to test it at a scale. So one thing we found out yesterday is that our test network didn't work with more than 30 people. So

• 28:50Anthony Pompliano?

  That's not good.

• 28:51Speaker 02

  It's not good at all. And, and we were, we wanted to launch it yesterday and we, someone said, you know, hey, we shouldn't, maybe we should, um, you know, just deploy a bunch of clients out, out in the cloud and see like what the traffic looks like. And it was just miserable. Like we were, we had something where For every message that was received, it was going out twice.

• 29:10Anthony Pompliano?

  Mm hmm.

• 29:10Speaker 02

  And this is like a exponential effect. So when you receive it, you give it to your peer who sends it to everybody and you also did the same thing twice. So when you sent it out, you're gonna get it back twice from everybody else. So it was just, it's just one of those like oversights where, oh, when I'm running with my two computers in my house, it works fine. Or when I run on like with my friends or on the team, it's like working just fine. So, I don't know, that was, that was pretty crazy. Um, and part like part of fixing that, we actually, we'd been running the chain for, for one week. We got to a thousand slots, right? The slots are the new um time interval for block times, right? It's huge amount, a hundred thousand slots is a huge milestone for us.

• 29:52Speaker 02

  Uh, and when we were testing out some issues with syncing, we accidentally knocked all the nodes offline. So if this is the, the, the World War Three disaster scenario, where if every node in the world went offline, you have nobody to talk to to get the latest data.

• 30:10Anthony Pompliano?

  Mm hmm.

• 30:10Speaker 02

  So we basically had killed the chain entirely. We couldn't get it back online. But we managed to recover one of the disks and then hacked, hacked one node together to say, just, just take the disk data and, and, and, and this is the latest data. It's sort of like a, a social uh consensus here where like, we'll accept what this guy says because we have no choice. So, we did that, revived the chain. It's still going. It's been going eight days now. Um, we were able to launch that test network and hopefully we can get a lot of people trying it out today. Cause we, we, I mean, we wanna see a break because if it breaks now, that's amazing. Now we have a new bug we can fix. And that's gonna like make it better when it finally goes to, to mainnet.

• 30:52Anthony Pompliano?

  Yeah, well, part of it is just like, if it's gonna break, let's break it day eight, not in, you know, day eight hundred and forty seven when there's billions of dollars on it, right?

• 31:00Speaker 02

  Yeah, exactly. So the, the, the timeline for this is that where the community, we're hoping to launch to mainnet at the end of the year. So, so maybe.

• 31:09Anthony Pompliano?

  You're like seven, eight months to, to continue on the testnet.

• 31:12Speaker 02

  That's, yeah, that's what I'm thinking. So. The spec, as I just said, is that it will be frozen at the end of Q2, which means we have, we, in two quarters, we can get everything. We're hoping we can get everything together. Where we're working, clients can, different clients can talk to each other. So our Prism client from Prysmatic Labs, it only talks to other Prism clients right now, just for this first iteration. But that's not very useful because we want it to be multi-client. So that'll be a big hurdle. Uh, and then trying to get things to work on a, on a really big scale. So running at a hundred clients is, is great and all, but the system, I think the minimum amount to start is sixteen thousand validators.

• 31:54Speaker 02

  So we're gonna need to run a test net with a lot more resources.

• 31:59Anthony Pompliano?

  Uh huh. And that makes sense. And then so as you move forward, what, what are the challenges or things that you're worried about? Right? Is it that it actually does break? Or are there other things that, that you're worried about?

• 32:09Speaker 02

  Yeah. Um, I guess I'm, uh, not too worried because I think we're, we're in a pretty good spot. Um, if things are coming along really nicely, there's a lot of work on the research side. They, it, this bit like in terms of the process between getting new ideas in, from, from someone's mind into paper. So we can all work on it. The process and the review has been getting a lot better. Um, I think that what, maybe what we're missing is, um, some velocity. I think we had all, I took a little bit faster. On, on Prysmatic Labs, we want to grow the team out to, to ten or so people. So we're really hoping that in the next six months, we can,

• 32:54Speaker 02

  you know, get more resources to build out the client and also the support infrastructure to help like keep it healthy and alive and bugs, you know, out. Um, so that's kind of in my mind is just how are we gonna, how are we gonna get there? And how fast can we get there?

• 33:13Anthony Pompliano?

  Got it. That, that makes sense. And then walk me through the scenario. Let's say that, uh, you are successful, right? And you build this Ethereum 2.0, uh, proof of stake with sharding and, and it works, right? And kind of Ethereum is ready to scale to the masses, if you will. What happens to like the Ethereum 1.0? And, and kind of what are the ramifications of your success in that scenario?

• 33:35Speaker 02

  Uh, right. So the, the, the migration strategy. So, so in my mind, I think that the Ethereum 1.0 chain is, is not going to be very useful going forward.

• 33:46Anthony Pompliano?

  Okay.

• 33:46Speaker 02

  I think, I don't think it's gonna go away.

• 33:48Anthony Pompliano?

  Okay.

• 33:49Speaker 02

  Entirely. Some people might wanna use it and might live on in some form, but, but I think the majority of people, I think that the DAP developers, the major applications are gonna move to Ethereum 2 and

• 34:00Anthony Pompliano?

  Okay.

• 34:01Speaker 02

  In uh

• 34:01Anthony Pompliano?

  And does that make Ethereum 1, like Ethereum Classic today? Kind of where like it still exists, it's just not used by anybody really, or, you know, am I thinking about that incorrectly?

• 34:12Speaker 02

  Uh, I, it's a good analogy where it's, it'll still live there and as long as people are mining it and, and blocks are advancing, it's still a blockchain that, that you can interface with. So I don't know if it'll have any value. I don't, I can't really speculate on that, but, but the, but that's what I think is gonna happen. I don't think we're gonna shut it down. Like we, obviously we can't, we have no power of that. The powers and the miners, if the miners are still mining, it still goes. We can't just say it's, stop using it, right? So

• 34:38Anthony Pompliano?

  Mm-hmm.

• 34:39Speaker 02

  And that's the, that's the, the beauty of blockchain, right? We don't control it. It's awesome. So, so the strategy is that we, um, maybe, maybe this is one idea anyways. We take a snapshot of, let's say a block 10 million. This is hypothetical. Block 10 million, we're gonna, we're gonna snapshot what the Ethereum 1 chain looked like. Break the connection between the two and then spin up a new shard with that state data. So if you hadn't migrated already, you're just gonna be copied over in that sense. So that's the one way that it can come, come over and, and, and, and sort of transition from 1 to 2.

• 35:19Anthony Pompliano?

  Got it.

• 35:19Speaker 02

  Yeah.

• 35:19Anthony Pompliano?

  And when that migration happens, are there negative side effects to this? Like, is there anything that you foresee where you're like, look, we think that a negative, you know, aspect of this is A, and we're okay with that, but we do identify it. Or, or is there really none that pop out?

• 35:36Speaker 02

  Um, some of the negative feedback that I've heard is that there will be two

• 35:43Speaker 02

  Ethereum tokens at or yeah, two Ethereum coins at the same time.

• 35:47Anthony Pompliano?

  Oh, interesting.

• 35:47Speaker 02

  Because because it's a a one way migration, there's not this um backwards liquidity or whatever you want to call it. But I think that problem is solved in in arbitrage, because if you really wanted to get your tokens back on that chain, somebody out there is gonna be willing to to to do that for some small fee, like maybe let's say two percent, like if if if I wanted to go to Ethereum two point O, I can get two percent more tokens by just trading with this person.

• 36:15Anthony Pompliano?

  Yeah.

• 36:15Speaker 02

  Then I someone's gonna want to do that.

• 36:17Anthony Pompliano?

  Somebody will figure something out.

• 36:18Speaker 02

  Yeah, if if there's an arbitrage opportunity, it's gonna happen. I'm not too worried about it. The the benefit here is that we keep the system simple. Having a one way bridge is significantly simpler than a two way bridge. So that's why that's why we're doing it that way. But that's some of the feedback that I've heard are people concerned that that Ethereum one and Ethereum two are going to trade at completely different prices, which is a is a valid thought, but but uh I don't think it's gonna be that.

• 36:43Anthony Pompliano?

  It's gonna be weird if there's like ETH one and ETH two on exchanges, right?

• 36:47Speaker 02

  Yeah, yeah, that's that's the other thing. So yeah, I don't know. It's something to think about and we're thinking about it, but I don't think it's gonna be like dramatically different, maybe just like a little bit.

• 36:57Anthony Pompliano?

  Yeah, well, and would it be fair to get I'm thinking about this out loud here, but I guess as long as the technology works and it gets to a scalable situation, the idea that there are different tokens on exchange, like all of that is almost on the periphery of the blockchain itself works and people can then build on the platform.

• 37:15Speaker 02

  Right.

• 37:16Anthony Pompliano?

  Right. It's like that's almost like it it it would appear to me without thinking about it too much, that's the most important thing, right? And then everything else is kind of second order effects of if technology works, then we got a shock.

• 37:25Speaker 02

  Yeah, yeah, we're yeah, we're not building this with uh like

• 37:31Speaker 02

  the trading part in mind, like that's important part of the system, like if this has to have value for it to work. But we're designing something that's supposed to be scalable and trying to keep it simple, scalable and usable, and then the rest will come come with that.

• 37:45Anthony Pompliano?

  I think that makes sense.

• 37:46Speaker 02

  Yeah.

• 37:46Anthony Pompliano?

  Um, before I wrap up, I usually go rapid fire questions. Uh, what do you think is uh the most important company in crypto?

• 37:54Speaker 02

  Uh, most important company in crypto, um.

• 37:58Speaker 02

  That's a hard question. That's first rapid fire one. I don't know. I think I think um I'll go with consensus on this because they have just done a lot of a lot of work. They're trying experimenting in all areas. Uh, having some success and failures, but overall like having a big impact in the community. So that's I really respect the all the efforts they're doing. Everybody over there is doing a great job. Yep.

• 38:20Anthony Pompliano?

  Okay. What uh, what's the one regulation that you would change or improve if you could? Anything that sticks out?

• 38:26Speaker 02

  Hmm. Yeah, I I hear that it's hard for people to invest in stuff in the US. I think the regulation on like KYC, know your customer and anti money laundering laws are are difficult. So at least compared to the rest of the world. Uh, I would like to see that different, you know, because like people, this is not like. I'm not like buying into someone's uh, you know, crazy company. These are like small amounts and and crypto that that's going to have, couldn't have a big impact. So I'd like to see that to be more accessible to people because the ultimate goal is global adoption. And if you have to have

• 39:06Speaker 02

  already a million dollars in your pocket to to get in, it's not going to benefit everybody, right? So.

• 39:12Anthony Pompliano?

  Got it. What um, what do you think your most controversial thought in crypto is? Like, what do you believe where you think a lot of other people disagree with you on?

• 39:20Speaker 02

  Hmm. I don't know. I try to stay away from those. So I don't know.

• 39:24Anthony Pompliano?

  Well, it doesn't necessarily be a controversial thought in the sense of uh, it is toxic, right? But but just in the sense like, is there something that you think will work that a lot of other people don't think will work? Or maybe you think that everyone's really bullish on something and you're not?

• 39:39Speaker 02

  Um, most of my thoughts are on the technical design. So the group of people that disagree with me is kind of small, just because it's a small sample set, right?

• 39:50Anthony Pompliano?

  Yeah.

• 39:50Speaker 02

  But in terms of like our um, interop architecture between the clients we're building. So the clients are two pieces. It's a beacon chain node, which is like the the the piece that handles proof of stake assignments and managing everything. And then you have the validator client, which is actually doing the signing. And the way these talk to each other, the technologies we're using is a um sort of controversial debate. Everybody wants to do it a different way. Um, I'm pushing more for this schema based.

• 40:22Speaker 02

  A framework called Protocol Buffers, but other people want to do JSON, which is like web technology and it's not like that controversial. So maybe this is like a disappointing answer, but it's the best I got, you know?

• 40:35Anthony Pompliano?

  If it's important, right? People disagree, it could be controversial.

• 40:38Speaker 02

  You know, I'm a pretty agreeable guy, so this is the best I can do.

• 40:44Anthony Pompliano?

  That's all right. What's the most important book you've ever read?

• 40:48Speaker 02

  I, um, ah, it's a good question. So I read, uh, I guess like in my career wise, I was doing a lot of JavaScript stuff and there was this series called You Don't Know JS.

• 41:01Anthony Pompliano?

  You don't know JS?

• 41:02Speaker 02

  Yeah, or Don't Know JS, something like that. And it was, it was huge for me because like a lot of people do JavaScript and they're like, they know how to do like get buttons to do things and just animate and do this stuff, but like to actually understand like scoping and how things work on the internal layer, uh, stuff they don't really teach you in school, especially me when dropping out. Like that was really huge, huge for me. So it, it kind of opened the doorway for me to like get, I, I actually, it was like the first time I got interested in how things work under the hood. So I was like always interested in building things, but like, how does it actually work? That's when I started getting, getting really into it in terms of my computer science career.

• 41:39Anthony Pompliano?

  That's awesome. Um, I, I probably will not read it just because I feel like I will, it'll go over my head too quickly, but, but it's, uh, it's fascinating to hear. Uh, not only is it an important book, you read it at the right time.

• 41:50Speaker 02

  Right.

• 41:51Anthony Pompliano?

  Right. So the timing is almost as important as the information, uh, which is pretty cool. All right. So to finish up, no matter, let you ask me one question, but first we talk about aliens. Uh, believer, non-believer, think they're real?

• 42:04Speaker 02

  Um, I think they're real. I, okay. I, I, I, it's, it's hard in, in this vast universe to think that we're the only, only people out there that's, uh, Intelligent species, so I'm I'm going to believe there are aliens.

• 42:18Anthony Pompliano?

  So I I'm with you. I think that just from a probability standpoint, it's unlikely that we are the only species. And and as many people on the podcast have said, it would be selfish of us to think that.

• 42:28Speaker 02

  Yeah.

• 42:28Anthony Pompliano?

  Uh, there is a group of people who are now trying um very hard to convince me that the aliens are already here. Right? And that uh maybe we just don't know where they are or all of that. So it's been it's been fascinating to hear uh people's perspectives on uh how deep their alien beliefs go.

• 42:45Speaker 02

  Yeah.

• 42:46Anthony Pompliano?

  Right? What uh what one question you have for me to finish up?

• 42:48Speaker 02

  Um, are you going to be an Ethereum two point O staker?

• 42:52Anthony Pompliano?

  Oh, interesting.

• 42:53Speaker 02

  You gonna validate for us?

• 42:54Anthony Pompliano?

  Um, I don't know, is is the short answer.

• 42:58Speaker 02

  Sure.

• 42:58Anthony Pompliano?

  Uh, I think that um something, you know, my partners and I have uh, we start out with very kind of rigid thesis, right? Which is um Bitcoin has a higher probability of being the global reserve currency than people give it credit for, right? So not necessarily even saying that it's a huge number, it's just it's a higher percentage probability than people think, right? So like, that's one piece. Two is that um a lot of what is happening right now is about automation. And this idea that every stock, bond, currency, commodity, so everything of of value is going to be digital in the future, right? Most likely cryptographically secure, most likely on a blockchain, all that, but doesn't necessarily have to be, right? So just can we use software to automate a lot of processes, et cetera?

• 43:41Anthony Pompliano?

  Obviously, with the smart contract platform, um that's that's a huge component of it. Um, we cut our teeth mining.

• 43:48Speaker 02

  Right.

• 43:49Anthony Pompliano?

  So actually mining Bitcoin and Ethereum and going through, you know, the GPUs burning out, right? And dealing with ASICs and all stuff. Um, and and so we we've dabbled uh personally in some of the staking stuff. Um, we find it fascinating. I think that uh what I what I am personally looking for is uh where is it working at scale? Right? And so I think uh Ethereum two is a great opportunity for that to happen. So so if the technology works and and um you see this. I don't see why we wouldn't be open to doing it, right? At the same time, part of this also is the idea of staking is very new to pretty much anyone in the financial world, right?

• 44:34Anthony Pompliano?

  This idea of like, hey, you're going to make an economic investment and then you are going to put that at risk in order to be part of running this net like, that's not a thing that they're used to thinking about. And so we've joked around about like, does that mean it's like the new fixed income type investment? Or, you know, how do you basically describe it to our institutional investors that they're not technical, right? Frankly, they think of Bitcoin and then they think of blockchain, right? They don't even think of Ethereum and all this other stuff. So like, we're still kind of, you know, educating and kind of trying to chip away at this idea that there's other things other than just Bitcoin.

• 45:14Anthony Pompliano?

  And so I think that, you know, it's a thing where maybe at the end of this year, I can make this promise. You come back, we'll do this again. And then maybe we can even if it's up and running, then maybe we'll, we'll stake some stuff kind of live and see how it goes.

• 45:30Speaker 02

  Cool. That's, that would be great.

• 45:31Anthony Pompliano?

  That'd be awesome.

• 45:32Speaker 02

  Yeah.

• 45:33Anthony Pompliano?

  Look, it's, it's pretty cool to, to kind of hear how, how is this going? And I think, you know, just our conversation today, there's aspects of this like Ethereum too, that I think a lot of people don't understand. And then two, part of what to me is so personally intriguing about this is like the amount of support that you're getting from different stakeholders in the Ethereum community, right? It's not just you guys saying, hey, we're going to go do this and like, I don't care if anybody likes it. You're actually having people who are saying, look, I'll, I'll provide resources, you know, financially. I want to be a part of this. I want to see this be successful. And I think that is, that's really unique and hopefully ends up, you know, accelerating your guys' success.

• 46:11Speaker 02

  Yeah, it's huge. And one thing I forgot to mention is a lot of the contributors to our funding are actual just individuals. So we've had people five dollars a month.

• 46:20Anthony Pompliano?

  That's awesome.

• 46:20Speaker 02

  It's, it's been huge through Gitcoin, we've received, I don't know, $10,000 from, from the community. So that was super cool.

• 46:26Anthony Pompliano?

  That is absolutely cool. And crypto is a wild place, right? You can never say never in crypto. And then I'm always blown away by just the generosity of people with their time, with their resources, etc. So it's cool to hear that you guys are benefiting from that as well. Awesome, man. Well, listen, I really appreciate you doing this. It's been very informative for me. Hopefully, people listening got a lot of benefit out of this. And then, end of the year, you come back and let's see where things are.

• 46:54Speaker 02

  Sounds good. Thanks for having me.

• 46:55Anthony Pompliano?

  All right, thanks. Hey, everyone. Pop here. If you liked this episode of Off the Chain and want to help us take crypto to the top of the Apple, Spotify, and other podcast charts, please do us a favor and rate, review, and subscribe. To review, simply go to the Off the Chain homepage, scroll down until you see the five blank stars. Taking 15 seconds to fill those stars in and leave a quick review goes a long way in helping us take the entire crypto ecosystem to the top of the charts. I appreciate you listening and see you next time on Off the Chain.