Hivemind: Peak Macro Uncertainty, Hyperliquid Bull Case & the New Era of Programming

Transcript

The big FUD about Hyperliquid is like this is a DDYDX 2.0. But I think it's reached like, you know, we're talking about 25 to 30% of Bbit's volume, like 10% even higher of of Binance's volume on like a monthly basis. It's like reached this kind of escape velocity. So if you walk that market share up to, you know, from 10 to 20%. And you have a little bit of a multiple expansion because like it's seen as not going away anymore, like you can pretty easily get to like 40 or 50 bucks. Nothing said on the hive mind is a recommendation to buy or sell securities or tokens. This podcast is strictly forformational purposes only, and any views expressed by anyone on the show are solely our own opinions, not financial advice.

[Music] Hello everyone and welcome to another episode of the hive mind podcast. The goal of the podcast is to provide an inside look into what we like to call the deli hive mind. Bringing together some of our brightest minds from each of the divisions to share insights, alpha and [ __ ] posts. Today with us we have the usual crew. Cedus Parabus, head of our institutional research arm. Yan Lieberman, managing partner at Deli Ventures and Ventures Associate.

Actually, Shitcoin scavenger Duncan, also known as Flood Capital. Uh, what's your what's your title at Ventures now? You're scout or something. Yeah. Yan, I don't know. Yan, what's my title? I don't know, Duncan. What's your title? Yan's personal assistant at Ventures. I like scouts, too. Like a boy scout for for ventures. Um, and yeah, I'm I'm I'm Jose. I lead Deli Labs, our protocol R&D arm focused on incubating new web3 primitives since we last spoke. When we last spoke was after Denver, right? Um, it's been pretty brutal markets, uh, choppy generally down, but as we're

talking today, Bitcoin's at 84K, ETH, are we up from then or no? Uh, Bitcoin might be flat, but like stuff like ETH and Soul are definitely Total Cap is definitely down. Yeah, I feel like we're down. Yeah, I'm definitely personally down. But as we speak today, ETH's bounced back above 2K. Um, Bitcoin's at 84K. Uh, hypes above 15 bucks. Uh, what are we thinking? Is this is is the bottom in like or what is this just a bounce? What are you guys How are you guys feeling? I feel Well, the thing is JPAL speaks after this pod today and

that's going to probably be the material driver of where we go next from here. So, I don't know. For me, over the past like few months, ever since all this tariff stuff and trade war and just macro uncertainty started, as I said, I've just kind of been going with the flow. No strong opinions. It's just like not my place. Chris Berniski style. Riding the wave. Riding riding the wave. It does. Like I was thinking yesterday though, like wow, I really forgot how it kind of feels with prices going up every day. Like if you think about it, even in 2024, beginning of the year was pretty

good. Sure. But then we chopped until the election, right? And then we just got two months of craziness of up only every day and you're just thinking, "Wow, this is never going to end. Everything is perfect. We're just going to keep going up every day all 2025." And now it's just Yeah. Now it's just been down only. And it feels like we've been chopping for the last week now. Here here's a good snapshot in time. I remember in March 2024, I was like criticizing Whiff and our head of HR called me a [ __ ] in the full full full company chat. So that's what it feels like when when prices are going up every

day. Yeah. Um so yeah, what what do we think? I mean there's a lot of people I feel like are seeing more macro threads than than ever which is generally like close to a bottom when people start uh uncertainties coming go to cash uh you know tariffs etc. Well, I did see well I did see this actually this chart yesterday that the the amount of articles in the news on like the stock market and economic uncertainty. Um it it charts it from like back in the 70s or 80s or something, right? Every time there's kind of like a big crisis and the right now is the highest it's ever been

articles on like economic buting financial. So, so it's like it's getting more attention probably than what we've actually seen in the market, right? Well, yeah. The biggest dislocation basically between like sentiment and data is is right now like everyone is max bearish. The economy's in a pretty decent spot. Inflation is is like surprising to the downside. Um so I I think you know these are pretty good setups to take some swing. I don't is a bottom in it's it's hard to say right like you know it's the future's kind of a probability distribution and and so there's always a chance that

um you know their their interest in in getting rates down uh supersedes like their support for the market to the extent that you know asset values come down. So basically like the market and and the economy is kind of propped up by the highest spending demographic and and the highest spending demographic really did that because asset prices were very high and so you have the risk of kind of like you know reflexivity to the downside as prices come down spending comes down and and it can get bad that way. So you know that that's kind of what can cause a recession realistically other other than that data looks pretty good. So, you know, if that doesn't happen, I I I don't see there being like a lot more downside. I don't necessarily

think we're in a in a world where we go up, but um I mean, I think in the long run, 84K is is a pretty good entry on BTC. Um Yeah. Oh, what are the kids calling it these days? A vibe session. Yeah. But no, there there are like I think for if you like corporate America um the reason why they're all worried is because the uncertainty is so high, right? And from from their point of view they need to know is the is the tariff thing for instance like the aluminum industry they need like 8 to 10 years to onshore that back to the US. And so from their perspective it's like okay we'll

build take 10 years I don't know but like what that's cap the US went from like being a farm country to the biggest industrial power if you listen to people okay I'm not an expert in the aluminum I'm not an expert in the aluminum industry but if you listen to experts in the aluminum industry they say it'll take like 8 to 10 years and so that's kind of like the biggest uncertainty Uh because even I guess if you get like a four-year commitment from Trump, right? Then you kind of need Vance to win too. Or you could see like maybe the Democrat like that's the that's the whole uncertainty in the economy right now is people don't know should we take

these very long-term capital intensive lots of upfront cost investments, right? like is can we get assurances that this stuff is is going to be permanent now for at least the next like at least the the whole term right and so I think once that kind of gets more clarity that's what the market just wants it just wants clarity right on like the absolute direction and that's what it seems like a lot of people are waiting on at least from like corporate corporate America yeah it's tough because if you if you if you like threaten tariffs then you need to be willing to follow through with it right at some point. And to follow through with it, it does need to be like

a long-term commitment because you're promising companies that they won't uh face tariffs if they build in the US and stuff like that. And those are like you said, very long expensive commitments and likely to be uncompetitive if tariffs get like revoked. So, if he's just trying to use it to get reciprocal tariffs, which it seems to be what most people think, um, then companies, it's tough for companies to make a make a real commitment unless they until they know what the what the actual rate is, right? Because like 30% or 40% to 10% is like massively different in terms of the number of invest profitable like investments you can make in the US, right? If tariffs are 40% then probably

a lot of production can be onshored, right? you have like a massive margin to play with. If it's if it's 10% then it's like totally different. Um yeah, I don't know. Yeah, but I think on the tariff side, you know, yeah, the goal is to you have to actually follow through the bark with bite and implement them otherwise no one will kind of, you know, resend theirs. But I don't know if that it's he's that is necessarily being applied to all of them in the sense that there are I think more permanent like tariffs on industries that he wants to onshore in the interest of kind of um you know long-term health of the US and independence.

Yeah. I think if we go back to the last episode, what we talked about um like the three things that are kind of scary or you know this terror stuff we we just covered uh the refinancing cycle coming up where it's like you know rates were at basically 0% in 2020 now it's like the first 5year mark and and rates are much higher. So if people have to refinance at this higher rate like that could cause it deleveraging. Um, and then also the Doge cuts, which like I don't think we've really got a ton of clarity on any of those. So, still probably like, yeah, like Yan says, I think there's not a great case for for up uh on US equities, but like, you know, maybe Bitcoin could hang in a bit

more, start to recorrelate to gold a bit. I think alts got really smoked um crypto specifically and there was like some coins that just got down to like a valuation that that is attractive even if like there's a longer digestion period of like uncertainty in the market. Um so yeah, I do think crypto kind of got overly bearish uh on on some coins. Obviously, I think a lot of coins are still super overvalued, but there's maybe pockets of of opportunity. Um, yeah. So, I guess that's kind of my my thinking. There's still like these pretty big uncertainties uh in macro land, I guess.

Yeah. I don't think there's a lot of uncertainties within crypto. I have a pretty clear idea of what's going to I think I have a pretty good idea of what's going to happen over the next year. Um, and like it's pretty I think it's also pretty clear what kind of like coins will do best if we go back to being risk, right? Um, well, I don't know. Do you think memes are going to do the best again? No. No, but I think like Soul will do well. I think Hype will do well. Uh because these are all kind of just trading coins that kind of make their profit off of like just trading activity, right? Yeah,

there's more activity. And so obviously people will use people trade like per and stuff when prices are going down, but it does hit a point of like apathy. No, people love pers in the chop like Yeah, but you still get more volume happening and everyone's a gambling addict. But I guess I guess It is youth in Asia roller coaster though because you know people are getting chopped. They're losing money and there's not new money coming into the space. So like it it it's still high but it's it's declining. But I feel like like this level is like is like pretty good, right? We're not like raging, but it's like there's been a lot of volatility. There's been a lot of volume on on these things. Like the the the meme coins have gotten hit pretty hard, but I don't know as as my proxy

for how the meme coins are doing. I watch like the pump fund revenue and and banana revenue and it seems to have kind of bottomed. And you know, it's not it's not terrible. Um, so I I think crypto specifically, and that's what makes this hard, is that crypto specifically, there's been a really like a ton of bullish catalyst like all these lawsuits getting dropped, you know, market structure bill, the strategic Bitcoin reserve, like that's huge, I think, long term. Uh, like really really huge, but it's just like been overshadowed by the macro stuff. So I think if like Bitcoin can maybe recorrelate to gold a little bit which is like pushing new highs or just like either equity markets around the world have been doing decently um then I think I think you know we're not

in a terrible spot especially for some of these coins that have just gotten hit really hard and and um like are trading at somewhat attractive valuations. Yeah, I think it would probably be good to, you know, uh, which mentioned in terms of, you know, having a good idea of what's going to do well when things start bouncing again. Might be fun to dive into that a bit, but more specifically, I guess, you know, you you do have the the kind of the um assumed blue chips now, right? Like you mentioned, uh, Hype and and, uh, Soul and some others. And so if beams don't necessarily do well, you don't have the dilutive effect and you kind of have the

majors potentially getting an outsized bid relative to everything else, which is is I think constructive for the space because we don't necessarily want everything to rise. We'd rather have good things do well because that motivates new capital to come into the space because there's logic to what does well and what doesn't and and makes people confident that they can kind of rationalize having an edge. Um, and then the question is like I mean memes probably come back. Um, they definitely will, but it's not going to be the same ones. It's going to be a coin we don't even know about yet. Okay.

But that it's different than saying a coin, right? Because that's not that's not memes coming back. Um, it's going to be like broadbased rallies. I think Yeah, I guess we can chat through how people see the memes. I think Whiff is fully dead. I think Whiff is just cooked. Sorry, Jason. But like something like that is not coming. I don't think it's coming back. Or like any of these old memes that had like insane runs, like I think it's like gonna be a Bowden type scenario where they had a crazy run and they're dead and it's like you're going to have different new new ones come back. Or do you think though that new memes can get back to these like billion dollar valuations? Yeah. Cuz like for instance,

you see alltime high? Yes. Like the liquidity on them is so thin. Like that SPX6900 one like when that was pumping it's like a billion dollar market cap with like a $3 million liquidity pool or something before it got any sex listings. And it's like those like you know those things get to insane value. I don't know if that's the the the perfect way to look at it though. I think like liquidity matters but that's more so uh the size of the door on the way out, right? But in terms of overall activity, it's just volume. Like how much money is churning in and out of that thing on a 24-hour basis is is more important for the context of like But it's easier to push it up, I guess. Like Yeah. But it's it's if like if you have, you know, 2003

$300 million worth of volume. That means a lot of people are dumping into this, right? So it's like, yes, it's easier to push it up, but you still need to offset that selling. Yeah. Like the delta between the two usually isn't huge, but you also can just like create these massive candles. like one guy comes in and buys like a few million it's like up 100%. And then it's like the whole reflexive nature of like people will pile in and then it just like oh now this is worth a billion dollars and it's like the whales or the devs who like bundled it are just like kind of slowly distribute like they don't they don't full stack. One way I think you can rationalize memes coming back is if the blue chips appreciate to a point where it's like, "All right, I don't know how

much more juice there's left to squeeze out of this play, so I need to go further out on the risk curve and and gamble." Yeah. Like that gig coin, the gigach coin, like apparently that was just I don't know if this is true, so this is complete conjecture, but apparently this is like super supply controlled. And they just like walked this thing up like every month. They had like 12 monthly candles closed green and then it just like went to zero because it's like the people who own all the supply like know that this is a lot about perception and stuff and it's like you buy like the whole thing with a memecoin is like you buy it if there's good memes and the price is going up.

Like that's the fundamental. So, like if they control the whole supply and walk it up slowly over time and just like slowly distribute, then I think like that that'll definitely happen again. Yeah. I legit don't know if I'll buy another meme again. Remember when you were [ __ ] talking me about Goat on this podcast? I mean, goat did go to like one and a half bill though. Yeah, it did. But now it's at zero. Yeah. as is everything. That's good to hear though. See, that's what we need. We need the marketer participants to be like, "Okay, I'm not touching memes again."

Yeah, I think memes memes will probably come back. I think like SPX I've been looking at actually. Um, just like the Mad Don't want to fade the Mad cult. Is Morad even tweeting though or does he just have auto tweets on that are just like they read like auto tweets? Yeah, believe in something. Um, he showed it to me at like 25 mil market cap in in uh token 249. I did not I did not buy. All right. Do you want to talk about like some of this other stuff? Wait, but one want to go through hyperlid. Can we talk about the mecoin's like last point on it and just curious for people's take. So, let's say like Bitcoin goes back to 100K like next

week. Do you think enough time has passed where people like have kind of forgotten how [ __ ] they got on memecoins where like they there's still appetite there or like do you think it's like something like okay this will reemerge in like you know six months from now? I kind of think there needs to be like a a time digestion. I don't know. I think people are like make it all back in one trade. We're going to alltime highs. Animal spirits like people forget very quickly. one pump and it's like, oh [ __ ] like like what's the next? But these these memes are just still so cooked though. Like if if Bitcoin goes up to 100K, like all these memes are still going to be down 90%.

Yeah. So people are going to see that as like, oh, they got tons of room to go cheap. Oh, okay. We'll see. Okay, that was my last thing. Yeah, but maybe like I guess what are people bullish on? Maybe we could do a quick round of that. That's what people really care about, obviously. Yeah. Uh maybe we can start with Hyperlid. You did a big post about it, Duncan. Um that got a lot of views among your your Scavenger followers. What What did you uh what what did you Yeah. Do you want to give us the TLDDR? Yeah, I put this out a couple days ago.

Um and yeah, so hype got down to like 12 bucks and I was just like playing around with the numbers and its growth and I was like, "Okay, this becomes like really attractive here." I think so. Right now it's about 10% and I'm I'm doing the market share. I'm doing like the volume numbers based on like market share verse Binance. So it's um like as a good reference point because I think that shows like the escape velocity it's reached relative to like all these other DEX plays and it's like be because I think the big the big FUD about Hyperlid is like this is a DYDX 2.0, right? like these volumes like someone's going to come out, you know, launch a new DEX and then all these volumes are going to

disappear. So like how are you putting like, you know, any sort of multiple on it? But I think it's reached like, you know, we're talking about 25 to 30% of Bybit's volume, like 10% even higher of of Binance's volume on like a monthly basis. It's like reached this kind of escape velocity. And also like you know those exchanges have a lot of volume that's like at a very very low fee tier like even like negative rebates. So every all the fees on hyperlquid are pretty transparent. So it's actually probably like for fee paying customers like at that fee tier on the per side could be an even higher percentage. Um and basically at like you know between 10 and 15 bucks and the 10% of of market share if you add the circulating tokens

plus the team tokens. So not so not the treasury tokens, you know, Hyperlink is trading between like an 8 to 12x multiple, which is uh pretty attractive and it was only at a 2% market share, you know, 6 months ago. So if you walk that market share up to, you know, from 10 to 20%. And you have a little bit of a multiple expansion because like it's seen as not going away anymore, like you can pretty easily get to like 40 or 50 bucks at like 15 to 20x multiple on 20% of Binance's volume. like 15 to 20% of Binance's volume. So I think that uh and then it's like okay like what's the case for this happening? Um I think that the buildout of their their spot markets is

really big. So this protocol hyperunit as essentially like onboard a native BTC where you can trade native BTC on hyperlquid. It's not it's not native BTC. You can't say that for people who don't care about it. It's native. You can deposit and and withdraw BTC on Hyperlquid. Um and then as native is Yeah. I mean, yeah, it's not native at all. I know, but it's not a wrapped version. Yes, it is where you can withdraw by multiG. Yeah. Instead of a custodian. Just because it's not tokenized doesn't mean it's not wrapped.

People who don't care about this stuff, it's like, oh, I can take BTC with no KYC. Massive qualifier and deposit it on Hyperlid like functionally daytoday. Obviously, the security guarantee. You could say that about WBTC though. You can say that all the versions of Bitcoin uh to to get KYC free leverage with WBTC. You can borrow against it on a but I don't think there's nowhere you can use it as like um collateral which is the the plan. So they have BTC live. They're launching ETH and Soul and then stocks next. And once those things get like liquid enough, then you can use those as margin on hyperlquid, which is like a huge huge huge market, which I think kind of like goes to like the

house all finance blockchain. Sorry, they're blockchain stocks. Yeah. So yeah, hyperlquid is very much like a zero or 100 type protocol. So it's basically that is going to help on the pers volume side and then you know the other side is if they have all this liquidity right and like this really amazing product on the per side and then the spot side it's like an order book as well just like fundamentally those technologies are like better than a Jupiter pers or like a radium and so I think a lot of teams who choose to build on EVM are going to choose to build on hyperlid because they have this product and build off of that product so like to think it becomes you number two like the third large like

most used or fourth largest moose most used chain I don't think is insane and that could be like another 50 bucks on the price of hype. So that that could get us up to like a $100 hype price if things just keep on trending the way they're trending which which is the bullpost. The the main thing also I'm excited about with with Hyperlid is just like you have kind of like the PERS opportunity which I don't know depending how how you buy like Binance and and all these like a 50 to 100 billion opportunity and then you have like the L1 kind of opportunity on top right which is another I don't know 100 to 200 billion market depending how you how you take the the the current incumbents um and then I think eventually they'll

also do a stable coin I'm kind of not sure. Did Did you read the thread? You literally just you literally just summarized the thread. That's that's it. 50 bucks for the PES, 50 bucks for the L1, and then they can also I'm not sure whether the stable coin will be successful just cuz uh that's actually really interesting. It made me more bullish. Athena like Binance launched their competitor, right? The FDUSD um which was like their Athena uh competitor like yield earning. You can earn yield on your collateral trading on Binance. and it's a year in and the thing's at like two billion market cap, you know, and that's the biggest exchange like like less than half of Athena and that's the biggest exchange on Earth. Um, I think it's really tough to do this as an exchange because first

of all, like it's hard to get adoption outside of your exchange and no matter how big you get, like it's there's always going to be other players. Um, and then also like you're just incentivized to only tap into the OI on your exchange, which just naturally caps size and makes the yield lower compared to if you tap into all of them. And you're kind of seeing that with all the all the exchanges other than Binance. I'm pretty sure Athena is in like three of the top five now, like integrated as as yield bearing collateral. But I do think they will do it anyway. They they'll do it anyway. And it'll be it'll be like a decent size. Um so it's just like yeah like in L1 it's like Salana with um like Radium or Jupiter integrated with like uh Drift or something integrated right it's kind of

like the the TAM um yeah S I know you're you're uh not bearish but like you have some some qualms with the Well there's just like tons of it's just like some things are just stayaways for me where whatever happens happens and that's like yeah you al you also said that you wouldn't keep any money on like L2s so it's like I think your risk aversion is pretty high also like once they get native USDC I think the risk of like a hack goes down a ton why because then they can just freeze it instantly it's native the USDC uh circle doesn't freeze unless they get a court order and by the

time they get the court order already gone. What? How can they not freeze? They can, but they don't. They don't. Tether Tether actually freezes a lot quicker, which is kind of ironic. You tag Paulo on Twitter and he's there. Bam. Frozen. I think that they would like Imagine the North Koreans took they would have frozen that. Yeah, hype definitely has a lot of upside. Uh if I were to say the bare cases for some time evaluation is that like uh people might not give it a 30x multiple if they don't expect it to be around in like a couple years. Yeah. Yeah. So it's it's like a 10x right now. I think obviously with there's like a lot more EVM chains

coming. Um you know the the hype the hype the hype VM and the hype EVM they don't work like atomically together, right? there's still you're still going between two different state machines and so it's not like perfect. Um and then you've seen like per protocols come and go a lot, right? And so those would be the main and then obviously all like the close source like you don't really know what's going on. All of those things are other risk factors. Yeah, I think the like we've never seen anyone reach remotely this close market share or just like these absolute levels of volume or OI or deposits. Uh so I agree like that that's definitely a risk. If the volume

goes away, it's going to go down. That's why I think it's trading where it is right now. Um in terms of like the EVM thing, it's like why would you as a user want to go anywhere else? And like as a builder, right? Like if it's a fast, like people just care that the chain's fast. That's all they [ __ ] care about. So if it's like there's three billion or two and a half billion no you care what you're gonna get paid to build on it too. Yeah. You you care about that as well. I I disagree with this. Like I I I think a lot of the really legit builders in crypto actually care about the values of crypto quite a lot. Uh and actually like the there's there's a lot of selection bias with like generally the the worst ones are the ones that

just care how fast the chain is or whatever. There's definitely exceptions. I think like the users have demanded the speed. Yeah. But there's a reason no one went to build on EOS or or like these these these other chains and they they chose Salana. Like Salana had to do a lot of like really solid engineering to keep decentralization while those are two different time periods though. Yeah. Like I think people those are years apart like people care about speed. That's the only thing people care about right now. So if you have like all this natively built-in stuff that's like hugely successful and you're a team thinking about where to deploy like why am I going to go to a different EVM chain versus this one when it's like they already have all the capital and all the users people want to be there. I think it's like a lot of people will

want to a lot of people want to build like uh trading products, right? And you don't want to build on a chain that's competing with you like the you know if if a lot of people want to build per they want to build but you literally can't compete with with what they're doing on the leverage side orders of magnitudes better. I think if you if you're building interesting ancillary financial products anything not trading it is compelling to go there. It's not a no-brainer, but it's pretty compelling. $100 hype.

Yeah. But so I kind of buck it into like So you have, you know, that the Buzz Lightyear meme uh where they're all in Iraq saying the same thing and it's like the case for BTC soul hype is probably, you know, what makes sense and and just because that's the meme doesn't mean, you know, you you should be fading it, right? So, um, those are probably just going to be your standout winners and then people will start searching for something that that that trails and and it's going to get activity. So, you're like I I think it kind of happens in waves where you'll have these run first and then you'll kind of trickle into some of the other ones.

Um, and what those are is kind of where the most money can be made, I think, because th those are decently liquid but have higher multipliers than than uh those major those three majors. Yeah, I think like Cedaris and Jose historically as I've watched you guys are really good at seeing things like early like from the tech side and I and it's like not my strong suit at all. Just like look more pragmatically like what does the market like? So like so in 2023 I was thinking Solano was completely dead. Like 2022 2023 I remember talking to Cedus about this because like you know FTX was gone a lot of their support was gone. I was like I don't really see this coming back and Cedus was super bullish on it

because he was you know thought the tech was great and he was early there. But something like this where it's like the market has clearly shown what is demanded in terms of like they just want to like people just want a fast chain. you know, this is or orders of magnitude more liquid and better than any other per product and spa product. I think it's like pretty clear like we've reached kind of an escape velocity here versus like it's not like an early stage thing like TI would have no idea on definitely. Yeah. So, this is I was going to I was going to bring this up. This is a good idea of understanding how like assets are very different, right?

Um, and I also don't know why Tia versus Hype became a thing because they're completely different like products altogether. But, um, it's like TIA here, I think TIA at this level over like a one twoyear time horizon, I think you'll like I think I'll like buying TIA here, right? But if the market turns around like hype is going to wipe the floor versus TIA on a price performance I would imagine, right? because it's just very reflexive and it does have like Tia still needs to Celsius still needs to even though I think the tech is good like they still need to prove a lot and like actual demand and actual people using it and building out this whole

ecosystem like hyperlquid has this right and so if you're talking like it's yeah I think it's like important to understand your time horizon when you're debating different assets right I also think Tia is more of a venture bet and hyperl is more a traditional like liquid markets bet. Yeah. Where there's kind of I think there's this interesting debate um you know between sort of Anatoli and and ETH people and Adati is like there's only people only want to trade tokens like that's the use case for this thing trade financial assets and I think if that's the case TIA's kind of cooked like I think a monolithic L1 or or a hype type thing is is the winner. um

like TIA really requires you to believe that people want to do more with blockchains than just trade assets. I think and personally I think that's the case. We're going to find a lot of really cool use cases to do with rollups. So that's why I also hold hold tier. Um but I I think for the people that are that are I think hype attracts like the flood uh you know of this world not I'm not not referring to you actually to the other flood although you as well but people who are kind of more pragmatic and also kind of like uh skeptical about block about crypto generally right think everything's a scam or whatever um yeah so to thing is like these six contracts matter and

that's it and I think the bare case on the modular side is that I see people that critique this viewpoint of totally but they don't have examples right they their reasoning is I choose to be optimistic and like I choose to envision there being more use cases but it's not like there are actually any use cases people have off the top of their mind that are going to come along right so it's kind of up to as of now to right, you know, like he's absolutely right that there's basically six contracts that matter and it's kind of up to the rest of the industry, the world,

whatever, to actually build things that are truly unique. And for Celestia, you also probably need a bunch of non blockchain consumers um of your product, too. like just people dumping data there. Can you explain what Celestia is really simply? I know it's hard for you, but really simply you have a really good report on it. Uh yeah, it's basically you I know you don't like to read anymore.

I mean, the whole the whole point of like a I'm not going to say the whole point, but like the original point of blockchains is that anybody could kind of verify that like transactions were essentially legit and stuff, right? And so, and that also has to do with the data being available, which is what Celestia does. And so, all these different blockchains can just dump their data onto Celestia and then um you can do this data available availability sampling, which is kind of this complicated. Yeah. See, I can't even explain it in like a simple Oh, try I it's it's the it's it's verification.

It's like blockchains are outsourcing the their verifiability, I guess, like and the more and the more blockchains that use Celestia, the more it scales because the way data the way Racer coding works is that it's like logarithmic, I guess. And so, uh, you can sample kind of like less and less. But the the but this is the the the problem with Celestia is that the network effects for Celestia only really come into play when it's massive and it has like a very big economy. And so getting to that point like if Celestia does reach escape velocity where they have like tons of high value rollups and applications using it then at that point

you can feel more confident on it. But actually getting from zero to one is something they haven't been able to do. Uh it's very it's like legitimately very challenging. But I will say on the other side that nobody has really they don't really have a real competitor when it comes to like a pure DA layer. Um and like people have kind of given up on that as a product, right? As being like a DA layer and just kind of thinking it's a commodity. So the bullc case would be if that kind of mindset changes and shifts uh like Celestia is in a very good position to kind of take all that demand if it comes. Uh so question so let me repeat it back my basic

understanding. So the current like L2s or something right now have to pay data availability to Ethereum which is really expensive. Uh and if they don't do that then you can't really verify the legitness of the transactions. Well, for take an easy example. Say you post it on Duncan's DA layer. Yeah. Okay. If you just keep all that data, nobody can verify what happened. Got it. Yeah. So, so no one can see like historically that everything is legit. Yeah. That happened on the chain. Change stuff or do something malicious. Well, you can withhold the data, right? And then nobody can progress the chain and then you could like charge like a ransom to actually give up the data again. So, it's just cheaper than ETH, right? Like that's the big thing like right now all

these L2s are using ETH for data availability. Yeah. And there's no it's like a very simple it's a very it takes the other extreme of like it's it's it's kind of like how Bitcoin just does money like Celestia just does DA there's no smart contracts or anything. And so there there are people that are talking about bringing rollups onto Salana but it's it's it's a bit complicated to actually use Salana as a DA layer because it wasn't built for that. And there's a ton of like competing transactions. They have all their vote transactions. They have all their smart contracts rolling and everything. And so Celestia just keeps it like very clean, right? No like tech debt. And they just basically are continuing to focus on this one thing of growing the blocks bigger and bigger over time, making the validators beefer and beefier because

you can make the validators beefy with like the data availability sampling and everything. And so it's very similar to Salana in the sense that they're scaling with hardware, right? Like they want to use beefy machines. They just want to do it where you can still verify. And so they don't make their blocks bigger until they can keep like those properties. Yeah, I actually think I did read this report. It's like um if if they if there's a certain if like all the transactions are posted, you just need to verify like randomly a few of them to like see the whole thing's legit. Yeah, that's basically the point. So yeah, we'll see. I mean, they take that pretty I think that was a pretty good explanation. Simple. They take a very like extreme standpoint.

There's obviously still concerns about like you know how much you can you know I think the bull case for Leia is like a bill a year of revenue and then what's it trading at then is it trading like 25 35 bill right that's like 10x from here and so I also think when people are like how can you be so bullish on Celestia not bullish on Ethereum it's like well they're kind of like moving in the same direction and Celestia is just very aggressively making a very good DA layer and is very cheap and so it's like the value they need to raise prices quite a bit though get to that level of or yeah, like how many chains they or Yeah, they could they could raise prices 100x today and nobody would notice. It's just um that's a good lever they need. But they

right now right now I don't think people should care too much about that. You want to see people actually wanting to use Celestia because that's like hasn't really been proven yet. I think if I think if you see a lot of new teams start to use Celestia, then you feel more comfortable about them being able to like actually change prices later. They have no bargaining power to change prices right now. So would this be like, you know, every company has a blockchain type thing and then like they all use Celestia that's like the bull case or like there just needs to be like a lot of blockchains for very specific use cases of different things and they all need to be using Celestia for data availability. Is that the bull case? Uh yeah, pretty much. I also think though that Celestia needs to make uh and I

think they've realized this. They need to make their own like Tia DeFi ecosystem. They need to kind of like in-house I think a massive chain that people actually want to use to kind of get the ball rolling. Uh I think we've seen like being neutral and taking a hands-off approach like their whole model is build whatever and it's it's like you have no ideas almost in a sense, right? And it's similar to kind of like Cosmos Hub was like oh anybody can be an app chain just do whatever you want. I think you need to take a more opinionated approach and I think they've kind of recognized that. So like how how many blockchains would need to like there have to be like thousands of blockchains that are actually used like hundreds of thousands like how what are we talking to get to to that billion

dollar number? Uh I mean I guess it depends like for instance you can fit I think you can fit 1.3 Salanas on Celestia right now. So it also depends what kind of blockchains. Oh, that's it's all gonna be it's all going to be it's all gonna at current capacity for Celestia. I mean, Salana has a [ __ ] ton of DA, right? Yeah. Um, so how much revenue would that generate if Celestia was fully maxed right now? Oh, it still wouldn't uh be it's just the price for the it's set very low like 0.001 or or something like that. Yeah.

Um, but they're going to keep increasing their blocks over time. Like this is the point of it, right? is like you keep increasing and you keep because as a DA layer you don't have like a ton of bargaining power to raise prices. Um even ethereum doesn't have a ton of bargaining power to raise prices on DA. Like teams are not willing to like So you're you're bullish on it here at $3. You're a fan. Yeah. Is that three billion FTV? Yeah.

Yeah, it's trading at a three times multiple setters. No, I mean if you three three multiple on like a crazy bull three times 20 year 40 40. Yeah. Yeah. Yeah. It's like listen, you need to dream. You need to dream a lot with Celestia. So just let's move on. Let's move on from Celestia. So uh we wanted to talk about SIMD. I feel like uh we've we've talked a lot already. Maybe we could do the RWA stuff like the A and the Athena. Also, actually maybe the Athena L1 is the most relevant to the Celestia discussion because it kind of is a counterpoint against the L2 thesis, right? If people keep if everyone keeps doing their own L1 um it kind of defeats the the whole point of like the Celestia

Initia like app chain. Well, um Celeste more so Celestia Initia thesis to be honest, right? Um yeah. So, so for those who didn't see, Athena announced a new L1 called Converge. They're partnering with with Securitize on it. Um, the idea is there's going to be like big institutions that are going to be that are going to be validators. Um, uh, Securitize is going to be issuing like the I think the wording is Athena is going to be the default issuance.

Athena Chain is going to be the default issuance platform for all their assets which is like the Apollo credit fund obviously Bidd and all the other assets that they they issue. Um so and then they brought a bunch of the kind of DeFi Avengers a Pendle etc to Morpho to build with them on this chain. So they're they're kind of like it's kind of a vampire attack on ETH. They're like taking all the best DeFi protocols and then uh putting like taking secure ties, issuing some some legit tokenized securities and building DeFi around that. Kind of like the grown-up DeFi or whatever. Pretty big announcement, but the price didn't react that much. Like what what do you think? I was Yeah, I I I agree. Um it did seem like a pretty

big announcement. The price didn't react. I think market's just just uh just bad market. Yeah, didn't really see it. I think it it didn't get like properly digested either. Um, yeah, I didn't really see it making too many rounds on Twitter. Alpha. No, but I think so the the I think the important takeaway here is Yeah. So, Athena's launching this chain RWA. They didn't choose to do an L2. They chose to do their own L1. And the reasoning behind this is that they have a bunch of banks as validators essentially. And so, could you do this construction as an L2? Yes, you could. But instead of having a centralized sequencer, you would have this shared sequencer with banks, validators. And then at that point,

you're reintroducing consensus back in. And if you don't care about kind of like doing interoperability with other chains or like if you're going to add consensus back in as an L2, at that point, it's almost like why don't I just become an L1 because that's a bottleneck. Like a big benefit of L2s is that you can run one sequencer. This is the whole consensus bank chain. I agree. I I do think that the Yeah, I think one of the reasons that it didn't the price didn't move is just that this chain has been announced a few times and I think people are just wanting the market just wants more details on on like what the construction is, what it

looks like because uh there's been like a few announcements and it it kind of pumped on on all of them. Um, wait. Is this different than that Ethereal thing? Yeah, that's a deck. Ethereal. Ethereal is Ethereal is like the Hyperlid of Athena, basically. True. Um, all right. Then it'll it'll it'll launch on it'll launch on uh it'll on Converge. Um, I don't know. I think I'm I'm obviously like a Athena bull. I'm not going to do the bull case again. I do think that the there is an advantage and I I want to wait for the technical details to really appline on it, but I do think the L2 is actually a good construction for what they're they seem

to be aiming to do where you could still have like a sequencer be the banks and stuff. Um, and you're right that you you'd kind of have consensus anyway, but I think the escape hatch is actually like pretty useful because otherwise you kind of end up with a private blockchain to some extent, right? Where where these these uh validators can can seize funds. I assume like Guy's pretty smart. I assume he's he's he's thought about this and something's going to come out next week to to to explain it. But for me, um, it almost feels like, yeah, like you said, if you're going to do an L2 and have to have consensus anyway, why not do an L1 and capture that premium? But I think chasing premiums in crypto is not

smart honestly uh like in in general because I think the the irrational premiums don't last long. Like it takes long to build anything and irrational premiums don't last that long. like there was an L2 premium and then with Blast and everyone was doing their own L2s and then Blast got absolutely wrecked and by the time all these L2s launched there was no premium anymore. It's having your own L2 and it's kind of the same with with L1. I just think you want to do the best thing fundamentally for the user experience. Um and it does feel like having some form of escape hatch for a institutional chain kind of makes sense, right? You Yeah. I don't know. Yeah. So, so one thing I guess I'll say on the escape patch is that they don't actually work. Um, it's actually easier to censor by just making

people's transactions fail forever than actually like excluding excluding their You can go back to the last No, there's a there's an article I'll send you on this. There's actually like if you deposit for instance into a money market on an L2, there is no guarantee that you can ever get that money out if the sequencer wants to censor you for the rest of time. Yeah. Because you have to get out of the contract first, right? you have to get out of contract first. The only thing if you're really get out is is an asset issued on the L1 that is not in protocol or anything. Um that's the only guarantee you have that you can like eventually get it out. But every other asset you can't. And so it's like you know how are you what's the

practicality if you're trying to get out? Well, it would just be you can get your crypto assets out, right? Yeah, I I know what you mean. They're going to be regulated institutional assets anyway. you're not going to be able to get them out anywhere. But I guess the idea would be you obviously your your securities and stuff get issued straight on the L2. And so, but it doesn't really matter because they're securities anyway. You can't get them out anyway in any meaningful way. But the the crypto assets would be on L1 and you bridge those over and like in theory if they're not encumbered in any contract, you could take them you could take them back to L1, right? Like um but I guess what's the point of having them there if they're not in any contract? I need to read this article. Yeah.

I haven't like caught up on the escape hatch mechanics in a long time. Okay. So on like that's pretty cool. They're building a lot of stuff this Yeah, they're crushing it. He Yeah, I mean with this plus Ethereal plus um Yeah. So is Ethereum going to be a new token though or is this going to be a new token or are they all No, this isn't will likely be ENA secured and then you know will be the all one token. Ethereum itself as a new token but it's going to be distributed to ENO holders similar to like BNB. He he sees ENA as like BNB and the idea is like a lot of stuff's going to launch in the ENO ecosystem and you know stakers get get a share of all of it or whatever. Nice. Yeah.

I see how you switched TIA with B&B there. Wasn't that the whole thesis? Buy Tia at $20 and stake it for air drops. Yeah, but that that that was not a good thesis. I know. I'm kidding. No, it sounds like played out pretty well. Like if you do the math on B&B like pool APR uh was it was absurd. Like cuz you got it. It's you're actually the only person that gets to dump like the high FB token and they ask for a lot. Like BNB drops like three to five% to those guys. Gave like 10% I think. Yeah. Yeah. And literally the launchpool guys just dump on everyone, you know, like they're the only ones that can realize the FTV.

Here's why the funding that's why the funding situations on BNB is always so funky. Here's a question. Some of these L2 tokens are getting like they're starting to get more cheaper and cheaper, right? Um, scroll for instance is trading at just on an absolute value they're getting cheaper. Not on like any sort of multiple or anything, right? But at a point and like I don't know if this is the right answer, but let's say for instance the EF could buy scroll for like 50 mil and nationalize it. Is that a good move or should they just like I I kind of feel like the EF should launch their own rollups and just have all the sequencer revenue go back to L1, right?

I think the reason why they haven't done this is because they didn't want to take they didn't want to like play favorites and um and all this. They wanted to be they wanted to be they wanted to be new. Well, yeah, but base rollups. Yeah, but base rollups can still be you can still have some of the funds going to another party. But yeah, like Tao, for instance, the reason why it's so expensive to use Tao is and the reason why they pay so many fees to L1 is because they're based, right? But no, I I am like starting to think, you know, will Ethereum actually nationalize some roll-ups at some point where it's a rollup, but it has no own token of its own. It's just sequenced

by Ethereum. Um, yeah. And then nationalize the per decks on the rollup as well. And then isn't that the squad exchange, it's not like Yeah, it it takes away from the whole credibly neutral thing, right? So that's why people haven't done it. But that's why that was a big argument in the cosmos days like the atom days right like the cosmosism whether they should implement cosmosism or not. There was things about them acquiring various uh like osmosis and stuff like this. Yeah. I mean you're kind like these guys aren't practical in terms of what people want.

It's like very idealistic and like maybe that is gonna end up being the vision like that's realized. But I just feel like going for like practicality of like what people want right now and then building that is a much better route. Like the EN hyperl strategy, build an app people want, build the product people want and then build all this other stuff is like way makes way more sense to me than the other way around. Yeah, I don't see the AF doing this. I think at some point it would be a good decision. I don't know about scroll. Uh yeah, but some kind of hungry team like me would have been awesome if they could afford something like a team like that.

Um just by I mean my is really hyper now. Let's see how it goes. Yeah, I mean something that's that's like uh has a narrative that that's comparable to like a Salana, you know, where it can be super fast on. I I saw on the talking list the vibe coding thing. I want Do you want to talk about that? I thought that was kind of cool. I asked Chad GPT, I saw everybody talking about vibe coding. I asked Chad GPT what vibe coding was and it was like casual coding, writing coding for fun, low coding sessions, putting on some chill music and getting into the zone. All right, I'm pretty sure this is not what everyone's talking about. It it's a term coined by Andre Karpathy about just like coding with LLMs, like coding with

cursor and and Claude where you're basically like not really coding. you're just instructing the model on on how to code. Um, and it's a super crazy experience. Like I I'm not a coder. I haven't written code in like 10 years. And even then, it was like some front end like CSS stuff. So, this has been like a learning curve for me. But, it's it's crazy the stuff you can build, how fast it is, and like there's a lot of [ __ ] that's crazy about. I mean, the one really crazy thing is it has taste. Like, it has the taste that a product manager would have when you're deploying a product. Like I did I like normally the product in when we do in labs we like wireframe stuff internally either on paper or using like wireframing. I

like to do it on paper which everyone gets really upset about because I can't draw. Um but people do it with with software and then you give it to a to a designer or they turn it into like designs and then obviously like once the designs are final then you need a front-end engineer to put them into CSS, right? And and then to link it up with the back end all this stuff. with with this it's I literally described my app idea in one paragraph and it derived the UX that was like 80 to 90% similar to what I had drawn out like uh it was it was wild like the thing has taste um so that's really crazy and the other thing that's crazy is watching because I've been doing it with a friend of mine and he's like one of the most beastly uh

vibe coders I've seen and it's also really interesting there's like niche vibe coding community on Twitter uh who on the bleeding edge of this stuff like using all the crazy new tools and some of the stuff these people build like insane games uh insane websites like with with is is crazy and the process like another interesting thing is the process of debugging and working through problem is so interesting because there's some background noise um it's so interesting you don't actually know the code right so like when the thing doesn't run you get like an error message, you have no idea what uh what the [ __ ] going on. And so most of the time you're like sending the error

message to the LLM and just telling it like basically coaching it like hey I got this error message come up with yeah I mean the please fix this is doesn't work that well often like what and the better people what they do is they literally coach the LLM. They're like um analyze the error uh come up with five to six different solutions like uh think about the trade-offs of each solution and just send me that don't execute anything now. And then you're like, "Okay, now of those solutions, which one should we try first?" Like, and you're you're like literally like a coach to this thing, like a like a teacher,coach just helping it work through problems.

And you actually like what what you bring is just and like the better the coder, the better intuition they have for how to solve problems. And so they're much better at coaching these things. Um, which kind of like a post I did today that you know like Tiger Woods has a swing coach, right? like no matter how good this thing gets at coding, it's still going to benefit from from a coach. Uh what one would hope at least. Um but right now it does seem like that's the case. Um and the [ __ ] you can build like I'm building just like an MVP of this idea I had and it's it like in two days the progress uh it's this meta feed thing. It's like a longer conversation, but it's it's basically just like um like right now to keep up with the world, you got to like catch up with

podcasts and and Twitter feeds and group chats and like articles subscriptions. So the idea with this is you just have um I think in the future that won't be the case. You'll just have an LLM that reads all your favorite content and and gives you summary summaries. So like you just tell it these are the people I respect in each field and and you just ask like okay macro I want a weekly report with like what Arthur Hayes and and Zultan and like all the whoever the people are I I respect are in macro. Um and it just gives it to you and then you can also like chat to it about it, right? It just keeps it all in context kind of thing. So I'm building this initially for just like YouTube and podcasts. So you don't have to listen or watch all of them. You could just like

ask for a summary or chat to it about the content. Like you don't need to watch your eight hour Lex Freedman or whatever. That's a good idea. Yeah. Yeah, I think so. Not just a hat rack over there. [Laughter] Um, yeah. And like I think it's going to change just how you I think LLM rappers are actually going to be like insanely valuable. I think most of the value will be in these rappers. Like people used to hate on these rappers, but I think people are going to be able to just build like automations of workflows that you do with LLMs like this one, right?

and just make them way simpler and charge like massive margins on your cost of inference and and indexing the data. I think you're gonna have like massive businesses that are literally just rappers. Um yeah, and I recommend all you guys try it like just just spin up curse. Ideally have a coder friend because otherwise there's a lot of like frustration that that comes up with just like random Yeah, you get stuck on just little things. I I've tried this before to to make a bot to pull like stuff from Coin Gecko and my friend was able to do it in like five minutes. I was getting stuck on just these like little bugs I know nothing about. Were you using cursor and and Claude? Uh we were before LLM. We were using No, no, we were using an LM. I think we were just using DBT

like 01 or something. Yeah, Claude's also a lot better weirdly than all those. Like I kind of compare them and most people who vibe code like are just Claude Maxis. Um yeah. Yeah. I feel like on this there's Adelfi invested in this company, but um it's pretty cool. Hightopia, it's like a game where there's like these, you know, people are vibe coding these websites and stuff, but they're also vibe coding games. And so the the founder just posted something on his Twitter. Uh his Twitter handle is imarchdev, but he basically like vibecoded this Fortnite like COD deathmatch type game in their engine.

And he said this took him like 15 hours to do. And you know, doesn't look obviously like, you know, a war zone. It's like kind of like a Minecraft Voxil style game engine, but like it has guns. They run around, you respawn, you pick up loot. um you know, it's pretty cool that they can make this in like 15 hours. And their whole idea is that they're going to make like a marketplace um to allow like content creators who are like doing this vibe coding game stuff to make a game, you know, monetize and then like instead of just like messing around with this for a day because it's fun to post on Twitter, like you can actually put like, you know, a couple weeks into uh into these tools. And you know, I have a friend who's building

functionality for it where like you describe in plain text what you want the game to be and it'll actually work on that like in the engine. It'll like create that game in the engine. So yeah, Deli Deli's invested in that and it's just so early because like there's even there's a bunch of like prompt scaffolding that makes the LLM so much better. like for whatever reason if you tell it like imagine you're a ex like senior software engineer or like economist depend if you're doing research um it'll just do a way better job and like there's there's just prompt um scaffolding you do that that makes its responses way better and like people are only just kind of scratching the the surface of that and and like when you're coaching it does it actually get better like for future use like when you say

like okay think through this and then like the next time like would it know how to handle that error more easily? Uh no, not right now. Um they I think they are like cursor and claude are definitely using the data and like they have a bunch of people just uh like sending basically pair programming with the LLM and sending like labeled data sets to it to make it better. But I don't think it it that's actually an issue right now with it. It's hard to um it's hard to have like the thing you'd have with an algorithm where it can like improve or learn your learn your preferences. The context window is the only thing you can you can use for that.

But yeah, because I was talking to the the the Htopia guy and he says like they're basically doing the same thing where you have like a text to to games. You type in like this is my game mode idea or like this is the world I want to build and then the game will create it. But as they have like but that pulls like references from other games that have been created. So as that library like gets bigger then like you'll be able to do more and more complex things by just saying like hey you know spawn me in this map with like this gun and do so I thought that was cool that it's like iteratively building on each other. It sounds like the LMM's slightly different thing but pretty cool.

Yeah it is cool. Guys you want to talk about anything? Do you want to talk about a thing or SIMD or or more AI stuff or should we call it? This is good right now, right? Probably just Yeah, call it. Mhm. Yep. Solid 15 minute hyperl liquid chill. Oh, the other thing is if you're building something in just the last thing, we're doing the accelerator. We're doing an accelerator with Deli Labs AI crypto accelerator. Uh it's backed by Salana, Ethereum Foundation, and Bass. So basically have access to everyone in the top crypto ecosystems.

Um you have a bunch of top mentors. You had up to 500k in funding. No ownership required to give up to just join the accelerator. And then there's a demo day at the end with all the top investors in the space. So we did our last one with near and companies raised like 20 million plus off the demo day. So, highly recommend applying whether you have like an idea or not or whether it's just like uh we we have a bunch of ideas for stuff you could build with AI. Um yeah, so if you're if you're building something in AI, building something in crypto or in the overlap, please apply.

You have until the 26th of March to do so. Um yeah, definitely definitely worth it. Sweet. Sounds cool. So 20 million last time in funding. That's a that's a good amount. So, how many projects was that? It was six. Six projects. Nice. Yeah. And is that like a mix of Deli and or like other VCs? It was actually A16Z did some Cyber Fund did some. I believe Dragonfly did one of them. Um, so they raised from a few different people and then a bunch of other smaller smaller VCs too. Nice. And we did some too, obviously. Yeah.

All right, guys. Thanks very much for tuning in. I thought this was a good one. Actually, there's some stuff we can clip here. We should we should make sure we do that. And uh yeah, see you all in two weeks for another episode of the Hive Mine. Yeah. Cheers. See everyone. See you. [Music]