Josh Bowen is co-founder and CEO of Astria, a decentralized sequencing layer for Celestia rollups. He was previously a Software Engineer at Celestia and Google.
[00:00:02] Hello, everybody, and welcome to the Crypto Hipster Podcast. This is your host, Jamil Hasan, where I bring you all the Crypto Hipster, of course, where I bring you founders, entrepreneurs, executives, thought leaders, amazing people all over the world of crypto and blockchain. And today I have another amazing guest. He is the CEO of Astria. His name is Josh Bowen. Josh, welcome to the show. Thanks so much for having me on.
[00:00:28] You're very welcome. Very welcome. I don't usually stumble upon my own name. Usually it's my guest's name. So let's kick things off. I'll ask you the first question. What is your background and is it a logical background for what you're doing now? Yeah, so my background, I guess there's maybe two components of my background that kind of tie into kind of what I'm doing at Astria and what I've been doing kind of in crypto.
[00:00:51] So on one hand, I found crypto and Bitcoin specifically quite early. I kind of grew up in a very kind of I liked computers. I liked electronics. I liked kind of building my own computers. And I came from that kind of building your own desktop and kind of overclocking computers. Through that, I found folding at home was a thing.
[00:01:10] And Bitcoin mining was found quite early in like 2009 by that community. And so I found about Bitcoin quite early and did a little bit of that, you know, way, way, way back when I was in high school. You know, unfortunately, you know, it's the classic, you know, I didn't keep that laptop or whatever. But, you know, so that got me very, very kind of early to Bitcoin.
[00:01:30] And I didn't work in that kind of professionally. And then I went through a relatively kind of like standard path of finish high school, go to undergrad for electrical computer engineering with more of a focus on, you know, call it like mechatronics and whatnot. But, you know, a little bit of software and whatnot. I ended up going kind of after school into software engineering. And I eventually ended up at Google and I worked under Google Cloud for about four years or so.
[00:01:55] And so that ties in a bit to kind of what we're doing in Austria in that, you know, I did spend a significant amount of my kind of professional career before I was in crypto doing broadly kind of cloud based distributed systems engineering. And so that kind of, you know, ties in from, OK, you know, we looked at, you know, crypto and blockchains decentralized systems. And, you know, from a classical software engineering background, it's they're different, but they are kind of just like you have a bunch of, you know, different nodes in a network.
[00:02:24] You want them to communicate with each other. You're optimizing network bandwidth, latency, et cetera. You're coming to consensus over it. And so that kind of, you know, carried over quite well for me. Awesome. Awesome. So we're going to get into Astria today. I want to find out what Astria is all about, including what decentralized sequencing it's about, too. Yeah, great. So Astria has been around. We started the project in mid 2022.
[00:02:51] It came out of my time as one of the earlier employees over at Celestia. And the original thing we were kind of focusing on was at the time called, you know, building like a settlement layer or a decentralized settlement layer. And the reasoning for that is that if you're familiar with Celestia, it's a relatively intentionally kind of minimalist blockchain. It just kind of stores the blobs, right? It propagates the blobs. It provides data availability sampling for them. But it intentionally does not do things that like Ethereum does where you can actually settle.
[00:03:18] And when we say settle, we're talking about generally like monetary settlement or really bridging, right? I can have like one coin. I can put it in some kind of escrow account, which we might call like a bridge account. And then we can kind of bridge it to another chain. That's not something that's kind of supported on Celestia right now. And so that's how we kind of started. We're saying, okay, we want to build something that is kind of Celestia native in that we didn't want Celestia to just be used as like an alternative DA for Ethereum with rollups that were then settling on Ethereum.
[00:03:46] And so that's how kind of Astria came about. How we went from there into decentralized sequencing was essentially when we looked at the network architecture. You know, we had like Celestia at like the bottom of the stack and we built the settlement layer on top. And then the intention was we were going to build rollups then on top of the settlement layer, which itself was on top of Celestia. And going through that kind of stack, you know, the expected way to build a rollup was to use, you know, a centralized sequencer. Again, this was mid-2022.
[00:04:13] You know, we had, you know, Optimism launch their mainnet in December 2021. Arbitrum launched before that in August 2021. So we were relatively recent into having kind of rollups live, right, using centralized sequencers. But when we actually looked at the kind of architecture diagram, the thing that stood out was just it felt excessive to have this like three-layer design.
[00:04:32] And so I started thinking like, okay, if we have this layer that is a settlement layer that we need for the bridging, can we also have that settlement layer act as the sequencing layer for the rollups themselves? And so that was kind of how we started approaching that. And also kind of important from like a time thing, right, you know, now it's like, okay, we're 2025, right? We've been running decentralized sequencers for, you know, not quite four years, but almost, right?
[00:04:57] And so, you know, when we started, we were, you know, less than a year into running decentralized sequencers. And so the promises of saying, oh, we're going to do progressive decentralization, we're going to decentralize the sequencers. You know, it was much more reasonable to kind of take those at face value and say, yeah, we just haven't gotten to that, but we're going to get it to it quite soon. And so kind of as part of that, it's like the past dependency, we said, yeah, we're definitely going to decentralize the sequencers.
[00:05:21] You know, the point of crypto to a large degree is the decentralization or at least the, you know, the censorship resistance, right, that is kind of enabled by being decentralized. And so that's kind of what we focus on. We have this one layer that is both, you know, a settlement layer, and that's not a term that's used very much anymore, right? And it is a sequencing layer. And then if we use the kind of more modern kind of language of the kind of technical kind of crypto terminology, the rollups on Asteroid would be considered based rollups of the sequencing layer.
[00:05:50] And that they themselves do not have their own block time. They fundamentally are just kind of pulling data out of the sequencing layer. So you can have multiple rollups. Each of them pulls a different subset of data at the sequencing layer. And then they actually use that. And essentially they execute that data where the sequencing layer is just kind of having it as, you know, we don't call it blobs, but it is like raw data or serialized data. And then essentially executed through one or more kind of state machines of a rollup.
[00:06:20] That's a word I've heard over the past few weeks. I heard the word blobs like five times. And I hadn't heard the word blobs for a year. So I'm like, okay. I have a follow up, you know. A lot of people build on Ethereum. They build on Solana. I haven't heard too many people building on Celestia. So I know it's had no problems in the chain. It's from what I understand, it's very sound, solid blockchain. What made you choose Celestia over everything else?
[00:06:51] So I chose Celestia, I guess going back. So I joined crypto full time in January 2021. The first place I worked was at The Graph. Specifically like Edge and Node was one of their development corporations, right? And I found Celestia because what I was doing at The Graph as kind of like a research project there was trying to figure out like The Graph indexes on chain data. They wanted to know, can The Graph also index off chain data? And the relevant bit was like, you know, me coming from like an engineering background, but not necessarily like crypto native in the technical sense I needed to understand.
[00:07:20] What does it actually mean for data to be on chain? Like what is the actual kind of mechanical distinction that you get for on chain data? And that led me to what we call the data availability problem, right? And it is the idea of like, what is data availability? Fundamentally, it is, you know, it's not this like, oh, it's not like a guarantee of storage. It's not saying someone is going to keep that data for an amount of time.
[00:07:41] Fundamentally, what data availability is, is it is saying there is an agreement via consensus among a certain number of nodes that a certain, you know, bit of data, blob of data, whatever, was made available to at least the entities at this timestamp. And the timestamp is stagnated by a block height, right? So that is kind of how I came about Celestia. And then why I joined Celestia was, you know, a little bit of it was kind of, it was very like past dependent. So, you know, I left the graph and I was looking to join, you know, kind of a new team. I took some time off.
[00:08:11] And really, I was between two teams. I was between, you know, Solana and Celestia. And so I thought of those as, in my view, kind of the two valid ways to scale blockchains. And the two ways I think about it is you have Solana just saying, we're going to do big blocks, right? And we're just going to bite that bullet and we're going to optimize the network. We're going to say, yeah, the validers have to be pretty large. They have to have a high throughput. They have to do a bunch of execution, right? But we're going to bite that bullet. We're going to have a network of, you know, whatever they have now, 1,200 validator nodes. And they're going to have large blocks.
[00:08:39] And then the way I thought of Celestia was, I think about it as like fractional block storage. What I mean by that is this data availability sampling and light node kind of system. What it lets you do is instead of having essentially this like relatively large kind of gap between, say, a light client, which is storing basically no data, it's just checking headers, and like a full node, which is storing all of the data of the overall ledger, right? Which obviously gets quite large, right? You can prune and whatnot, but it becomes quite a large amount of data. You have these light nodes, which sit in the middle, right?
[00:09:08] And through the kind of erasure coding data availability sampling, you can have a same amount of data for the network as a whole, for the ledger as a whole. And then individual light nodes can store a fractional amount of that data. And then third parties can query, you know, light node to do the sampling, right? To get guarantees that all of that data is actually present. But each of those nodes can only store a fraction of it, right? And so those were the kind of two ideas. And one of the primary reasons really I joined Celestia was because of a phasing of the company.
[00:09:38] You know, at this time, Celestia was about 10 people, whereas Solana was much larger. You know, this would have been November 2021. Solana was trading at about, you know, 253. And essentially I said, look, you know, if I go try to get a job like Solana Labs or whatever, right? You know, I'm going to go get a grant that is going to be priced at like 253 or whatever. And, you know, I'm still like a huge, you know, kind of like Solana bull. But like, it was very obvious to me that I'm like, I don't want to get like a four-year grant or whatever, you know, from, you know, an equity token or whatever thing at that kind of high price.
[00:10:07] And so I chose to join Celestia, which was more at the start of phase there. And then Astria has largely been an extension of my kind of distribution. And essentially kind of, you know, playing out the commitment to a degree where I was saying, I fundamentally think, you know, Celestia focusing just on, you know, this Blobs, right? And just this high throughput DA. What I was worried about is if it was only used as like alternative DA for Ethereum,
[00:10:32] then fundamentally the like progression or the speed or like the kind of like quality of projects that could kind of be built on it, like how scalable they could be, would be kind of constrained by Ethereum speed. And that was something I wasn't really interested in. So the effort of Astria was like, what can we actually do kind of on Celestia without having Ethereum kind of in the loop so that we're able to kind of move more independently? And the scope has obviously crept a bit on that, right? Again, you know, I started as, you know, the project called Sevmos when I was at Celestia,
[00:11:02] which is now kind of, you know, the framework there is now called RollKit. And then we built Settlement Layer. And the Settlement Layer is also a sequencing layer. Now we have our own roll-ups, right? And so the scope has crept. But it's all been in this kind of service of we want to kind of take Celestia and the exploration of that ecosystem as far as we can, kind of, you know, without being dependent on some other existing ecosystem, which may be more conservative. Makes a lot of sense to me, especially with your engineering background.
[00:11:30] So I'm seeing how you answer it. It's a lot of engineering focus. So pretty cool. So I want to find the history. You shared a little bit about this, but the history of sequencing, what has changed over the landscape, and then is there still a need for it, for sequencers? Yeah. So the kind of history of sequencing, and I'm going to get my dates, like, a little bit off, right? But when these were originally designed, or the original design of, like, a roll-up,
[00:11:57] you know, as proposed by John Adler, who's, like, I don't know what his actual title is, but, like, head of research of, like, Celestia or whatever. You know, he's the kind of guy noted for, like, inventing, like, roll-ups. You know, the idea was that they were going to be what we now term based roll-ups, where they would have a โ they would just pull data out of the base layer for โ so in the original design, right, it would be, like, Ethereum. You know, it could be Celestia. It could be Astrid at this point. And then they would execute that. And so the data would still be stored on the base layer,
[00:12:24] but the actual kind of executed state and the actual ledger would be the thing that is on the roll-up, and then you essentially have this bridge mechanism, right, which is what allows you to kind of safely move between that, right? What people realize is from at least, like, kind of, like, a competitive market positioning, right, there were just some limitations through the kind of base roll-up design. So, you know, Ethereum's kind of issues, you know, are around โ you know, it has 12-second block times, right? And those are rather slow relative to kind of, you know, kind of competitive standard of the industry right now.
[00:12:55] As well as it was still relatively expensive, right? This was long before flops. And even if you just had to submit call data, you still โ it was relatively expensive to submit this data to East L1. And so sequencers came about to kind of solve, you know, both of these problems, where essentially the sequencer is, you know, an off-chain entity from the perspective of the L1, which receives transactions, right? And then, you know, it orders them and it produces a block. Very specifically, it produces a block at a faster time rate or a faster block time
[00:13:24] than the L1. So, you know, East runs at a 12-second block time. You know, something like Arbitrum runs at a 250 millisecond block time. You know, Optimism Rolls run at about a 2-second block time, right? And then it makes these blocks and then it batches them, right, into, you know, now it would use a blob. Previously, it would just be like a batch of call data. And it posts that to the L1. And essentially what you're getting is you're able to provide what we might now call pre-confirmations, right? They weren't calling it, you know, a sequence or pre-confirmation or whatever, right?
[00:13:51] And that means that a user's fundamental kind of like user experience is better, right? They submit it and they get an ACK back from the sequencer. They're making a trust assumption on that sequencer to submit that data. So there is some like, you know, distinction in like the trust model, but they run much faster. And then you also get to amortize the cost of posting data where, you know, in just kind of data posting generally, very often you'll have like a static amount of metadata, right? Where it's saying, okay, I have my actual data I want to post. And then I have, you know, whether it's a header or whatever, right?
[00:14:20] And that's just a static amount. If you can say, okay, I'm going to post, you know, whatever, five or six blocks together, right? You only have to post the header once and then you get the data. And so you get to amortize that cost kind of thing. And so that was like, you know, kind of one of the reasons for the sequences, right, was this user experience and also this kind of cost amortization, which is carried through into, you know, blobs significantly reducing the cost. And there's a bunch of drama in like Eastland about, you know, the economic kind of viability of like blobs from the L1 layer versus the L2, right?
[00:14:47] And to the point of like, you know, is there still a need? I think if you're in an Ethereum landscape, yes. And I've argued with various people back and forth on the kind of, you know, based roll-up kind of designs where if you're familiar with kind of the MEV supply chain of Ethereum or your proposal builder separation, there's some designs nowadays that would say we have based roll-ups. And then the base roll-ups still didn't necessarily have to run at the block time of the L1 because they can use the PBS, you know,
[00:15:16] system as essentially a mechanism of pre-comps. It's my perspective that that is structurally the same thing as having a sequencer. It is just a slightly different kind of way in which you choose who your sequencer is through maybe an auction mechanism rather than having a static sequencer. But from my view, at least from kind of what we see as chains having usage, I do think sequencer, they're kind of important and necessary because, you know, the bean chain proposal that Justin Drake of the EF put out in,
[00:15:44] you know, November, you know, that's, you know, relatively far out and that's only four-second block times, right? So even from, you know, again, Arbitrum running at 250 milliseconds, you know, Optimism running at two seconds, you know, and they're primarily competing against Solana, which is running at about 400 millisecond block times, right? So in that vein of saying, even if you were post-beam chain and you run a base rollup that's pulling the blocks out of just L1, you're still kind of having a much slower, you know, lower or higher latency user experience.
[00:16:12] And to that sense, I do think sequencer is kind of, you know, necessary. Again, there's trade-offs like a trust assumption, but the reality is we don't see users, you know, we wouldn't expect to see users moving towards, you know, a rollup that has much slower block times, right? If you look at like Tyco's kind of usage compared to, you know, the top three rollups right now, which are like base Arbitrum and Optimism, right? It doesn't see a lot of usage. And so the Tyco guys are working on Gwyneth, which is their based rollup design using kind of pre-coms.
[00:16:40] But again, my view is structurally that is the same thing as a sequencer. You're going to have some set of off-chain kind of entity that is going to give you a pre-confirmation on this thing. Whether you call that a sequencer or a pre-confer, you know, to me, that's more of a semantic distinction, but kind of from a technical design perspective, those are the same things. So there's every chain has its nuances to get towards the end goal of building decentralization. Now comes along President Trump.
[00:17:09] And, you know, we are claiming to be at an existential crisis in the industry, right? So some people want to use decentralized technology. Some people don't want to use it, right? You know, there's a โ where are we at as far as the existential crisis of the decentralization to begin with? You know, what's the future of it and why are we at that crisis and what do you see happening now that Trump is president?
[00:17:38] Yeah, so I spent the last couple of weeks, you know, I was at East Denver and then, you know, this last week I was at Digital Asset Summit. And kind of my goal there was to essentially just do like a vibe check and kind of understand like, you know, what is the general feel of the industry? Are people bullish? Are people bearish within that? What is the kind of nuance? I think there's a couple things, right? So, you know, there was this degree of almost this like naive kind of take of like, oh, Trump's just going to buy all our coins and like pump our bags kind of thing. And so, you know, we did see maybe a bit of that, right?
[00:18:06] You know, a couple months ago, right? You know, now there's, you know, always been this thing of like, you know, is crypto actually like a โ like is crypto actually exposed to like macro environments, right? Or is it like an uncorrelated kind of pricing? It's very obvious, right? Like crypto is correlated to like macro environments, right? You know, and the market as a whole, right, is not particularly happy with the tariffs or, you know, certainly the on and off of the tariffs, right? Where it may be even more chaotic, right? If people don't โ you can't move, right? If you don't know if a tariff's on in the morning and off in the afternoon. So we've already seen the macro environment kind of down, right?
[00:18:35] You know, we have recessionary fears. And that's just going to impact crypto kind of writ large, right? I think more concretely, the thing that is kind of, you know, maybe potentially like existential is, you know, there's a bunch of like โ I'd say like low-hanging fruit things in crypto. And I think about crypto as primarily a financial thing. I'm not big on like, you know, crypto as like a social network kind of thing. I think fundamentally the kind of costs you have to pay for decentralization don't necessarily kind of justify themselves in non-financial applications where, you know,
[00:19:05] lack of kind of counterparty trust is, you know, you're willing to pay for that. And so I think about kind of DeFi and, you know, okay, we see institutional adoption kind of coming in. The concern really is โ I think just putting it bluntly, like are we the like crypto natives, the people who are doing, you know, startups and crypto have been around? Are we actually going to be the ones to benefit from that? Or is it actually just going to be, you know, pick your name, right? But, you know, the name that comes top of mind for a lot of people really is like Robinhood, right?
[00:19:32] And is Vlad just going to kind of roll in and kind of, you know, take all of our market and be the one who dominates the kind of order flow here? And okay, great. We made a bunch of, you know, neat software and neat tools. And none of us are going to get to monetize that. And, you know, Vlad and Robinhood or the other institutions, they're kind of going to monetize that all, right? And I do think that's a concern. But going maybe back to your other question on like, you know, does decentralization matter? You know, I do think there are significant benefits to decentralization,
[00:19:59] even if you take a relatively kind of, you know, cynical and less ideological view of, you know, the censorship resistance perspective. There's a degree where like, you know, one of the things I like about kind of like working in crypto versus working in maybe like a normal fintech is the reality of, from an engineering perspective, I don't really want to think about custody. Like, I don't want to actually custody anyone's money. Like, if that is the thing where saying, oh, I want to do, you know, financial transactions,
[00:20:27] or like that is the kind of like, you know, design or product space I'm working within. Fine. Maybe that is the interesting product I want. But I would prefer to do that in a way where I just don't have the option to steal anyone's money, right? I'm trying to build a product. I don't want to steal anyone's money. I would prefer that to not even be an option available to me. And so in that way, one of the things I think crypto gives us is this option to have, hey, you can build this thing in a way that it's decentralized. You don't have to go figure out how to be custodial.
[00:20:53] But I do think there's maybe like, you know, the vibes are bad, like bluntly, right? Like this morning, you know, the coins are up. So like, everyone's like happy this morning, right? But like, you know, we'll see if we make it in the afternoon. But I do think there's kind of an existential crisis of who is going to profit, who is going to monetize out of these things right now. Because, you know, crypto is a relatively small industry. You know, our market cap is sitting at what? We count all the coins at like $3 trillion or something like that. You know, that's a lot of money. But from, you know, the TradFi kind of institutions, it's not that much money from them, right?
[00:21:22] If a lot of these things went away, they wouldn't notice it. It wouldn't cause, you know, some global macro crisis. You know, it's like, oh, all the crypto coins kind of went to zero immediately. And so I do think there's a worry about, you know, institutions moving in and being kind of overly extractive. And specifically, again, the Robinhood point is nothing kind of against Robinhood. But like, you know, it's a tourist kind of thing, right? And it's the, they were not here under the Gensler regime when it was obviously, you know, very not kosher to kind of work in here.
[00:21:51] And even then, right, you know, the biggest kind of fault they made, right, was, you know, when the markets were really, really down mid-2022, right? They forcibly sold kind of everyone's assets, right? They just said, you know, we're just going to sell you at the bottom, right? And, you know, Sol was kind of the biggest notable one because they sold everyone out at like, what, like $8 or whatever, right? And Sol is sitting at, I don't know, $135 today. That's, you know, in two years, you know, if you just held and forgot about it, right? You know, people forgot about it. They go back and they're like, what do you mean you sold me at $8? It's like $135. I just left it there. Why not? So, again, I can kind of go in circles here,
[00:22:21] but I do think there's an existential crisis from a, okay, are we going to be the people who profit? Are we going to be the people who are able to kind of maintain, not just like a moat, but relevance here? Or is someone else who's better resource, who has more connections institutional, kind of going to be, you know, the dominant player because they're already in that side and we're just going to be kind of payment rails for them. And that's going to be something where, you know, they can just really trivially, you know, go buy a couple projects, get their foot in the door.
[00:22:49] And then there's a dominant player in the market. You know, when I got, when I came into crypto in 2017, blockchain had three promises, right? I want to find out they still matter. You know, people forgot about them or don't know about them. Voice for the voiceless, which is where I come in, you know, identity for the unidentifiable. Unidentifiable. And banking the unbanked or in Excel, in Excel's case, unbanking the bank.
[00:23:18] You know, you know, are those still promises or are we shifting towards something else, just finance? I do think generally we're seeing a lot of people trend much more toward this just being a different, essentially just like financial rail layer where, okay, this is a different way to move money, right? I mean, what I look like, you know, what is the amount of like stable coins, right? Just Tether is probably sitting at somewhere in the 150 billion
[00:23:48] and USDC is probably sitting somewhere around 60 billion today, right? And so out of a $3 trillion market cap, we're probably sitting in the, I don't know, 250 billion of that being like dollar backed stable coins. It really was like T-bill backed stable coins, right? That's not, you know, DAI, that's not, or whatever, Sky or whatever, right? That's not the like, it's an on-chain, it's collateralized by ETH. That's just like collateralized by T-bills. And at some level, right, you're like, okay, clearly like we don't get like, you know, non-state sovereignty if you're like,
[00:24:17] your coins are all backed by T-bills, right? And so to that, I think definitely that's the trend we've gone, right? And, you know, I'm kind of in this camp, right? Of like, you know, I raised venture capital to kind of, you know, do a crypto startup, right? And like, what are the things that kind of generate like revenue or fees in crypto? And a lot of it is, you know, essentially swap coin A for coin B and do that in an efficient way with a market and you're a trading venue and you collect dips on, you know, the front end or the back end or whatever, right? And if you're looking at, okay, we raise venture capital
[00:24:45] and we want venture-sized returns, right? That's kind of the area you're going to go to at this more kind of, you know, traditional finance style thing. And what we're seeing now, you know, from, you know, Digital Asset Summit, you know, last week was the vibe I got from a lot of these events was some of the larger projects were essentially positioning as essentially the technology partner to large institutions. So it was like kind of a flashback for me. I walked in a room, you know, there are guys in like plaid shirts and like a vest, you know, kind of thing.
[00:25:14] And I'm like, oh, these guys were like B2B enterprise SaaS sales guys like six months ago. And like, they don't really care necessarily as crypto, right? Again, maybe they do, right? But their job was, I have someone who is less technically sophisticated, who wants to use the technology. And I am the intermediary who is going to sell a service who is going to benefit them kind of getting on chain, right? And whether this is issuing stable coins or whether it's payment rails. And it's very much the same as, you know, I used to be in kind of cloud doing enterprise B2B SaaS, right? Where you say, okay, we're going to be a technology partner
[00:25:43] to Geico or whatever, right? And they have some software engineers, but they're mostly outsourcing. And this was what I was starting to see, right? Of, okay, I'm a bank and I want to use crypto. I don't want to like figure that out myself. I'm going to go partner, right? Or build or buy, right? And then when we see that, right, there's a question of, you know, from a total dollar asset amount, right? What percent of the industry is kind of going to increasingly be that versus this kind of truly decentralized bank of the unbanked, right?
[00:26:13] And then I think the other thing is, you know, if we think about, you know, bank of the unbanked, right? Or kind of, you know, serving this like longer tail of people that are not in the like, the like normal economy, right? We call it like global south or whatever. There's an interesting dichotomy in crypto where ETH is not the place where people do that. Sol is not the place where people do that. Tron is where people do that, right? And it's one of these things I find interesting. People always like to point at remittances as like a specific use case where like, oh my God,
[00:26:42] these people charge so much money and crypto can just undercut them and provide a huge service, right? For cross-border payment. And there's some truth to that. But also if you go look at like any of the large remittance companies, right? Like, you know, some of these are like publicly trade companies. You can go look at like their 10Q or whatever. You know, in the ballpark of like 50% of their cost is like, you know, last mile kind of point of sale style systems, right? Is you send the money, but actually the person in whatever, right? Small town, village, whatever can actually go get the money from a point of sale system where they're actually living.
[00:27:12] And what crypto has not done is that, right? To some degree, Tron has done some of that, right? Where they're actually doing this kind of, yeah, we have people on the ground in South America, in Central America, in Africa, right? Actually doing that last mile. And I think this is one of those things that crypto is kind of underinvested in. And a little bit of my concern is that like, you know, there's a question of the existential, like, is it too late? Did we lose our opportunity? But there is something useful to think about, like, you know, did we just not go kind of far enough from the investments we made
[00:27:41] to actually get kind of crypto in front of people, right? For example, like the Solana phone, you know, got a lot of crap for like being, you know, whatever an Android phone. But I did think that was actually useful because at that time, it was very difficult to get a crypto app, right? Whether it was a wallet or whatever in the Apple App Store or in the Play Store. And, you know, classically, right? The way you do that, you say, hey, I got to partner with one of these guys. If they won't partner, essentially, you can just apply pressure by saying, I'll do it myself. And do you want to kind of, you know, give me the partnership? Otherwise, I'm just going to become
[00:28:11] a competitor in the market. And right, obviously, right? The Solana phone is not a true competitor, right? Like Android's not threatened by them. Apple's not threatened by them. But let them do it themselves. And when we think about, you know, the treasury, some of these crypto projects have, right? Or the foundations have. And the amount of money they have. I do think we should think about what are the ways they're spending them that actually do or do not kind of increase the adoption of crypto, right? If we look at like, you know, what, you know, block, which is like, you know, Jack Dorsey's, right? Like formerly Square
[00:28:40] or however that kind of thing. You know, they did a lot of last mile work to get the point of sale systems kind of distributed, right? There's a question of why is one of the, you know, foundations in crypto that has whatever, you know, billions of dollars of kind of treasury. Why are they not actively kind of pushing point of sale systems, right? Because these are things that actually let you be, you know, bank the unbanked. If you're always saying you only pay for a thing by going through like a fiat on and off ramp kind of thing and you actually have to, you know, your crypto just stays in crypto land and you never use it for real world transaction.
[00:29:09] I do think there's a question of the ability to actually kind of, you know, own your destiny and kind of bring it there because you're always going to be downstream of a regulated kind of entity if you have to do a fiat on run or on or off ramp, right? To actually use it in the real world. So these are the kind of things I think about from a, okay, you know, it's not something we're doing just frankly from like a cost to sizing, right? A scope of thing. But I do think it's something that some of the larger crypto projects should think about is saying, well, maybe that is a valid use of $10 million a year over many, many years to try to kind of get some adoption
[00:29:39] so that, yeah, a person can go tap to pay at a thing and it's just crypto to crypto and you've kind of resolved that and that actually gets you kind of one hop closer kind of to, you know, retail or whatever and it's just more boring kind of low margin thing but it is the thing that, gap between off-chain and like on-chain, right? And all your on-chain stuff that Jeff used for crypto stuff and if it can never touch physical goods then you get in this, okay, I guess we'll just trade
[00:30:09] the coins for the coins because we can't touch physical goods. Yeah. You, you, you went to DAS last week so you're around the institutional players and I see, I see a crisis of being the, the difference between the institutional players in, in, in Web3 and the retail players in Web3 where the institutional players want the retail players to be separate. You know,
[00:30:38] how do we solve that so that everybody's collaborating together to bring Web3 into the mainstream so that crypto doesn't stay in crypto land it goes all over. So I think, you know, the institutions can be like allies and, you know, when I think about it, I think about crypto in like two categories. I call it like the tech side and like the money side, right? And the money side is like the institutions, right? It's like, I think about people like Anchorage, right? Where it's like, you know, Anchorage is like literally a bank, right? They actually kind of hit that bullet, right? They went and got
[00:31:07] a banking license, right? I can build the tech, I can build the blockchain. I don't know where to begin on getting a banking license. I don't really want to know where to begin on getting a banking license, right? If that is a thing I have to do, that is something I want to kind of partner with or kind of outsource or whatever, right? So when we think we can have like, you know, an existential crisis about like, you know, Trump and people kind of, you know, the institutions trying to kind of just like charge fees or rates on us,
[00:31:37] but it's also the reality of we're allowed to do stuff now, right? Whereas before there was a very, very gray area, everyone was a bit skittish, right? And you couldn't ask questions because you go to the SEC's office, you get a Wells notice a week later, right? Now we're allowed to clarify and say, hey, I want to do this. Here is my argument where I think that can be good. You can go pursue it. The thing that needed to happen though, is people need to like, actually do the work still. And I think we got in this little bit of like, nihilistic kind of, you know, and I'm not like an anti-meme coin guy,
[00:32:06] really like at all. I think the meme coins are like fine to some degree, right? It's like, it's a game and it's like, the rules are written on the tin kind of thing. You know, yeah, people could understand a little bit more, but like, same thing as options, right? People don't understand options. People don't understand like liquidity. There's just a degree of like, you know, retail gambling, they're going to miss things, right? But I do think there was a little bit of a period of crypto got in this place where it was pure nihilism and it was just kind of trying to make these abstract games to like pump a coin and get exit liquidity.
[00:32:34] And now what we're seeing is, okay, the institutions are going to come in, you're going to have the ability to do real stuff, but that means you have to do real stuff, right? That means if you don't do real stuff, you might not have an opportunity in the market. And a little bit of the concern is, you know, are people kind of up for it, right? It's like, are there enough people who know how to do that, whether it's partner with the institution, whether it's actually kind of build, you know, the integration or build products because of the reality of like, you know, crypto is like so scared of like, you know, Robinhood or whatever,
[00:33:04] but there's a degree of like, look, if you're like, you know, a web to Y combinator, whatever startup, like you can just go take a shot at Robinhood anyways, right? Like it's like Robinhood is a mobile app kind of thing and they have some big integrators and they're a big player, right? But like you can just make a better mobile app or comparable mobile app to Robinhood, right? And so I do think there's a little bit saying, okay, we have to build real things and that means we have to compete, but that means you have to just build products that are competitive with the existing players. And I think people need to make sure they're not,
[00:33:34] like they're not in the mindset where that's impossible. Like they're not, oh, we're the amateur league and we can't. You're like, no, no, it's just, you just hire good engineers, good product managers. You know, a lot of these projects have like comically large treasuries, right? I'm not sure I see it as like them being explicitly separate. I think it's that the institutions have different
[00:34:03] priorities than the retail, right? And it's a question of like, what does retail want? And a little bit of my view, if we think retail and crypto, generally we're thinking gambling, right? Like really what these guys want is they want to be buying lottery tickets. It's not just crypto, right? It's prediction market. It's sports gambling, right? It's draft kings. It's options trading. It's zero day options, right? It's meme coin. And they want to gamble. And to some degree, ah, you know, I have my qualms, but like, you know, that's their business. They want to gamble, right? But when we look at institutions, right? To some degree, their job is to have a large
[00:34:33] amount of capital and to get returns on that capital. And crypto can benefit from injections of capital. I think one of the like lowest hanging fruits that would benefit crypto is you go look at like what the cost of borrowing like, you know, a dollar in like an over collateralized like loan on chain. And if you're like lending these and borrowing dollars, it might be like nine and a half percent or something like that. Like that's a very expensive dollar and over collateralized loan, whereas like a T-bill is like four and a quarter or whatever, right? And so like maybe it would be nice if some like, you know, large institution just said I'm
[00:35:03] going to take, you know, whatever, five billion dollars and I'm just going to close that arbitrage and maybe five percent is actually the risk rate on, you know, East for an over collateralized loan. But, you know, the liquidation markets work pretty well, but we could get significantly kind of more efficient markets in crypto and that would make the whole economy work. And that's something where we do need institutional capital because there's not someone else in crypto who's just going to inject five billion dollars, but that's like a trivial amount of money for a large amount of institutional players. So I do think we have to think about again, right? It's, you know, what can the institutions benefit for? Know their kind
[00:35:32] of preferences, right? What do they try and get out of it, right? What's their game? What's their benefit? You know, why are they here, you know, and just be aware of kind of what they're trying to do, but try to get benefits from them. And then again, don't be like overly scared, right? Like, you know, like the big boys are not like, you know, it's not impossible to compete with them. Like you can just, you can just have a competitive app, right? Like startups still outside of crypto, They still can like, you know, get a wedge into the market against a large institutional player, right? It's not an insurmountable goal. I do think crypto just has to buck up
[00:36:02] and say like, okay, fine. We have to go compete with the institutions on an even keel. Just go make a better product for them. Yeah, I agree. I agree. I see a lot of products that are better. We'll find out they are in the coming years. So awesome. So I want to thank you very much for your time today for speaking with me. I enjoyed this conversation. I have one last question. How can people find out more information about you, about Astria? How can they do that? Yeah. So, you know, crypto for better or worse is primarily a Twitter, I guess we call it X, you know, a base kind of
[00:36:32] communication industry. So, you know, I'm J Sky Bowen on Twitter or X and that's where you can find me. Astria is going to be Astria.org. So A-S-T-R-I-A-O-R-G on Twitter. That's going to be the best place. You know, we have a website Astria.org. You know, you can use, you know, our rollup if you're interested at flame.astria.org. But that's going to be kind of where our stuff is. But generally, you know, Twitter is going to be where our kind of content is again. For better or for worse, that's where this industry runs. So really appreciate you having me on and appreciate the time.
[00:37:03] Awesome. Thank you very much for your time today.