The Exchanges

Q You also mentioned on the product side, as you were learning about product, that some of the things that you did aren't what people would have expected, or what the customer wanted aren't what people would expect. What's an example of that?

A So basic stuff at that time, people just wanted higher limits. They wanted to buy more Bitcoin. And people wanted fast delivery. So one of the things about the Coinbase product back then was bank transfers in the United States between a US bank to a US bank using the ACH system are comically slow. It takes about three days to clear. And this is based on a check clearing system that's been adapted to computers. It's like a joke how bad the system is. What it means is that we couldn't deliver people's Bitcoin until that cleared safely, but then we realized that there were all these little optimizations you could do, so we created sort of an algorithm that would detect the probability that somebody had insufficient funds, and we would deliver the Bitcoin faster based on that algorithm. We would also algorithmically place someone into a tier of fraud risk, so the thing preventing Very high limits was fraud. The bad guys hurt the good guys ability to buy. So we created a tier system where we would create an algorithm. This was a bit later. We obviously had more engineering power after all that manual review, and we put people into a certain sort of bucket and the bucket would allow them to buy or sell more or fewer amounts of Bitcoin. So all those sorts of things, they're small optimizations around the edges. They don't feel like a new product. You're just refining that core product…

AI assessment note: “basic stuff at that time, people just wanted higher limits. They wanted to buy more”

Q So let's turn to 2016 and you started talking about a different kind of venture fund or different kind of investment fund, crypto native. What was your original sort of investment thesis going in?

A I knew all of these engineers and researchers that were working on incredibly cool projects. But the end thing was like a crypto protocol. This was on Ethereum. This was projects like MakerDAO. And outside of Ethereum, these were projects like Tezos, projects like Cosmos. Polkadot was being sort of conceptualized around that time. And these aren't businesses in the traditional sense. There's not really an equity entity like a Bitcoin Inc. that sits behind the system and extracts revenues from it. It's a peer-to-peer protocol system that's being developed. These engineers and developers, I thought that a lot of these concepts were really brilliant. I understood why they could be important. I had spent years looking at this stuff and going very deep on crypto tech, but there was no funding model. There was not one person in the world that could wrap their head around what it was and understand that to fund this, you can't think about it like traditional revenue business. You have to just buy a digital asset. So it was a, in hindsight, a massive vacuum in the market. So I sort of started Polychain and was doing deals immediately. I think the first deal I did three days after I launched the fund was to fund Tezos. Shortly after that was Cosmos. And this was also during the beginning of the ICO or crowdfunding wave. And I was a massive participant in those crowd funds throughout lat…

AI assessment note: “to fund this, you can't think about it like traditional revenue business. You have to just buy a digital asset.”

Q Which two people have had the biggest impact on your professional life?

A Definitely Brian Armstrong of Coinbase. I just think he is so insanely smart and calm and just one foot in front of the other for like 10 years just grinding to build the financial institution of the future. He taught me a lot at Coinbase. I just think it's insane the story of Coinbase and just having been really a part of it from the super, super early days is just something really special. I don't think I'll ever have that chance again. The second one, I would probably say the philosopher David Pierce. He's a transhumanist philosopher and just talks a lot about motivation and hedonism and happiness in sort of a philosophical way, and a lot of what he says appeals to me. I think he's a hundred or 200 years from now going to be remembered as way ahead of his time, but today almost thought crime from a cultural perspective.

AI assessment note: “Definitely Brian Armstrong of Coinbase... The second one, I would probably say the philosopher David Pierce.”

Q I'd love to start going back 10 years or more, and how did you first get interested in the whole crypto world?

A It was the summer of 2011, and I read a Gawker article about Silk Road, and it mentioned Bitcoin as being the catalyzing technology that made that possible. So I started reading about Bitcoin, and Right away really was thrown into a bit of an obsession with the entire concept presented, which was this, at that time, not really an industry, but more like just an open source piece of software you could download and a community of users on forums. And there were a couple very small scale businesses built around buying and selling Bitcoin, but at that time, it was really more just an open source software project. This concept of a sort of non-state money that was algorithmically determined monetary policy and a decentralized user base that would determine what sort of software to run in sort of a decentralized consensus. It was all just extremely fascinating to me, and I felt like I stumbled on this amazing, amazing new technology, and I really did have a feeling that the world didn't know about this at all. So I really got quite into it that summer, and I was going to my senior year in college and had to decide a thesis topic and decided to write it on cryptocurrency, which only took me further into the rabbit hole.

AI assessment note: “It was the summer of 2011, and I read a Gawker article about Silk Road”

Q What was your path from your couple of years there to starting Polychain?

A Once I joined Coinbase, I got a really intense, just hardcore, rapid lesson in scaling hypergrowth company. I started as just frontline customer support, and for basically anybody who wrote a support ticket in the year, 2013, very good odds that I was replying to you personally. Every time. A year later, I had a team of over 40 people that were reporting to me, managing effectively all of customer support, of course, and then our anti-fraud process and operations. To an extent, we got a chief compliance officer in the latter part of that year, but I was also in charge of looking at a lot of the compliance stuff. I mean, this was very early. The company was like five people, so somebody had to do it. I was not qualified to do these sorts of things. But I'm not sure anybody was qualified to really do anything on paper. It was just an insane lesson in how to scale a company really, really rapidly. I also learned a lot about what it meant to think about product. And because I was interacting with every single customer problem, I learned very intimately the ins and outs of the early Coinbase product and how to make it better, basically, and what people cared about. And it's not always obvious what you think people care about. But if you talk to users, you can always figure it out. I think that experience got me even more interested in entrepreneurship and just the idea of building s…

AI assessment note: “that experience got me even more interested in entrepreneurship and just the idea of building”

Q Over the last couple of years, evolution of, as you said, equity in businesses, tokens, different forms of investing in the space. How did you think about structuring your investment strategy so that you could participate however you wanted to in the space?

A So I designed the fund from a structural perspective to match what I wanted to do in terms of investing in that fund exactly. So instead of doing a traditional venture drawdown structure, I structured it as an evergreen, more like, quote, hedge fund. Again, we're long only venture style approach, so not exactly hedging, but I designed the whole fund to be oriented around this type of investing. When it came to hiring people, I hired people that were oriented around the specific type of investing. So from the beginning of I never was targeting people from Sand Hill Road with a traditional VC approach, nor was I targeting anybody from Wall Street. I was targeting basically other people like me that were grassroots crypto people that understood the technology, understood the market, had connections in the space, just designed everything from the beginning with that in mind. And then over time, we got really serious about on-chain participation. Now we've leaned into that super deeply. So we've always had engineers, More on the practical engineering side, people implementing staking, DeFi participation, on-chain governance participation, all that kind of work, which is just part of being a crypto investor. When you hold Google stock, it's not an asset that gets sort of activated and used in a system, whereas in a crypto asset, it really is a useful asset a lot of the time.

AI assessment note: “instead of doing a traditional venture drawdown structure, I structured it as an evergreen”

Q What was it that resonated so strongly with you?

A This was in the wake of the great financial crisis of 2008. I was studying sociology and asking a lot of petty questions that you have a lot of time to think about in college, and a lot of these questions were things like, how does private property work? What's the relationship between private property and the state, and what is money, and who makes these decisions about things like The 2008 crisis management bailouts and things like that. So I think I was a bit jaded with large scale institutions of all types in general, the financial sector and the government, especially the intersection there. So when I started reading about cryptocurrency, it just felt like this incredibly powerful vortex where once you learn about it, you can't Unlearn about it. You sort of have this alternative to the state issued money. That's sort of a better product for the end user. And at the time I wasn't really thinking very much about decentralized applications or web three or things like that that we talk about today. I was really just thinking more about basically like an electronic gold sort of thing. I just felt like this is going to be the unified money of the new internet based and highly globalized economy. And it wasn't going to be tamper proof. There was going to be no manipulation by even the people who created it. All of it was just extremely fascinating to me. It led me to learn about …

AI assessment note: “You sort of have this alternative to the state issued money.”

Q What are the categories that you like today?

A This is a common one that people are talking about, but I really do think that this concept of attaching a proof of work type system to different types of behaviors This sort of consensus building the way we've done it in the past. So the way Bitcoin hashing leads to a block reward, it's basically what other types of behaviors could you coordinate people around using like a protocol block reward type system. So a lot of times people call this play to earn, like play a video game, earn the block, earn the reward. But really, I sort of view it more broadly as is kind of like the future of work and labor. And the relationship between capital and labor. And I think it can be applied to a large swath of categories of labor over time. And that is a very big idea to me, which is basically, we could not just program something like Bitcoin mining, but we could program the whole industry potentially around these work and reward systems. So I do think that Video games are sort of the first sandbox for that because it's a very neatly defined type of work because you're basically creating fake parameters. It's like a fake setup followed by a fake reward system in that artificial setup. But I think that that type of design mechanism can be applied to more and more types of basically labor done on computers. So if your job is on a computer, I think that there's some chance that you can replac…

AI assessment note: “I sort of view it more broadly as is kind of like the future of work”

Q So beyond this play to earn block reward category, what are some of your other favorite big idea categories going forward?

A I don't think that this is a huge novel insight on my part, but this idea of these centralized platforms that have this Very tumultuous relationship with their users and creators. I just think this is one of the big broken pieces of the internet today, is that YouTubers who make their entire career on YouTube don't like YouTube. They have a bad relationship with YouTube. People on Instagram who make their whole career posting on Instagram don't like Instagram. It's this very hostage relationship. It's like a Stockholm syndrome. I rely on this platform But I do not like it, and I am held captive by it. There's no alternative, there's no forking off, there's no mechanism to break the rules, because even if my followers or users want me to. So that just feels like it's a really broken piece of the internet, and it's a little blurrier to me exactly what it looks like and exactly what role crypto plays in this, but it does feel like if you can create an alternative distribution and alternative monetization mechanism, And I do think part of it may be the disentangling of those two things. So today you monetize an Instagram and you distribute through Instagram. It's not like you distribute on Instagram, but you monetize with NFTs. The idea that you could disentangle distribution and monetization potentially and refine each of them separately. I think that there is something there and …

AI assessment note: “this idea of these centralized platforms that have this Very tumultuous relationship with their users”

Q Use might be an easier Question to answer as an investor than value accretion, and I'd love to get your thoughts on, in that great example you used with Bitcoin versus Ethereum, but across the different chains, where you think conceptually value will accrete?

A I think that every blockchain in the long term needs to have a tightly coupled mechanistic value accrual mechanism between use of the system And the price of the asset. So this is, in the Ethereum context, you actually burn Ether the more people use the system. I just think every blockchain in the long term is going to need that type of tight value accrual mechanism in order to appropriately scale. Because the types of values we're talking about, if these systems are going to host the global economic activity of the world, They're going to be insanely valuable. And the scale here is beyond what I think most people really comprehend. Even crypto bulls, I don't think really realize where we're actually going. And once we're there, this isn't going to be a buy your favorite coin. This is like the global capital of the world allocating their capital to these blockchain assets. In that world, you need to be able to predict What value accrual looks like based on use, and you need to be able to model use and basically make asset allocations based on that. So the historical way we've done this is basically, I see a lot of users, so speculators are going to speculate. And I think that has crudely worked, but we're going to need a system that the use actually drives value to the underlying asset. And that's going to be, I think, a critical component of every blockchain that makes it.

AI assessment note: “every blockchain in the long term needs to have a tightly coupled mechanistic value accrual mechanism”

Q And how about these bridges to create the interoperability across layer one blockchains or into layer two?

A These one-off bespoke bridges, I don't think are the way. This is, okay, we want to move from Bitcoin to ETH. We want to move from Ethereum to Avalanche. I think it is bridge protocols that are based on economic security in the same way that Bitcoin or Ethereum itself is based on economic security. You can't just exploit three servers and take down Bitcoin. There's this incentive system that you have to exploit, and it's way harder and way costlier to exploit that incentive system. So I think the future of bridges is bridge protocols based on economic security instead of basically multi-sig or server security, rather than this bespoke one-to-one style of bridge that is more popular today. I just think that that's the early system. I also just broadly think that where we're going architecturally is going to be highly modular.

AI assessment note: “I think the future of bridges is bridge protocols based on economic security”

Q And how about DeFi? Where are you excited about opportunities in DeFi?

A I think DeFi is just chipping away at the global financial system. I really do view it as where every financial transaction in the world will take place. It's going to be a long path, but I think that people are just continuing to chip away at new products and new concepts. I think one of the very interesting categories here, and an important one, is the coupling of offline assets to DeFi protocols. So being able to take a plot of land and use it as collateral to borrow in a pool, for example. And that is mostly legal technology. We can't just write software that'll make that perfect. It's gonna require a lot of entrepreneurs operating in a legally complex area to enable that, but if we can couple offline collateral and assets with DeFi, it's at least a 10 X, probably closer to a hundred X unlock from where DeFi is today.

AI assessment note: “one of the very interesting categories here... is the coupling of offline assets to DeFi protocols.”

Q So what is getting forged in this bear market?

A I think a lot of amazing ideas. Really were created throughout, call it late-twenty-twenty and all of twenty-twenty-one, basically, but I don't think they were really tested in a sort of, does actually anyone care about this type of way? Including stuff like Play to Earn. There's very powerful feedback loops both directions with Play to Earn. If the price is going up, there's more users. If the price is going down, users leave. And it's like, how many of these users were here because they like the product and not just because of Some sort of speculative mania. I just think that we need to really test all those ideas that came out of this previous cycle. 2017, it wasn't that different. I mean, we had this concept of DeFi. The word DeFi didn't exist yet, but we were seed investors in zero X, which was the first order book system embedded inside Ethereum. We were seed investors in compound finance. In DYDX, and a lot of those deals happened around that peak mania, late 2017, early 20 18, but we had to sort of test them and really refine them, and we didn't fully get them refined until 2020 for that DeFi category, and that's when it really blew up and people realized, wow, this works, and we can actually embed the future global financial system inside a blockchain, asterisk, once it scales. We just need to test it out. A lot of the ideas that I'm talking about, conceptually, they'r…

AI assessment note: “we need to get pragmatic and real about the implementation”

Q What's been your biggest mistake and what did you learn from it?

A The biggest mistake I've made is ever really like doubting myself. I think you just have to 100% invest in yourself and just fully commit all in 100% to what you believe in and what you're doing. And a million people will tell you you are wrong. And you need to actually not care at all what they think. And you just need to say, I don't care. I think this is right. I'm doing it. I think I have mostly done that actually, but even the like five percent of the time where I'm like, man, so many people think that what I'm doing is a poor decision. Should I really go through with it? The answer is just, if you think the answer is yes, the answer is yes.

AI assessment note: “The biggest mistake I've made is ever really like doubting myself.”