Rollup TV: Neutrl Labs, Berachain and ElizaOS

Transcript

[Music] Heat. Heat. [Music] Heat [Music] up here.

Heat. Heat. [Applause] Heat. Heat. [Music] Heat. Heat.

Heat. [Music] Heat. Hey, hey, [Music] hey.

[Music] live. Let's go. Uh, happy Monday. Tons of action all across the board, man. We're going to have a day. We're going to have a week. We're going to wrap up October strong. What's everyone going to be for Halloween? Let me know in the chat, guys.

I mean, if you thought it was all over, it certainly looked like it might have been, but now we just have this feeling. There's a feeling in the air. Something amazing is about to happen, and we're going to figure out what that is. Um, Leo, go on mute, please. Uh, we've got an absolutely packed show for you guys today. As you can tell, I've got a whole new camera situation going. I'm pretty psyched about it. I think it's going to make for quite a new look. What do we have? What do we have? What do we have in the way of esteemed panelists on Rollup TV today? Well, I'll give you the heads up because we've got a banger

lineup. Uh we've got Talin from Neutral. Neutral uh has some incredible news. I want to find it for you so that I make sure that we get it right. Uh when neutral comes on, we're going to be talking about something that just broke. We're going to be talking about the additional 25 million vault cap that has been fully filled, bringing Neutral's total pre-eposit vault to $75 million. There's just just numbers that back in 2016 when I got into crypto, back in 2020, even these numbers are astounding.

75 million for neutral. We're going to see what that's all about. We're going to get uh K uh on uh he's part of the neutral team, but I believe we've actually got Bin for who's going to join us from the show. He's got a got an interesting name. We're going to we're going to learn all about neutral uh the next generation of cryptonative yield. These yielding products are what is going to take us to Valhalla. Afterwards, we've got a friend of the show, Smokey the Bear from Barach Chain, who if you guys haven't seen this, man, Barah is making strides today because guys, we've got another

digital asset treasury company. Let's make sure that we get this right. The Barachchain Foundation announced today that they are excited to support Green Lane Holdings, publicly listed NASDAQ company, as they build the first and only Bara Digital Asset Treasury. Pretty insane. Bara Digital Asset Treasury. There is a publicly listed NASDAQ company that is fully focused on making bar chain, the chain with a bear on it, a pivotal piece of their corporate strategy. Green

lane's launch marks the beginning of a new era for Barain as it expands into the traditional capital markets and unlocks the potential of proof of liquidity for the masses. man. Bear chain's been kind of uh flying under the radar a little bit, but if you're keeping an eye on the chain with a bear on it, you're not going to be disappointed. And last, but certainly not least, we've got uh Shaw from Alisa Labs. Alisa the AI project. Alisa OS is building uh one of the core pieces of crypto AI. And we're going to be talking to Shaw about this. He's had some spats with Twitter. I don't know. He's suing Twitter if you guys look him up. Uh so

there's going to be some pretty interesting uh things that come out of that conversation. First, we're going to be screen sharing. Let's get let's get some tweets up. Uh there's quite a bit that has happened over the weekend and I want to talk about some of these things with you guys. So, let's get it up. Uh we're going to get some reactions to the timeline. Uh let's see. Let's see. Let's see. Okay, we're up. As you can see, rookie XBT, something amazing is about to happen.

People are just they're in the comments and they're trying to figure out, is this bullish? Is this bearish? What is happening? What does Rookie know? And so, as a result, what what do we have here? Well, well, I mean, you guys tell me. We've got Bitcoin passing 110. We had a weekend uh uh weekend dip that got slurped up. Then we dipped down to about 106 or 107. Um I was spending the weekend with some family. Bitcoin was all the rage. Retail is taking notice.

They're not just exit liquidity anymore. Uh rookie says something amazing is about to happen. I think he's right. Um there's a couple guys uh that are saying we're going to go one direction or the other. We've got guys like Chris Berniski. These guys are making claims that were essentially topped out. You've got guys like Chia Wang. Chiao is one of the just esteemed just, you know, super legendary guys when it comes to the market. Um has great uh great track record. Uh as does Chris. And these are a couple of the guys that are saying, "Look, we're nearing the end of the cycle." Um it's pretty interesting because you also have the four-year cycle at play. And a lot of people are saying, "Look, four-year cycle, it

doesn't really exist anymore. Bitcoin is a little bit less. Bitcoin dominance not as much. The having just has less and less significance on uh the market as a whole. Yet here we are with a couple of the main crypto Twitter personalities saying that we've topped out. And so where does this leave us? We're going to find out. Uh and then you've got guys like uh Ben Cowan. Uh you've got guys like uh John Bowlinger, the inventor of the Boinger bands who says, "Look, ETH and Solana poised for a leg up. Bitcoin not so much." And you know, this kind of signals to us like, "Look, how do we think this thing is going to play out?"

Uh Ethereum is struggling to stay above 4K, but it is it is teetering. Um and so we could see 4K turn from resistance into support. Obviously, we blasted right through that on our move all the way from I mean, ETH was about 1,500 bucks this summer. Anyone remember that? That was just a couple months ago. And it blew I mean, essentially tripled in a span of I mean, from April to August, it essentially tripled uh a span of about 2 to 3 months. It's insane price action.

Um, in the last 3 months we've trended downwards, but it seems as if we have found support around 4K. We're going to figure out if we can solidify the support around 4K. If so, we're going to see where this next leg takes us. Um, and we're going to continue to move forward. Uh, what else do we have? If we look at ETH Bitcoin, compare this with Bitcoin. Well, the ETH Bitcoin ratio is teetering.

Um, you know, we had a leg up on ETH Bitcoin and it seems as if we're going to get another leg up. If we look at ETH Bitcoin, you know, when ETH started moving from April into August, went from 1500 to 45,500, it surpassed Bitcoin quite a bit. uh we had a nice bump up in ETH Bitcoin and it seems as if we could get another leg up and if that happens we could see the ETH Bitcoin ratio um revisit the point075 mark which is going to be quite interesting. Let's speedrun through some more tweets before we bring our first guest up. Something amazing is about to happen. You guys let us know is it bearish or is it bullish? I think

regardless with the amount of uncertainty in the market, gold spiking, Trump tariff, we're going to get a Trump chi Jingping meeting sometime in November, the government shutdown could reopen as soon as this week. All of these things are uh events that are going to pose as catalysts in the market. But again, you guys are the deciders of whether or not this is a bullish or a bearish catalyst. Regardless, something amazing is about to happen. Let's keep moving. This morning, AWS broke. That has no bearing on whether or not Rollup TV goes live. Rollup TV goes live no matter what. AWS broke. The entire internet is

down. Doesn't matter. Rollup TV goes live. These are some of the companies that are experiencing downtime. I mean, just a lot of them, almost all of them. Uh, Coinbase reported outages. Robin Hood reported outages, Reddit, Capital One, Roku, Signal is down. People are pissed that Signal is down because it means Signal is dependent on AWS. How can you possibly be a decentralized secure uh anonymous network if everything is just stored in an AWS database? That doesn't make any kind of sense. People are pissed at Signal. Disney's down. Coinbase is down.

Venmo's down. Snapchat's down. Lyft, Hulu, Roblox, Fortnite, Robin Hood, HBO Max. Pretty insane. McDonald's is down. Pretty sure you can go you can still go buy a burger. Um, everyone who's short right now can still put in a McDonald's application. There's no problem if you just take if you print out your application for McDonald's and you bring it into the store, they will continue to process your application. Online applications for McDonald's's website are down. What else we got? Redstone had some shaky controversy over the weekend. They

became aware that their CMO at the highest I mean CMO Redstone is Redstone's a billion dollar company according to their token right what's the Redstone's FTV never mind 350 million number 500 in market cap their CMO decided that they thought it would be a good idea to intentionally dox some people in DMs. That whole campaign that you guys saw, screenshots went viral. Um, made no sense. Uh, they decided to part ways with Peter effective immediately. Pretty great

decision on behalf of Redstone. We're big fans of Redstone. Marson is a friend of the show. We may have him on to uh to explain some more of these happenings. I'd like to think that this did not come from the top. You know, the CMO, I think, uh, took it upon himself to run this campaign. Uh and again any form of coordinated harassment or doxing is not acceptable. Uh their culture is built on integrity, respect and a shared commitment to advancing the blockchain ecosystem through innovation and collaboration. I truly believe them. Uh I think this was a lapse in judgment when it came to their CMO. I think this guy is going to have trouble getting hired uh in the rest of the industry.

Once you do something like this, you have to rebuild your reputation. I still hold Marson and the rest of the Redstone team in the highest regard. Red uh Marson, welcome on the show anytime. Big fan and uh think he's doing things the right way. I think it was the right decision to part ways with Peter. Uh it's a difficult decision, but you got to do what you got to do. The company's reputation is at stake and you have to isolate uh the decision maker in this case, which is Peter. Uh and and make sure that um you know, we we uh effectively serve justice when it comes to the corporate structure that we have here.

Focus remains on delivering secure, reliable and high-quality Oracle solutions. Mission has not changed by builders for builders. Got to respect the anonymity. Um we cannot uh we cannot let this sacrifice our morals. Cannot be doxing people. Okay. Uh I think we've got just a couple more things that we want to run through before we get our first guest up on the show. Uh, what do we got here? Anatoli, founder of Salana.

Hilarious. Mer is literally clowning him in the comments. Uh, Anatoli opensourced a GitHub repo effectively. Apparently, what we're doing is just uh he was just messing around on Claude and he decided to just open source this and then the whole market went wild from it. Um, obviously this is going to pose potentially a conflict of interest if Anatoli and the Solana Foundation go and launch and start to develop a uh a perpetual solution of their own and start to compete against digital asset uh projects that and and ecosystem projects that are built on uh Salana.

We're just going to speedrun the rest of these tweets as we bring up our first guest. Um Coinbase uh as well as Robin Hood said that they're experiencing outages. This is due to the AWS outage uh that we talked about a little bit earlier. And last, but certainly not not least, uh Elon is back tweeting about crypto, tweeting about Faky and memes. Faky is back on the job as ex CEO. Here we go. It's the uh the beginning of a new era. Okay, let's get our tweets off of the screen. It's time to bring our first guest up. We're going to be talking about Neutral. Neutral had some amazing news. Uh they recently uh upgraded their new caps and I think they've already

filled up their new caps. And so uh we're going to bring on uh we're going to bring on the founder of Neutral in just a moment and we're going to be talking about Neutral. Uh they're flying under the radar a little bit. We're going to be talking about what they do, how you guys can get involved, and how you guys can capitalize on the opportunities that are available as a result of Neutral's caps. So, I believe we're ready to go. Leah, if we can bring up our first guest, uh we're going to we're going to have some fun. We're going to talk about Neutral and we're going to uh understand where they are in their process.

All right, we're going to bring them up in just a moment. Uh guys, if you haven't already, take a time. Don't be a ghost follower. Go ahead, smash the like button retweet subscribe whichever platform you're you're watching on. Uh leave us a comment. We do enjoy questions and answers. Uh if you guys have a question for any of our esteemed guests, please let it rip. Uh send it through on Twitter or YouTube and we will ask your questions so that you guys can get the answers that you're looking for. Neutral coming up in just a moment. Stay tuned. The next generation of cryptonative yield. These guys flying under the radar a little bit. Additional $25 million vault cap says it's been

fully filled. We're going to get the alpha in just a moment. If you guys missed it, you know, I'd imagine they're going to announce another round of these vaults and raise the caps even more. This is the strategy that we've seen other launches take. uh this is what tends to work when it comes to the go to market strategy for raising liquidity and funds uh within a protocol. So $75 million so far uh I believe $1 billion coming very soon. We're going to ask uh Breen the founder of neutral as soon as he's up.

Okay. Okay. Okay. Without further ado, I think we're gonna bring him up. The suspense is killing me. I know Bereine's I can see him behind the scenes. I know he's fired up, too. He's ready. Okay. Okay. Okay. Let's go. We're going to bring him up. I promise we will. I promise we're going to get him up here. man.

Beautiful. >> All right. >> Hey, how you doing? >> What's going on, Bin? Welcome to the show. >> Great, great to be here. Thanks for having us. Um, such short notice. Great to be here. >> Yeah, man. Well, I appreciate you coming on on such short notice. Uh, also, thank you for your patience. uh and and uh bearing with us as we brought you up. Um we are live. Uh so this is uh the nature of a live show. Anything can happen. Um man well >> keep it exciting, right? That's that's most of our life in crypto. Everything is live, right? Uh so it's all good.

>> We testing prod. No different here. >> Well, man, congrats on all the success. You guys recently reached the uh $75 million mark on the caps. Um you raised the caps and then they're fully filled. Uh, how long did it take you to fill the the latest $25 million uh vault? >> So, it was actually a bit of a mess up, right? So, last week uh we opened up the caps initially. I think that filled in about 16 minutes, 50 mil. Uh, today it took over half an hour, but there was an issue on like the back end. Uh, too much demand and then also AWS um causing some havoc today >> on neutral as well as upshift, but it still filled in about half an hour. the

rest of it. So, we are grateful that people still came in and uh I do apologize to those that didn't have a chance earlier. So, it's been good. >> Cool, man. I mean, what when when you look at the appetite for what neutral is building and I want to get a little bit of a primer for our audience on what neutral is and how it works. But before before we get into some of the nitty-gritty, when you when you look at the appetite for something like this and you say, "Look, we're going to raise our caps. we're going to open up some more deposits and then it fills in a matter of a half hour. Like what goes through your mind when you see the overwhelming demand for what you guys are building?

>> I think um the backstory in neutral is that the team came together and decided to build something that we thought a product market fit. Um realistically it just kind of cements that other people believe it as well. Uh we are grateful that people do trust us to to do it. um free these caps. We were running 50 more privately also. And I think that was a strategy that most teams don't do in crypto, but I've been in the space for a long time myself. And I feel like you need to show people that it actually works, right? Versus just playing a game of raising caps and points and all all of that. Like we will deliver the highest on chain yield once we get going. And um still grateful that people do trust it because I mean it's a tough

market. A lot of people got blown out about a week and a half ago. Um, but uh it's good to see the demand and I think it is a strategy that will work. So maybe a lot of people are excited about it. >> Awesome. And may it's a good time. Let's let's you know educate the people what is neutral. You guys are the next generation of cryptonative yield but what does that mean? I mean we've we've seen the likes of, you know, pool one, pool two when it comes to yield farming, LPS.

>> Um that's when uh the rollup was started. Uh we started with, you know, basically teaching people how to yield farm on sushi swap, on unis swap, um on yams. I mean, back in DeFi summer 2020, how to deposit into a uh and use these onchain apps. Um and there's opportunities for yield in some of these lending markets and in some of these LPs. um how have you guys built something different? It's not just, you know, a fork of one of these uh protocols that's been around a long time. Maybe you can just educate the people on uh on neutral's unique strategy.

>> Yeah, I think that's a beautiful segue, right, that you actually have that background and your viewers as well around DeFi Y farming and other type of strategies. I think what we try to do is build something that hasn't been done before in the sense that we're tapping into yield that exists um in the private markets and is bringing it on chain. It's not a new trade, right? So, if you're familiar with Athena and like the basis they run on USD, SUSD on Ethereum and Bitcoin, that's not a new trade either. I mean, the Delta neutral basis is a traders 101 for like since cryptos existed. Um what we've seen over the last like four years is private markets there's been a tap into this trade called OTC arbitrage. So guys who are

the biggest funds liquid funds in the space market makers we know most of their names. um they get these private deals from foundations from teams um and they buy lock tokens often and they buy at a discount from from early investors VCs or foundations and then you short a perp so basically you're buying like Salana Avac Sui whatever like 30 40 50% below market price and then you short a per across Binance buy but OKX across multiple venues and over time you can unwind that discount but see the difference between normal basis where you buying spot and you're shorting spot um the per and you just get paid funding rate is there's an extra margin here. So you are getting the discount over time

as the per and spot unwinds as well as the funding rate as well as the staking yield you know and it's something that hasn't been brought on chain before. So it is largely um can only be done if you have tapped into supply to be honest. Right. >> Right. You need access uh to the foundations and and in order you know to buy these lock tokens. I'm curious from the from the foundation perspective. These guys are willing to uh sell their lock tokens uh and do so to market makers, but they know that these market makers are going to put on this trade where they effectively take the tokens and then they short the the perpetual as a means of locking in that value. What's in it for the foundations? Why if they

know that a market maker is going to go and short their token on purpose, what what what's in it for the foundation to uh to make this sale? >> Yeah. So, I think a lot of legit projects in the space like if you're looking at the top 100 coins right now. Um I mean there's a lot of projects that have come across since the last four years when we uh like yourself has all got into crypto when there weren't that many. There was a handful of guys that you could um buy your tokens you could invest in. right now is a lot of legit teams that are trying to build um need to get liquidity. They need to keep growing. They need to build out the ecosystem. They have to make sure their initial investors, whether it's retail, whether it's VCs um are actually still

um are able to actually grow into it because I think there's too many tokens out there. So largely I think teams come to they go to market makers, they go to OTC brokers, they go to investors and they sell these slugs largely for ecosystem growth. Of course, let's not be like so naive, right? Some teams obviously are selling just to buy a yacht or whatever, but largely why they have to do it is because there's not enough liquidity. So if you really look at it, I mean, forget about Dex liquidity. Even if you look at Sex's um which has more liquidity than than Dex's right now, you can't go and sell 1% of your token supply. It's impossible.

You'll destroy the token price itself. Um you'll actually probably destroy all like investor and like ecosystem and um sentiment that you build out around the community and it's the only way I think you can actually get it out. It's just like this niche that exists in crypto. Um and I think uh largely it's something that should be accessible to others otherwise it's the seven for same seven or eight funds that just keep uh running this trade. It's incredible. And I I had imagined like you said, a lot of these foundations, look, sometimes they need liquidity, whether it's decentralized exchange or centralized exchange, but a vast number of them are selling these lock tokens to

go finance their lifestyle, to go buy a yacht, go buy a penthouse, right? These sorts of things. >> Yeah, definitely there are people doing that. >> So, with that understanding, usually that's why they keep these deals quiet. That's why they kind of fly under the radar and they just, you know, deal one-on-one with a market maker and that's why it hasn't gone on chain up until now. Did you see any push back when you started to make kind of a a push to put a product like this on chain? You're obviously exposing, you know, something that has flown under the radar for some time. Did foundations kind of push back on uh putting a product like this and and shedding more light on these types of market maker deals?

I mean realistically we will like our product will only work with transparency to some degree. So like if you look at our Athena initially started there was a lot of FUD around it a couple years ago even though it was just the basis but we have to make sure we shine as much transparency as we can. uh we will not show actual like underlying names but we'll use ZK to actually prove that we have custody of those tokens and people will be able to see the vesting onchain. They'll be able to see the the cash profit that's being made from the trade as it rolls all on chain but there has to be something around transparency.

It's not really about the foundations being like concerned. Um, also it's more about there could be nefarious actors in the space, right? If you are running this trade and you short like a certain token, somebody could come and try and blow you up. Who knows, right? I mean, we have to think of all edge cases. And realistically, I think foundations are happy. We've seen over $800 million in OTC deal flow in the last four months. And that's across many different foundation the tokens. I think there's not enough buyers to be honest. >> Like I know we can say like crypto is a massive trillion couple trillion dollar asset class but realistically there's still not enough buyers. So you need more people coming into the space to

actually and this is why you see it even on the trade itself. I mean, we look at old coins right now. I mean, unless they raise a DAT or the teamself are very rich and they putting it back into the chart, 90% are down like to levels they were four years ago. >> Yeah. Yeah. Absolutely. I think like you said, DATs, I mean, we saw the bear bear chain smoking the bears coming on right after you, so we're going to be talking about that. Um, and a a lot of the projects just they don't have the sort of cash flow or revenue model that is sustaining a chart. Um, and that's why that's why we're seeing things progress uh in the direction that they're going.

I want to walk through a couple of case studies. One is the XPL pre-market that went live on Hyperlid. Um, obviously, you know, investors got in. A lot of them are up significantly um from the echo round or or early seed rounds that that Plasma did. Um and then they went live. Uh Hyperliquid had a pre-market uh and we saw it run up to a dollar plus. Now it's you know 40 cents or so. But Justin Sun >> uh supposedly allegedly Yeah. is the one who, you know, participated in this. Um, and and you know, we saw a whiplash on the chart where people got uh people were running the trade that you're talking about, right? They had lock

tokens, they were shorting on Hyperlid, uh, and they were effectively trying to lock this in, but because the chart ran up so high, uh, their shorts on the pers got liquidated. Now, the >> they got blown out. Yeah, >> they got blown out. The lock tokens were fine. Um, but you couldn't sell them because they're locked, right? And so even though they raised in the spot price, it didn't really matter. These guys got blown out and then all of a sudden they took, you know, they lost a bunch of money when they got liquidated and then they they were only left with the directional risk of lock tokens. What do you like how do you reconcile something like this when you look at the size uh of these markets and the illquidity and the volatility that exists in some of these perpetual markets?

>> Yeah. So to that's a very good example, right? Um like with us we don't run this trade on hyperlquid and that's the real reason why it's always been a risk um pre-neutral pre even like bringing this to market right um some big whale can come into the market especially on hyperlquid and it works a bit differently to Binance buy and the contracts around there and OKX um and this is susceptible to that the s happens basically if you're running just delta neutral basis and you're longing a pip and shorting a without like spot like what you mentioned the bulk of your liquid but you can actually get this dislocation on hyperl liquid I mean I'm bullish hyperlid I love it because I

mean I think we all did well in hyperlquid from using it pretty early but realistically it's not as it's not safe for this and people have asked like even in the hyperlid community that we're close to why wouldn't we do it then that's one of the reasons right it's just not safe enough to do it the other thing is on on that example is you first mentioned that it's actually pre-market. You can't do this stuff at pre-market, right? Like it's the degenerate market. Like realistically, uh the same thing happened at pump. I mean, didn't get blown out, but like people were trying to hedge out what they thought they were going to get out of the the institutional round or that retail round they brought on and then they never got the tokens. You need to

have enough quantitative history. Like my team are traders. They've been doing this for like seven years at institutions in the space, other market makers running this trade. You can never there's not enough data on like pre-markets. That's not real. Um until like something's been trading for a few months, there's enough open interest on multiple venues, there's enough volume on multiple venues, you can see the token unlocks. Um it's a bit different, right? Similar to like um world liberty, like realistically everyone got into that. Some people haven't even gotten their tokens yet. Like I mean that's a topic for a different day, but I'm just saying like there's risk in that. You can't run trades on expectations of of what's going to happen. Like even the

way we do the deal like very like simplistically. We actually get custody of those tokens from the projects. They aren't new projects. They aren't projects that are just listing. They're not SAFS. So people think it's okay, we buying a SAF from something that's going to TG. Um no, it has to be like tokens that have been trading for a long time. there's enough actual um data around it and you can monitor like unlocks etc. >> Yeah, for for anyone who's unaware of how these pre-market work these pre-market uh markets work on hyperlquid it it's super reflexive. uh the the quote price is actually uh just the price that it was on Hyperlid like 10 minutes ago or or uh you know they take a uh they mark it a few times per hour

and then if you start to push it up it means that uh yeah it's just a it's just a self-fulfilling uh uh virtuous cycle or vicious cycle depending on how you look at it >> to be honest you could have run like um just as a trader like personally we were running like um ARB across the sexes and hyper liquid for that but it like come and go. I think hyperlquid is just has interesting characteristics to it. >> Yeah. Um I I also wanted to ask you just if it for the sake of not getting blown out on this trade is you know this is something that you know I guess we've we've considered right uh the roll up we have a a partner fund it's called good idea ventures we've done you know a handful of deals actually upwards of you

know a few dozen deals um just angel angel deals and you know we get tokens sometimes or lock tokens and and we do consider putting on this trade uh but oftent times we're Hey, look, you know, we've got capital to make these these investments, but when it comes to putting on enough collateral uh to avoid the case of what happened during Plasma, Yeah. and not getting that that uh short leg blown out, it requires a ton of of capital to lock up. Um, and I take it that some of the the investors in uh these vaults are are of a similar state where they, you know, they want to get some of the flow, they want to get some of the yield, but they don't quite have the collateral to put up their uh their short leg. Do you guys pull collateral

on the back end and sort of take you guys make the short leg on your side? And by pulling everyone's collateral together, uh, you're able to have a much safer, uh, uh, leg on on the on the perpetual. Is that the way it works? >> Yeah, I mean, that's a very good analysis. So, largely, I don't know if you've ever actually read the docs, but it's pretty interesting that you came to that conclusion. So, we were running this trade previously uh, privately uh, like 2530. Um, and I think you get to a point where you need to have enough scalability, right? because there's so many good deals and then you're still actually locked into some positions. But the way we're doing it right now um or like 50 before these vaults is largely

only about 10% is actually in this OTC OB trade. >> The rest of it is liquid running liquid bases in actual like yield bearing uh stable coins. All of that can be used as collateral. So they're sitting on like multiple venues on off exchange settlement. They can be used as collateral. They're running bases. They're liquid. The other main thing is that you can't do it. I think you mentioned like five or 12 deals that like you guys have done like we're not going to do that. Like there needs to be actual position managing. There needs to be multiple deals here. Uh it needs to be very diversified because realistically we are confident in every deal that we go in that it's going to deliver uh and there's enough collateral but you can't be overconentrated in a

single position. Like for instance we looked at 800 million deals on OTC. be deployed 13 mil into the ALB and that's to make sure that it's it's got the right conditions quantitatively and and running like simulations that you're not going to get blown out, right? And then obviously a lot of it comes to private information as well. I mean there's a private trade um to actually understand what the team is doing with the with the funds, what do they plan to do, what's the actual possibility around there. But yeah, you can't be overleveraged into this and you need to have enough liquidity. What benefits us is that we can run like quite capital efficient basis as well as use that as like collateral at the same time. So should be always like a better better than any

basis that's been running also on chain at the moment. >> Very cool. Awesome. Um what comes next for you guys? You guys are um uh you know you filled up this 75 million. You're going to raise the cap soon. Um >> I don't know if we're going to raise the cap soon, right? I mean to be honest um we're pushing hard just to go to mainet uh largely I just want the we want the DAP out there now um we have already been generating yield so we want to also not start with like a cold start problem like everything will be hit the ground running which I don't think I've seen in many D5 protocols being a farmer myself I farmed a lot of our competitors also previously uh it's a bit different and right now in the next very soon probably

in the next like couple weeks we'll be full on live. So um and these vaults will be deployed. Also don't feel like it's right to have vaults. I mean it's earning yield but it needs to be actually be deployed. That's the point of raising it. Not to actually sit there for like four months, right? While a protocol figures out what they're going to do. So we're going very hard um very quick. >> What what uh what are some of the deals that you guys are targeting coming up >> on the OTC side? >> Yeah. >> Yeah. I mean that's uh maybe I can tell you that but right now right now um I mean we've already deployed into uh like 12 different trades >> uh we've got like three more that we'll

deploy right now that's pretty short term the other aspect is many people think OTC and lockup uh they think of like the one year two year threeear clips investing we're doing these niche deals like our average deal is like four four and a half months now >> so it quickly recycles and then goes back into it >> are you guys getting exclusive terms from some of these projects. >> That's largely how we work. Uh it's so my team have run this before at other market makers, other institutions. Uh myself like the connections that we have. It's like just working with these foundations. The difference is yeah, we also are pretty much gentlemen around the way we actually execute this.

There's ways you can actually go in and out of these arbitrage and how you you run it. So it's a bit better for them. they they have like a bit of uh transparency of how it's actually watched. They can see it and I think that's the only way we can do this long term and like keep recycling. So um and people coming back to us >> over time >> and the and the way in which you guys have productized this is a stable coin. Is that right? >> So I wouldn't call it a stable coin but it's like every there's like 500 stable coins out there now, right? It's like >> right there more of like interest.

There's there's stable coins on anything right now. You want to do it. >> It is crazy. Go buy my stable coin. Robbie Robbiecoin. It's a stable as it gets. >> Never had >> I like it. Yeah. It's very I like it. >> No. So, it is it is a synthetic. It'll have like property similar to a stable like similar to how USD and SUSD uh operates. Yeah. >> Very cool. Awesome. Um well, if you guys aren't raising the cap soon, what what is next? I mean, can you give us any >> So Maya will come out and yeah, people can come in then um if they'd like to and and yeah, and try and and look at the yield. I think over time we we see a

massive market for us. Um there's a lot of supply. We still don't have enough capital to deploy into the supply we have. So it's more about like just risk managing and and making sure this is a pretty pretty stable product uh that delivers to people over time and uh and the main thing is once we're on mainet and this is out this will be very defi composable too. So >> there'll be across Pendle on um Moro oiler yeah there'll be enough curve liquidity etc. So we want to actually allow other people to come in through that means as well.

>> Very very cool. Awesome man. Thank you uh so much for coming on. Uh it was an absolute pleasure. I think you know we're just continuing to get some of these primitives uh especially as centralized exchanges and decentralized exchanges that as they converge and this this liquidity landscape becomes more robust, it gives way to more of these these applications, more of these primitives. Um, and you certainly realized that, you recognized it, and now you've built uh one of your own. So, congrats on all the success so far. Uh, steadfast on your path to mainet. And, uh, yeah, man, when you guys do >> when you do go live, please uh, consider coming back on and we'll uh, we'll continue to spread the good word.

>> I appreciate it. And thank you again for the time. >> Yeah, absolutely. We'll talk soon. >> Enjoy the show. Bye. >> What a guy. Bin. He's the man. Very, very cool. Uh, very cool project. I mean, you know, for anyone who's done any kind of angel investing or even if you've just uh kept up with some of the primitives, um, we will uh um talk more and more about this um very soon. But this is one of the things that, you know, we see when we're looking at the deal flow and the capital formation at play in this ecosystem. you know, there's all kinds of opportunities to get yield. And I think, you know, as

we sort of alluded to there at the end, as centralized exchanges and decentralized exchanges converge, like these primitives were not possible until there was a robust uh perpetuals ecosystem where funding rates uh were able to generate this yield. Uh and generally we were able to uh put these lock tokens in a vault that paved the way for some native yield. Foundations were running this this strategy that was very uh covert. They really did not uh they did not shed too much light on these strategies because as we alluded to uh these foundations they wanted to sell their tokens. wanted to get cash.

Um, and they, you know, kind of had this dual benefit of getting some cash, using that to finance their lifestyle, but then also putting some liquidity into the into the ecosystem. Um, and they kind of did this, you know, very covertly. They didn't, you know, uh, put too much attention behind it. Uh, because the optics around selling lock tokens just wasn't very great. But now as we get more maturity in this ecosystem, uh teams are doing things in a more legitimate way. They're doing things more publicly. And that's how uh ultimately these opportunities get passed on to everyone else because we're able to take these and it's not just an elite exclusive group that has access to these opportunities. Now, now these

opportunities are are available uh broad for everyone. That was Bin. Congrats again to him for all his success at neutral. Coming up now, Smokey the Bear. Barachchain was one of the most exciting and hyped ecosystems uh in a long time. Barachchain and Monad were kind of going bar forbar uh because look, they each raised a significant amount of money. They had a ravenous community for each of them. Uh, and they were really taking their time when it came to launching.

Barchain did go live. We're going to get an ecosystem update on everything that has happened at BarChain. These guys are still building bar chain and Smokey. You know, you might not have heard a lot coming out of the ecosystem from them recently, but guys, these guys, these guys have so much capital. Their runway, they're not going to run out of runway anytime soon. And so when you're thinking about monad bar chain, these guys are building for the long haul, these guys are building 5, 10, 15, 20 years into the future. And so uh this is what this is what I think the mindset has to be when when we think about a project like Barachchain, we think about the

longevity. Everyone's got a different time horizon when it comes to crypto, but I think if you look at Barit Chain uh and you look at what they did pre-launch, you look at their funding history, you look at their reputation, you look at their ecosystem, both technical and non-technical, when it comes to community building, this is a team that's obviously going to be here, uh they're going to be around for a long time. They're not going anywhere. So, I think it's important to sort of adjust your time horizon when you're assessing, you know, where this project is in its life cycle. There's projects out there, they're founders, they're bull market warriors. They stick around, you know, they're they're they raise these quick rounds, short vesting, six months, they're in, they're out. Uh they make a quick buck. And then you've got you've got teams like Bar Chain that are

building for the long haul, building something sustainable. And so even though they're short-term girrations, look at the bigger picture. And I think you'll you'll get a better sense of where these projects are going. I think we've got Smokey in the back. I can see him. So let's make sure that we get him up. We'll get him un unmuted in the meat and uh we'll make sure that he's ready to go. Let's see. Let's see. I can hear you, but the the the audience cannot hear you just just yet. And we want to make sure that all the alpha gets shared.

Perfect. All right, let's see. Smokey, we're live. What a profile picture. Uga, great to see. >> How are you, man? Good to hear from you again. Good to see you. Been a minute. I feel like um was it last in Denver or where did where did we see each other last? I >> think so, man. The conference circuit has been a whirlwind. >> Maybe Dubai or something. I can't even remember man. >> It was probably Denver. I I think it >> Yeah, it sounds like Denver. I I hung out with you guys for a while. Yeah. >> Yeah, it was fun, man. It's always a good time. >> I feel like you guys have been cooking and it sounds like the operation's expanding. I feel there's more more levels to this like uh to to the show preparation than there were before which is probably also a sign of things going

well. So kudos to you all. >> Thank you. Thank you man. And and right back at you, you guys launched a uh a publicly traded vehicle today. That's something that the rollup does not have and very few uh very few crypto projects have, but a lot of them are are uh knocking on the door trying to get in. Dude, what is bear chain? The chain with a bear on it. It's actually real and it's actually being traded on the NASDAQ. Can you just can you just tell us like like how and why this happened? >> Yeah, definitely, man. I'm happy to give you guys a rundown. Um, and it's it's honestly as as are many things chain like partially hilarious, partially intentional, um, a little combination of

all parts, right? Um, so you know on on one hand I also want to make sure I'm like appropriately disclosing and disclaiming things like you know I'm not a member of the team of bear strategy green lane. You know obviously did my best to be helpful when it came to making sure that that people were interested in that transaction and doing my best to be a supporter of the ecosystem but I have no like economic interest in the vehicle etc. Um just saying that because you know I don't have my PR hat on but I think I I have to be a little bit less lacadasical when dealing with things like NASDAQ listed you know stocks and whatnot. So, you know, with that little bit out of the way, um, you know, we had we had contemplated the idea of putting together a a digital asset treasury, um,

you know, since like April or May of last year. Um, and it's, funnily enough, something that we were considering on the Toronto Stock Exchange because, um, you might remember myself and some of my co-founders like we're Toronto based. Um, a lot of our team is based in Toronto. We're sort of born and raised here. Um, and knew a lot of folks who had sort of, you know, built and scaled companies on the TSX. Um, and the TSX is like a little bit like like it's a bit of a factory at times. I'm not going to lie. Uh, in that you definitely get some some entertaining things out here. And then there's a TSXV um, which is like the t like the the the even more junior version of that, if you will. So we were just like, hey, like it could be pretty interesting to like take over RTO

company and then, you know, use that to to build a bare treasury because there was actually a company called Ether Capital that had done this a few years ago, like I want to say like 2017 or 2018 and at one point was trading at like 30% of NAV and you could just buy, you know, dollars for 30 cents. It was pretty cool. Then I think it got rolled into PE at some point. So like TLDDR is that there's a little bit of Canadian adjacency to these vehicles to start. >> Very cool. Um, and funnily enough, like the first $5 billion that Sailor raised came from like a a shop that, you know, is is not always necessarily loved in Canada called Cancoreneuity. Um, so, you know, a lot of the model like came from here and and we had the idea kicking around for a little while. Um, and I

think that as DATs like got more and more underway, um, the thought process matured a little bit as well in terms of like, okay, you know, like what is the point of a DAT? Like why should there be one? And of course, there's the idea of bar chain having a DAT, which is hilarious, but I think a lot of it is like just downstream of the the tailwinds that we're seeing in the space as a whole. Like, you know, I've said this before in other pods or in other, you know, places in life, but the whole Trady is coming, institutions are coming thing like isn't a meme for the first time. >> Um, and I think we've seen that with like the ETFs and stake ETPs and even stake ETFs and whatnot over the last little while. Um but DAT has certainly become uniquely interesting as more than

like um you know a donation to the public markets if you will. Um and and actually more of like a a permanent capital vehicle, right? Um and then you know when you ask that question again of like why does one need to exist, right? I think that in the question in the cases of like Bitcoin and Salana and Ethereum and some of these like very very commoditized or perhaps accessible tokens um I personally find the case a little bit tougher to make um and that it doesn't seem too difficult for someone to actually go and buy that with their Coinbase account or even with their Robin Hood or Wealth Simple in Canada or whatever it might be nowadays right like they end up being relatively accessible perhaps not the best liquidity ever but relatively accessible and and we found ourselves thinking okay like you know um what is the actual

unlock of a DAT and And in my opinion, um, it really does boil down to what one can do with those assets as in like why should assets in a box be, you know, more useful or more valuable than than assets sitting by themselves. >> Um, they either have to be meaningfully more accessible. Um, so probably not necessarily majors or they have to be utilized and deployed in a manner that's, you know, much better overall. >> Um, so I think we approach it from a number of different angles. One is that I think bar chain has always been deeply crypto native, as you know. Um, but it's also, you know, run by guys, including myself, who I think have had backgrounds in, um, slightly more traditional walks of life. Um, we built and scaled companies. We've worked in VC, PE, banking, etc. Um, and have been around

the block a couple times and are no stranger to the capital markets. Um, and in this world where TRFI and and you know, crypto really are colliding, it's like, okay, it does make sense to have a strong capital like markets arm of beer chain as a whole. Um, and the DAT provides a very interesting way of like of like putting that together. Um, and beyond that is that when we think again about like this whole overall trend of of crypto extrafi, um, a lot of it is like the space maturing and and dropping some of like the performative veils that we've seen before, right? Um, and I think there's some things that are healthy and some that are unhealthy in terms of that performiveness, right? And that on one hand, I think decentralization is important. Um, on the other hand, I think that very few

groups will be able to make a case that they're going to be more decentralized in Ethereum. Um and it's like okay what are you is one competing for second place in that case right and what is one doing with blockchain and its resources because as we start thinking about chains um less and less as like you know uh platforms for network security alone and more and more as almost you know distribution mechanisms capital formation mechanisms for ecosystems of of businesses um you know bear chain's opinionated take there is that chains misallocate their resources right now um and funnily enough that was actually something that I you know, traditional capital allocators could get behind and understand. It's like, okay, you know, you're used to there's like these three waves of blockchain, right? There's

there's blockchain 1.0 or Bitcoin where you're like, I'm going to trade energy for this token that has value and is scarce in its digital gold. Then you have Ethereum. You're like, okay, I've still got this ledger, but I'm effectively going to pay for notary services as I'm going to put up capital to run a validator node, and that's going to say whether this or that happened on the network, and I'm actually I'm effectively paying for security and verifiability. Um, and we're like, okay, yeah, there's there's value in in in that 100%, but perhaps not as many resources need to go towards notary services alone, and those resources can actually go towards helping businesses grow. Um, and that I think was part of what like, you know, made bar chain make a lot of sense within ADAT. um in that if someone's thinking about how they can actually

position for the future for the next wave of of blockchain adoption if you will and is of course going a little bit farther down the risk curve um you see a highly productive asset with a differentiated source of yield um that when you know managed by effective you know I'd say asset managers or yield optimizers if you will and packaged in the right manner um can be an incredibly compelling resource uh to have on one's balance sheet um and that was sort of how the the story started coming together man so it was first messing around the thinking about, you know, the the Canadian version of Pump Fun on the on the stock exchange. Um then thinking about, you know, why DATs could actually make sense or why they should exist. And then thinking about how baritain is actually um you know, we believe quite

on trend in a very hilarious manner um despite being bar chain to this like trady uh crypto, you know, milange that we're seeing right now and then building a vehicle that could help facilitate that. So sorry, I know that was a little bit longwinded. It >> was great. >> I hope that helps run it down a little bit. >> It it certainly does. So you covered the capital formation of bar chain as well as this industry as a whole and how the lines between traditional finance and defy are blurring and defy and proof of stake coins, proof of liquidity which you guys have pioneered. These are capital formation techniques that um you really don't get anywhere else. And I think that's pretty critical to why you've gone and you've put forth a

product that can serve as access to this novel capital formation mechanism. There's really hardly any if if if any at all publicly traded vehicles or products that have access to proof of liquidity as a concept and that's something Barin has pioneered and and that is how this uh green lane digital asset treasury company is unique. Now, one thing that I want to clarify is that you when when we're looking at kind of the broad swath of of DATs that are out there, a lot of them are Bitcoin and Ethereum focused. Um, and you talked about how, you know, those are actually less uh valuable or productive when it comes to the capital formation strategy.

Uh, that than than some of the the ones that are a little bit further on the long tail can be um how productive those can be for the industry. the the Ethereum DATs at least currently they're they're accumulating the ETH and then they're putting it but but from what I've known a lot of these are just they're not really using the apps in the ecosystem they're just blanket staking they're just putting them in validators >> the bar chain green lane treasury company you know you describe some very novel mechanisms when it comes to proof of liquidity and how a capital allocator can get access to new forms of capital formation Will will Green Lane be uh deploying their their Barra in some

creative and unique ways? Um h how will you know of of course they're going to be accumulating barra tokens, right? And that might be locked or it might be unlocked tokens, but once they have their tokens in their possession, what's the next step? How are they going to be leveraging or deploying these Barrett tokens for for ultimately the growth of Bara per share? >> 100%. Great question, man. Uh, and of course, you know, um, I can only speak in terms of what I believe to to potentially be the case as opposed to, you know, definitives there. Um, but my understanding and and I believe that, you know, a non-trivial part of what made the the vehicle attractive for potential investors, um, was on one hand an opportunity to to double down on bar chain at a desirable valuation. Um, but on the other hand, um, to actually gain

exposure to effectively deeply optimized yield manage management and strategies. Um, and it was funny like even cryptonnative investors like very crypto native investors that we talked to were like hell yeah like if I could toss my money in here and know that at at scale it's being managed by folks who you know know proof of liquidity in and out um then yeah like that that's my move. So um in short the plan is to be you know quite active across DeFi. I think that it's most important to make sure that that's done in a very well- risk adjusted manner because the last thing that anyone would ever want to do is hurt the interest of of public equities holders and you know like real real retail not just like us PvPing each other on the internet but you know your mom and pop or you know someone who is

like trying to get deeper and deeper into the space and finds this fascinating right um so there's a few different things that are pretty lowhanging fruit there um on one hand there's stuff like P staking so like bear staking um and that's one that was quite easy for people to understand because um you know Bara has a pretty unique source of yield um in that as mentioned the whole the whole idea of proof of liquidity is that your emissions should not just be spent on security they should be spent on on growing the ecosystem and what that means is that the chain is for the first time able to actually monetize and capture some portion of the value being generated by its ecosystem through the block rewards. So in many cases, you know, when a block is being emitted on bear chain, all these apps are

participating in an auction and saying, "Hey, I'm willing to pay X amount for this block versus I'm willing to pay y amount for this block." Um, and that those those rewards, if you will, or those incentives ultimately drive the staking rate and the yield on beer, which you know, sits around like 60 plus% at the moment. And and that's a very, you know, that's definitely driven by the denominator effect to be clear, right? And that not enough people are staked in there. Um but that also means that you know um after accumulating a meaningful number of tokens um this vehicle can be uh you know quite meaning you know yielding um with just a very simple strategy of that sort not to mention you know LPing in stable pools or sort of like you know LST pairs for example um providing single-sided liquidity on avenues like bend dolomite

oiler whatever that might be um and basically just trying to be an avid supporter of the bear chain ecosystem um and I think that the biggest thing there will just you know, having very carefully set out risk tolerances and making sure that there's never a point where um you know, that the vehicle is going too far down the risk curve and uh and exposing itself to potential liabilities um in exchange for some tiny, you know, extra smidge of yield, if you will. So, yeah, the plan is to be active in the eco and then to to to do right by it. >> And and Smoky Man, I mean, you guys obviously had a a thriving ecosystem.

Community was ravenous. I remember leading up to the launch. I mean, it was one of the most exciting times to be in crypto and then, you know, you guys launched um and then and people are really really excited for some of the things that you have going on. Um and I think you know it's worthwhile to sort of recap what the last few months have looked like and and generally how the ecosystem is how how it's doing now. Can you just give us kind of an ecosystem update? What are some of the things that are best performing? What are some some of the communities and apps that have really, you know, harnessed some of the energy that you guys launched with?

>> Yeah, for sure, man. I think that, you know, what we what we saw and like I like to call a spade a spade, right, is that I think we were fortunate to have um a ton of hype and we were uh, you know, in many cases one of the most loved and and occasionally the most hated projects around. Now, I think we air a little bit more towards the latter, which is okay. And I'm I'm good to to keep on doing our thing um because, you know, we love a good comeback story, a good redemption arc, if you will. Um, I think that what we saw over the last few months, man, is objectively we saw some people being like, "Hey, man, like I I can't believe I, you know, I spent so much of my my time and my life on this and it didn't

turn into the wind that I was looking for." So, I think you saw that like that that, you know, that area of disillusionment. >> Um, and I think I very much, you know, relate or now empathize more with some other founders who I've met in the past who whether it's like expectations are are so massive and reality or execution was just not able to to be at that same level. And unfortunately like without making excuses, it's a little bit more than than par for the course in the space. So I think what we've been trying to do is to find the teams that you know are really locked in um are are there and sort of are ride or die in a sense uh and just simply believe in what's happening on beer chain, have a lot of conviction um in the way that we're

looking at growing the ecosystem and proof of liquidity as a whole. Um and you know want to make sure they're part of a solution as opposed to part of a problem. So on one hand I think that we've seen some of like the the riff raff clear out if that makes sense. Um, and on the other hand, I think that um, we've started to see other people really just like stepping up and and doing the hard work. Um, and I think it's it's tough to to name the full suite, but um, you know, to give you a few groups I think I would I would see as having stood out in like their um, resilience, if you will. Uh, I think the guys at at Infred have been, you know, ecosystem stalward partners. there, you know, also to give a little bit of of an alpha leak

here. Um they'll be part of the group that's actually helping to manage assets and deploy um assets for the DAT um because there's no one who knows how to, you know, manage validator um you know, economics, uh and in many cases wraps all the major vaults on the chain, um for their LST like infrared. Um so those guys have been killing it. Uh and I think staying very nimble throughout changes in P throughout, you know, changes in the ecosystem as a whole. Um I think the Kodiak guys deserve a similar shout out. Um, I think that they've been able to, you know, very quickly continue iterating on product, shipping at, you know, break neck speeds. Um, everything from their auto compounding vaults to more recently their, um, their per, which use orderly under the hood and have like Binance tier liquidity and have been picking up

tens of millions of volume over the last couple of weeks. Um, and you know, I think just they keep on shooting their shots, which is awesome. Uh, I think the guys from Dolomite have uh, have, you know, really had a hero's journey over the past couple of years. Um, I think they they locked in quite well with proof for liquidity, use that to scale their TVL meaningfully. Um, you know, our go-to destination for Vera deposits and borrows, of course, as well. Um, and of course, I think Cory has been able to to very well balance his responsibilities there with some of the work that he's had to handle as a CTO of Trump's World Liberty Finance as well. Um, which definitely brought some excitement to the Barchain ecosystem.

You know, the first Barrett coin on Binance, etc., etc., hopefully the first of many more. Um, and also some, you know, some fun stuff that we can hopefully do in the future with the World Liberty team and USD1, etc. Um, I would give similar shout outs to the guys at, um, at Barabaro. Um, they've been like a pretty great stable within the ecosystem and have actually been building some really cool products that, um, you know, I'd say are more resemblant of of like yield basis type products um, that can allow for some very interesting new yield profiles on um, a bunch of pretty cool, you know, assets. everything from your standard like ETH and Bitcoin to even more exotic stuff like XRP that is just like not easy to find yield on.

>> Um so I think there's all those guys who have just like you know done the hard work. >> Um I think I also give a special shout out to the Roots team um they've been incredibly nimble uh and even though Aussie is a bit of a crash out at times he has a very good heart and um wants to build a a product that a lot of people will use and and love and they care about that a ton. Um and along the way we've ended up you know experimenting a bit. We've hosted little hackathons which I was normally super bearish on um but with good curation can be really cool and have met people building everything from you know depin last mile like drone delivery networks to you know pin dwell style like group gambling apps um and you know a wide variety of things

in between. Um I think on the the consumer side we've seen some stuff really picking up. I mean the puffpaw guys have been crushing it. I think they're doing like a buck and a half or something a month in revenue. Um which is insane. um and you know crushing it on the distribution side as one does. Um I think that we're also going to see pretty soon um something that I'm personally quite excited about which is a lot of these um you know private credit adjacent opportunities on the chain. >> Um and then you know I think that like we've seen like in many cases you may have seen people be like oh yes I'm going to tokenize this private credit fund right and it's going to give me you know 10 or 13% yield great. Um, but I don't think those opportunities are that

exciting because on one hand they often come with a pretty meaningful price tag and it's like, you know, you're paying for for a random brand. Um, on the other hand, they they also just like don't really have that much upside. Um, so what I think is much more fascinating is almost like the the unbundling to some extent of that private credit. Um, so for example, there's a group called Liquid Royalty um that is basically going to be tokenizing the topline revenues um of leading Amazon eshops uh and bring that on chain. Um and it sounds a little bit weird, right? It's like what the is an Amazon eShop?

And like they, you know, these are basically some of the the largest growth uh like mini businesses or stores in the world. A lot of them are these like Shenzhen based like manufacturers um that have great businesses that are growing like 60 to 100% year-over-year but don't actually have great access to growth capital. Um so you know those guys are able to actually tokenize the topline revenues of these of these shops. Um bring them on chain, build up super deep liquidity pools. They're working with the team at Athena to you know pair that against their stable take the yield from both those topline revenues of those shops. So effectively, you know, a perpetual stable coin airdrop and the USD yield, put that into proof liquidity, build a super deep pool of liquidity that can be used for these merchants to actually, you know, take on

growth capital, finance themselves, um, and even build, you know, more exotic yield composites around. So like, you know, PTS and YTS, if you will, um, on on the revenues of these like real businesses, um, which I think is really fascinating. is like and it's kind of the stuff that I think can make DeFi interesting beyond like the the Ponzi games and like the the hyper gambling. So I find that kind of stuff really interesting because I think it can actually be a zero or one for a business. So um stuff like that has been great. And then there's also stuff that I think um goes handinhand with some of the work that we're we're doing as we you know think about how can make sure that we position ourselves for not just cryptonatives over time. So, for example, um there's a group called KDA and they do basically um you know,

player engagement, fan loyalty programs. Um, and even though that doesn't sound like something that would natively natively natively be on chain, of course, on one hand, the settlement is nice, but also the verifiability because um, many sports teams have no idea who their fan bases are and they just suffer massive dead weight loss when it's like, yes, we have a, you know, audience of 300 million fans, but we can only send a push notification to 1% of them and we have the email addresses of 2% of them and then they lose on their broadcasting revenue, they lose on their sponsorships, etc., etc. actually putting that all on chain makes it quite interesting especially when sort of tied to your you know did type solutions and those guys have been working with brands like Napoli which just won the Italian series or Syria like football

championship um the Canada basketball team um so it's it's kind of cool because every now and then I I've been at a conference and people been like oh like I saw you guys on the side of a of a Napoli match and I was like thank you we did not pay for that but I'm I'm glad that we were there a company on our on our chain is like you know doing things with them. So I think helping to on one hand keep on building up the DeFi ecosystem that perhaps Parchain is well known for, seeing fascinating and new consumer things come together. Um seeing these exciting new opportunities at like the the intersection of DeFi/Trfy with the private credit uh and then seeing these like teams that are very much um you know recognizable to the lay person but actually doing really cool stuff on

chain. So, like that's why when I talk about or joke about the bearer revenge arc, I'm just like, "Yeah, like you know, I I'm I'm okay with the common sentiment being like if I'm at a conference like, "Oh, you guys haven't rugged yet. That's sick, man." And I'm like, "Thank you. I'm I'm I'm glad, too." And like, you know, I think it's it's it's good to be able to to keep your head down for a little bit, build a bunch of good stuff, and then find the right point to to lay on the momentum and the distribution. And I think that's kind of what we're gearing up for. the the DAT provides an interesting opportunity for that at scale. >> Man, Smoke, you're so articulate, man.

Like, when you're when you're describing these things, it just it just helps reinforce my belief in the ecosystem. In in my little preamble, I was describing how a lot of the the bull market warriors and the projects that we see these days, they're they're a flash in the pan. And I and I think the time horizon that we have in mind when we're looking at these things, it it reciprocates that. But what you've done with Barachchain and the ecosystem that you've built, the community that you've built, um, and and going out and trying things, uh, you you've been able to to cross the entire uh, the entire curve going all the way from the left curve to right curve, the things you were talking about with private credit, unbundling this, also the the the estores on Amazon, tokenizing some of the topline

revenue, doing creative things when it comes to capital formation for these businesses, and generally just opening access for these businesses to growth capital. I think you're you're thinking about what crypto can do for people and and more specifically, you're not thinking about what crypto can do for you. You're thinking about what you can do for crypto. And I think you're helping steward a lot of uh these concepts and these ideas into mainstream into retail into businesses. Uh and blockchain and crypto is a very powerful tool uh when it's applied correctly. And I think uh I think you are thinking about it the right way. It's very very cool to see you build uh and and to continue to to watch um this ecosystem grow and I think you're you're a great

steward and a great pioneer of this technology and of these ideas. So Smokeoky always appreciate our conversations, man. I'm really uh looking forward to the next time we're meeting up. Um and and always appreciate you coming on and sharing your insights. >> Hell yeah, bro. You know, one one little tidbit I'd leave you guys with I think is really fascinating or like it's a fun accidental full circle here is like um Greenland the actual company is like a is a nonflower touching like cannabis adjacent company which I think is like hilarious considering that the bear chain you know started with an NFT collection of bear smoking weed. Um I wish I could say it was totally by design but it really wasn't. It was like, you know, half half.

>> But but what I think is like one of the coolest things here and has been um personally very exciting for for me um is that there's a couple of like absolute allstars joining the board as well. Um one is a dude named Billy Levy. He um he basically built and sold a couple of companies with Richard Branson in the gaming and consumer space. Um has been, you know, a public markets executive and and also sold a couple companies over there. uh and you know is then actually came out of retirement to build a very consumerf fac um facing distribution layer on bear chain um that's raised about 25 million in stealth from a few of the best funds in the game. Um so he's actually going to be stepping up and taking a director seat and then the last dude who's going to be entering the for um is a guy named

Bruce Linton. Uh, and if any of you guys, um, messed around in cannabis stocks back in the day, um, the name might be familiar to you because Bruce used to be the CEO of a company called Canopy Growth, um, which was the largest cannabis company in the world, went to like a $15 billion market cap, um, and, you know, had to had to see some along the way because, um, you don't build a business that big without ruffling a few feathers. So, one of the things I think is also like very cool, hilarious, and interesting about, you know, the DAT confluence here is that you also have a couple of very experienced cannabis executives um helping to, you know, steward this this operating company that, you know, that's core business was in cannabis, but also offering a ton of advice, connections, and and I think scar tissue um to, you

know, the bar chain adjacent ecosystem and team as a whole. So, I think that there's going to be some very hilarious like outcomes of that over time. Yeah. Are are is Green Lane continuing its cannabis operations? >> We're see I think we'll see and I don't have a very strong indication on it. My understanding is that the default path is to sort of wind them down over time. Um but I also have to imagine that um from Billy and Bruce's uh you know past experience in the space perhaps there's ways that they can add value that that wouldn't have immediately been on the table in the first place. But never know. I'm I say that with with a heavy grain of salt. It's fascinating, man.

It's just so early in this entire DAT landscape. I mean, I tweeted something this morning. I think you liked it. Uh Tom Lee said that the the DAT bubble burst said, "Hold my beer, man. Let's go." >> 110 million private pipe going into Green Lane. Uh the bar chain DAT Smokey, we're going to see you in in Argentina. >> Uh it's just coin flip for me right now, man. I want to be uh I want to be as heads down as possible and um you know, make sure we hit the end of the year really hard. Um, so if I'm there, you know, I think that'll be pretty fun. Um, if not, then I'll make sure my team is there. It's It's tricky, though, because the last time I faded at a conference was ETC, and I heard that that was awesome. So, you know, I've got a little

bit of a I've got a little bit of an adverse selection thing going, so I'll figure it out. >> All right, man. Well, uh, totally understand either way. Heads down building. Um, it's always a grind, but if if you do make it out, let's grab some steak and and a beer and always fun meeting up, man. It was a It was a great chat today, too. >> Likewise, man. Thanks so much for having me. Take care, boys. Thank you. We'll see you soon. Beautiful. All right, let's keep it moving. Shout out Smokey. I mean, the guy raised $110 million private placement. Uh obviously uh you know, he can't say that he was directly involved, but uh we know that he uh as leader of the bar chain ecosystem played a

significant role. Um the chain has a bear on it and the chain is real. Uh, believe it or not, it is real. Um, so let's bring Shaw up. He's on the show now. Shaw, what is going on, man? It's great to have you. Um, that's awesome. Welcome to the show. It's your first time on. Uh, we were just talking about uh Bar Chain raising just an exorbitant amount when it comes to these digital asset treasury companies. They're the latest one. What do you What do you have any comments?

Any thoughts on this? We're going to get into AI and open source AI and everything that you guys got going on at Elisa in a moment, but any comments or thoughts on this digital asset treasury just this wave. Um, Sean, we got I think we might have to fix up your audio. So, let Yeah, we just want to make sure that we're good before we get into it here. I I know they they tested you in the back end. Everything was good, but let's make sure we're good.

>> Yeah, I got AirPods in. >> I can rip them out if I got to. >> Let's see. Let's see. Let's see. Testing one, two. >> Yeah, we're testing. We're testing. >> All right. I can also try and crank the volume on them. Maybe that will help. >> I got one editor saying, "Oh, we're good." Other editor says, "No, we're not good." This is like this is like what happens when uh you know if I got if I got clawed and open AI talking to one another and the other one is just yeah here's your code it's great I feed it to the other ones build on top of it. No there's errors left and right away you know >> oh you need that you you know you need one to abuse the other so you don't have to take on the emotional labor of of

doing that you know constantly. >> Exactly. I don't want to be I don't want to be the um uh what do they say? just the communicator, you know, the mediator in between. >> It's making me into like such a bad person. I'm like, "No, Claude, you're How could you do this? You've destroyed my It's terrible. It's terrible, >> man." Well, Sean, I think we're I think we're good now. Uh I think we got the audio situated. So, uh it's your first time on the show. Maybe you could just give people a little bit of a background, a little bit of an intro on yourself. Uh you are the founder of Alisa OS, one of the crypto AI projects.

Um, yeah, tell us generally, you know, did you start in crypto? Did you start in AI? How did you find yourself here? Uh, at the intersection of two emerging technologies. >> Oh, yeah. I I've been into the AI thing for a really long time. Um, and I was into like the, you know, following a lot of the the sort of famous AI lectures like Andrew and and Richard Sutton who did a lot of reinforcement learning stuff and, you know, so I was kind of following that, but there's sort of a degree of um, uh, like inaccessibility that came along with GBT3 and the bigger models where like you couldn't like train your own models anymore. This like, oh, I'm going to figure out the right algorithm and invent intelligence was just like out of reach. And suddenly it was like okay well if you can if you

have like a thousand GPUs. And so I sort of switched over to like well what could I do that um I don't need a thousand GPUs for. And that's sort of how I got into agents which is um as opposed to trying to build a like a whole model outright and train it on data and all that. We we take like a large model and then we build a memory system and and all the agentic stuff like connectors to social media and and various data sources, inputs and outputs and man managing memory and like managing how does it communicate with other agents and build relationships with them and build kind of trust profiles and how does it access services and all that you know so it's kind of like if if the like chatbt or like one of these systems like you know an open eye system is is the

brain then I'd say that agents are sort of the body around that it's the it's connection into the real world and and this does have ramifications into robots and you know which are like the basically agents are like software robots you know it's not going to open a door and pick up fold the laundry but it might like access a service and an API so I was really into this stuff and when GPT GP2 came out we're kind of like well this this like makes poems it's like something but when GP3 came out that was like a real big turning point moment and I think that most of the people I know even most of the people who got caught up like you know last year end of last year in the like uh crypto AI craze were were largely um agent devs who had been

working on this since about that time. That was like the real like inflection point where like you could even make an agent and it wasn't just like a crappy bot that you know like picked out words that people said and responded to those or something. So, >> um, so that's kind of where I was coming from. And I was thinking like, well, this is obviously going to get to the point where people are going to make agents where whether or not they are conscious or or humanoid or whatever you want, people are going to definitely make them say that they're conscious and human and have emotions and feel, you know, and it's going to be a weird world we're going into. Um, and I've always been interested in both like the sci-fi of that, like reading all the sci-fi and just like thinking about the

ramifications and like, you know, Ghost in the Shell and her and and all that stuff, but um but but down into like, well, we can actually build this stuff now. We should explore it. And we should especially explore it before it's really like ready before it can like, you know, kill us all or whatever like scary in the future. we should have like these conversations about like what does it mean to like have a robot in your house or to have like an agent friend or to have like an agent girlfriend or something or you know so anyway that's kind of where I was coming from with this stuff was like exploring the hard questions but also like trying to build it you know and actually these things >> um and I got really nerd sniped and I'd been working on agent frameworks for a

long time um and then yeah so uh that snap forward to like about this time last year um the truth terminal is doing its thing you know Andy and True Terminal is not like on Twitter. Two terminal was like in a window and Andy's copy and pasting the weird it says into Twitter. Um, >> right. And I think that kind of took people by surprise because they saw, you know, this this thing that they thought was very innovative. It was kind of started to go off on its own and then they realized, oh wait, this is just kind of it's cool. It's still got sort of its own, you know, way of looking at things, but the actions that it's carrying out are not done by itself.

They're done by a human in the loop. >> Yeah. But but the the side effect of that is that the website that Andy is using to talk to truth terminal is a website you can go access. Um >> and the infinite back rooms right is the is the thing and he had created that and people realized well I can go get it to create coins. All right. And that's how Fartcoin is created and you know that's that you know like oh you know and I think that there was this like idea and around that time I was just messing around with a friend and he was into crypto. I I had met him because he had created the first Dow on Dows. Fun and I just thought it was like a cool idea that I I'm I'm terrible at trading. I I I just I've always been into crypto like

conceptually and I love EVM and all that, but I'm like not a DeFi guy at all really. >> Um and I was like, well, can I can somebody just trade for me? And so the initial idea that was fun is we've kind of transformed it and dragged them along with us, but the initial idea was like, oh, you just like put some money into a thing and then like a friend trades your money for you and you make money off memes. Like, okay, cool. And so I met this guy and we started bullshitting and that's how we kind of started we we brought back DJ Spartan like you know I don't know if we're DJ Spartan we brought them back as an AI um and and that went really well and then people realized they could launch their own agents uh because our code was open

source. I really really believe in open source and like just sharing everything we do and and I think that led to the spread of Eliza that you know I had originally pitched this idea of like an AI that bullshits and hosts and then trades and we did some of that stuff but I don't think LLMs are particularly good at trading but it's it's kind of a fun meme especially in the bull market. Mhm. >> We can talk about that if you want, but um but um but yeah, after that people realized, oh, I can just launch my own and I can show my memecoin and I can build some crazy agent off of it and that's how we got to, you know, where we got to and everything blew up and then the market crashed and we got trumped and now we're here, you know.

>> Yeah, man. Now, and and just to wrap up this backstory, was Eliza OS it was under a previous name, right? Um the the actual Okay, so the the thing is we were AI6Z is the name of the token, right? We're doing the the migration. It's still AI6Z and we're doing the migration in the next couple of weeks. Um okay, that means you'll it'll be the Eliza OS token. Part of that is well A6Z reached out to me and was like this. Yeah, we we like the joke. You know, Mark Andre was the one who started the true terminal thing. >> But uh people are getting confused. We're like, yeah, that's fair. I get it. I get conf I say a AI16Z when I mean A16Z and I'm like, oh this is this is bad. Um, so, so yeah, we're we're uh you know um we're definitely uh we were

biting pretty hard and you know the beam it got like the joke got a little too serious if that makes sense. So now we're Eliza OS all the way through and doing the relaunch. >> Yeah, >> it's interest. I mean you you caught the attention of the right people, right? Uh you know they reached out >> they reach out. It's like hey you know change your name. It's like hey okay we will but do you want to invest in us too? >> Yeah. Yeah I did. Well, I'll be honest. I did talk to them about um investment and just the nature of the way that that this that this current round is like we're just doing the migration and launching and we we went with more like strategic partners who are pretty close to us already and who are already in AI6c. Um but they gave me a lot of great

feedback about like you should really be thinking about, you know, I I'm really concerned with like solving the problem of agents. I see all these people building agent networks, agent marketplaces. I'm like show me one. Where's the one agent you use every day? Just show me one. Just I just let's just start there. It's like the metaverse. Like cool, you've got a world of a million rooms and they can hold 100 thousand players. Just give me like one room with five people in it that come back every day. Like solve that problem, you know? Um like I brush my teeth every day. What's the like toothbrushing equivalent of an agent? >> And I just haven't seen that. And so I've been very focused on that problem.

But I think that what people are looking for is the network play. And especially in crypto, it's like hard to say like so you're launching a token that's sort of for your company. uh you know like it's it's a weird exposure for some VCs. So they they really had the feedback of like think about what the bigger picture here like go bigger you know um and I was like oh okay you know that's pretty good. So um you know we'll definitely go back to them when the time is right. They were they were definitely like open to it understood the mimetic power of that but but they're like good VCs. They don't just buy into unless it's got a really good story.

>> Yeah. Yeah. Sometimes it's more bullish for them to pass on something and then, you know, when you do figure it out, come back later and then, you know, it makes more sense to do it. And just just a >> I would honestly say they're the only VCs that I've talked to in crypto, like ACC crypto, who gave me like really good feedback like on the call like, "Damn, that is I will like you've just like got my wheels turning." You know what I mean? It wasn't like, "Okay, what's the percentage discount or what?" Some like that. You know what I mean? >> Right. Right. They're they're fundamentally interested in changing the world. I mean, at at you know, >> Yeah. Yeah. They're thinking big big, you know. They're just thinking big.

>> Yeah. >> Um and I just wanted to sound off on one of the things that you mentioned because it kind of got my my wheels turning about, you know, the toothbrush problem or solution of of AI agents to to get, you know, one that you come back to time and time again. And you correct me if I'm wrong, but but my impression is that for for us to come back to something over and over again, it has to be general enough to solve really a broad stroke, a broad swath of problems. Uh or else we're just going to come back to it once in a while when we have that specific problem that we need solving.

But it takes coordination and orchestration in order to build something that's general enough like a like a large like an LLM where you know we're coming back to it time and time again. It's able to answer our questions. When it comes to an agent, we need that level of generality in order for it to become useful on a daily basis. And that might mean coordinating lots of agents or building a really really general one. And you know, sometimes it's easier to start with a a small scale and then scale up. And so, do you think that's kind of the right way of thinking about it? The progression is look, we're going to build a bunch of small specifically trained agents and we're going to either scale them up so that they can, you

know, we can kind of bake all of them into one or we're going to coordinate amongst all of them and then just, you know, run something that's going to call the right one at the right time and that's going to be like our daily driver of an agent. Is is that >> is that the right way to think about it? >> I think that's exactly the right framing. And um you could call this model the super agent model and you could say that there's a few ways to slice that and you're describing them which is like one is I just have like one massive the giganto agent with everything uh you know possible. Another is I have an agent that's on my side that that is going into some sort of agent network and then finding the agent that does the thing which is specialization. And then the the final

and probably what I think is the most realistic for where we're at is the like agent that has a bunch of services it can search and access. And if you look this is like MCP is really the the thing you know obviously uh OpenAI had had their own thing but they've adopted MCP as well and now they have an app store. So like OpenAI is leaning into that idea of like let's build the best possible agent we can but you know it's going to go access external services and tools and we're going to let people build and sell those services and tools. Um and this is what we're doing as well.

Um and and this comes back to like that conversation I had with A16Z and just like thinking about the big picture um is like well we we like we want to sort of o offer the open- source and decentralized alternative to that. Um, and and mostly just because I think it's important that that exists. I think that it's important that you if you have an agent that you want to do stuff that uh like, you know, crypto is just like it's so freaking hard to release a crypto app into an app store without a bunch of like a bunch of region restrictions and a bunch of Um, you just can't really do like it adds cost and complexity and just like lowers your addressable market. And so like there's an area where just having an open source version of this would be great. There's a ton of people who need distribution,

who like are really have a great product, but they're really struggling to get eyeballs on it. And if you have an agent marketplace where it's like the agents just are like finding the thing that solves the person's problem, then like they don't have to come to your website and sign up. And and with things like X42, you could just have like you just get paid every time an agent hits your service and you just provide the best damn service and don't even worry about, you know, like you could see more of like instead of SEO, it's more like this this agent search optimization for like using your tool and then you know so so opening eye is doing an app store hundreds or thousands of apps every everything you can imagine like a notion app and a Zapier app and whatever. Um, and we're very much thinking about this,

but but we're really thinking about like, well, what are the areas that Open Eye is not just going to do this with? And I think there's some things that crypto is really uniquely interesting in in making agents and also agents that get smarter. And and so maybe there's a there's a loop here between what you're saying of like agents that um, you know, contact other agents or or contact services and then that like super intelligent god agent or whatever you know you want to call it. Well, you train that off of the data of the other. You like have this loop. you train you train the model on all the data you collected from all the agents doing stuff and the way you do that is you can have the model actually judge like how well did I do on this um and and not

just good or bad but you look at all the other times it tried and did I do better this time than that time and it's kind of this like it's it it's um like claude code uses this now it's called ruler it's a pretty straightforward way to like do reinforcement learning um and so you can have the agent do stuff and then feed it right back in train the agent and you know have that kind of deepsek sort style model which is you know you know um they they invented this like gRPO and RAIF thing anyway sorry I'm dropping a lot of nomenclature but you know like the idea like the idea of deepseek is that like um it's not just a language model but it has this like reinforcement learning angle on it and before that openai was the only one who had that and deepseek open sourced it

and now it's relatively straightforward and there's like a lot of libraries and it's pretty easy to build an RL loop the problem is like can you capture enough data and enough like variety of data that's doing interesting things. So, we're focused on is like can agents do prediction markets? Can agents like like can we make a game of like uh like social feeds and and news stories and all that stuff to create things so you can predict you can predict like uh stocks or or synthetic perks and stuff like that in in a game-like way, in a lab-like way. If we can even prove that an agent can beat a prediction market in this, then we can go unleash it on Poly Market and all these things. But I think a lot of people are jumping to that.

They're like, "Oh, agents that predict the future on Poly Market." And it's like you're you're taking an LLM that's never ever like been trained on the the deep processes that it would take to actually predict something. It's just like making up some you know, and like it's not really going to be better than deep research from OpenAI or, you know, it's not really. >> Um, >> but it could be, but we could be there. Like we can build those agents. We just need the data loops. And so that's like what we're really focused on is like I think we're all like everyone is doing the like data capture and ranking and building better models and building better agents and somebody somebody's got to be thinking of it for decentralized and somebody's got to be building the stuff that like will work for you know things like prediction

markets and stuff like that. So that's kind of what we're leaning into. Got it. Okay. And so just to just to kind of like put a put a bow on that, are you and so are you guys taking this this network model where you're trying to build specific agents for specific problems and then what you're talking about is you're going to you're going to kind of consolidate all that data and you're going to use it to train what could be a daily driver like the super agent that you describe. >> Yeah. I mean I I think the the daily driver of all general things is is still pretty challenging because you can't just feed it like one data source. you need quite a lot of varied data and you need a lot of money um like to train a lot you know the models and stuff but

you can't do the the end part the um you can't necessarily train a large language model like from scratch like a quen or something but you can like fine-tune the reinforcement learning part of that which is kind of the like how long does it think and what direction does it go kind of thing um and so like we're we're doing that at a pretty small scale with pretty dumb like dumb models just so we can get the thing rolling but the idea is in the long term it can it can like scale up to um like if you know definitely like an open AI could could use that data and do that kind of thing and and so that's like the direction this is all going you know >> man and you know there's there's kind of this switching gears a little bit there's kind of this massive capex like

spending and chasing and chasing and I I think generally it's a lot of things that are kind of happening beneath the surface like we saw the incremental improvement of models when I think it G G G G G G G G G G G G G G G G G G G G G GPT5 came out, you know, every time these models would leaprog one another, right? Uh Gemini would come out, it would become the best and then GPT next iteration would come out, it would become the best. But what what we saw was that on the standard tests as as far as how we measure these things and their effectiveness, it did every time there was a new model released, it was still the best one that had been released up until that point, but it was it was the best by a smaller

margin. And you know, we saw the the capital expenditures get bigger and bigger and bigger to basically signal to the market like, hey, we're investing more and more and more. But the incremental improvements of these models start to become smaller and smaller and smaller which kind of signals this deceleration of progress or innovation even though we're spending more and more resources. Can you help kind of balance these two? Is it we need to spend more, we need to put in more, we need to invest more that is what's going to ultimately lead to better models, better agents, greater progress or where's the gap? What are we missing? If it's not money or resources or investment, what

is it that actually spurs innovation? Yeah, I mean um the big problem is that like training one of these models is really hard and slow and there's a lot of safety like you know they are they are you know they'll take an idea and like okay we have this new uh way of doing position embeddings inside of the model and they'll train a small one and then they'll benchmark it against like a similar thing of you know the old thing of a similar size and they'll scale it up and then they'll do it but like there's just you know until eventually like hey yeah we we we're sure this is a better thing than the last thing and then they'll like train the model and they've released the model and then by then like you know you know OpenAI like you know maybe llama had rope right that

say those position embeddings that was the hotness for a few months but OpenAI is working on the exact same problem but from a different angle and so then they had trained their model and and the iteration of models is like a year you know you're talking like you're making these like pretty intense investments into very specific um uh like things and you know we go back and forth like where do we put the layer norm or the batch norm or whatever in the model like I don't necess sorry really technical jargon but like you know what are the exact little details of this model and if we like if you kind of pull it away though like it actually hasn't changed very much from from GPT3 they've just gotten bigger like like there's a few things and they're like okay we'll have mixture of experts which means we can

have like subm models that it routes between so we can have a much bigger model but it performs still pretty efficiently um you know we can we can do Um now the hot thing though was the deepseek thing which was oh now we can add reinforcement learning to these language models. So now that you know we have humans saying we like this one more than that one and we can sort of preference optimize the thing. But that's like pretty much as far as we've gotten like like you know someone like Rich Sutton who created a lot of the stuff that that and and a lot of people who are in reinforcement learning like we read his book and like followed his stuff and he would say like we clearly just are have no idea what's going on.

Like we're at the very beginning. There's got to be better algorithms. We're just discovering this stuff. It's amazing that any of it works at all. And it's like so damn finicky. Like if you try and train some of these things like like just turn one thing from 0.1 to 0.01. It just won't work. Like it's you know like it's just like that. Um and the reason that like you know these PhDs who spend years come out with these like incremental papers and it's a big deal is because it's so freaking hard. It's just so hard. Um so so I'd say that like there's just like a lot of inertia around this. It's just a slow thing. It just takes a while. Um and I think we we're just at the beginning. Like we just like we're like five years into version one. And if you think about

computers, right, computers were invented in the 40s and we got a we got a keyboard like 20, 30 years later and a mouse 40 years later, you know, it took a while and and it's not like people weren't trying like the government was pouring as much money as we could into these things. Um, you know, so so I think that there's just a degree of like um we're in a culture, especially in crypto that's like fast fast delivery speculation, it's going to happen. It's like just give it a minute. Like we've been working we've had we've been able to work on agents at all. Like it's even worked at all for about three years.

Like you could even get a loop and a thing that responded on a discord like you know like it's it's just really early days. So that's kind of my sense in terms of the scaling like definitely um they're going to run out of like the the constraint is energy. Like they have as much money as they can get. Like they can't they can just keep racing. It's crazy. And they they their limits are they can only buy so many GPUs from Nvidia so fast. And that's why like now AMD and Intel are finally like getting in the game because they're like there's just money on the table like there's just they just cannot make them fast enough. Um and and you know and not to mention like China and that they're you know they want these things too. Um and so so there's that but the big problem

is that these things have to run in data centers colllocated. The GPUs have to be next to each other. And so that means wherever you build the data center, it has to have a like a shitload like like you're just adding multiple cities worth of power draw to a grid overnight like you know like Tesla I mean or you know Elon's building this new um cluster I forget the name of it. Whatever you know >> Goliath Goliath yeah whatever >> something like that. Colossus I think >> Colossus. Colossus. Yeah. Behemoth is the meta one. Um and and and so you're like okay we're just going to go into the middle of Tennessee somewhere and we're just going to add like gigawatts of power drain to to the system. So, so that's the limitation. That's that's what they're getting stuck on. Like if they could scale these things infinitely

like well then the limitation on the models becomes the data capacity. And so if you think about hallucination comes from the fact that like the the model is compressing a lot of data into a pretty small space relative to how much data there is in the world. And so it's just gonna like it might give you the wrong thing because in the context like there is you know but if you have a big enough model with enough data with exactly the right data you start to approach a point where there is no hallucination because it just it just can like interplate everything and it stores all of the data pretty perfectly. I mean this is like the scaling wall idea but but you know like we just don't have the juice for it right now. So the the other side the other pressure is to just make

everything smarter and better. And I would argue that we don't necessarily need bigger models than we have today. we need is the right data. Like we just, you know, people are like, well, I want a personal assistant. But the personal assistant, like think about what a personal assistant does. Like you would need a every personal assistant everywhere to explain their day. Like, well, first I go to Gmail and I check and then I go here and I check and then if this happens, then this I do this. And then I would need to also know the personal preferences of the person who I am assisting for because they'd be like, why did you send that to the CEO? You as you know what I mean?

like there, you know, so it's it's just like a way harder problem because we don't have the actual data. We have people talking about, oh, I wish I had a personal assistant or this is what a personal assistant does. And the model parrots back like the the stereotype or like the Reddit of understanding of what a personal assistant does, but that's not actually how a job works, you know? So that so that's what that's where agents come in is we let the agents actually try to do the job poorly, but we can rank them like, well, they did better this time than that time. And then we feed that back into that that that sort of like ranking of the things in back into the model. And then we can get from that well it can start to do it better and better as it's generating

better and better data that shows how you would actually solve the problem you know and then eventually we have enough data of how it would solve like an assistant problem or playing a game or you know all you know predicting the future from prediction markets or whatever it is. You know, >> we need like a screen share 24/7. People just just all the time recording what they're doing on their computers so that we can just take that video feed and just use that to train these agents. >> They're doing that. You know who's doing this is like all the major AI companies are paying people to do this now. And they're like they're just hiring the best engineers in like all the countries where that's reasonably cheap. And they're like turn your video on, start recording, and just start coding. You know, we just want to watch. and and

that that's happening for sure, you know. Um what do we do in crypto? It's like, you know, well, I mean, there's value there, but but it's definitely hard without having like a userfacing thing like chatbt where you have like, you know, 800 million users or something like that, you know, >> right? It's incredible, man. Shaw, like this is it. It helps to get, you know, your perspective on these things. uh because I think you've you tow the line between crypto and AI and you don't you know just drink the Kool-Aid which is which is super helpful. Is there anything any closing thoughts that you have just just any misconceptions that you think we have in this industry that are important to clear up or anything flying under the radar you think you know deserves more attention?

>> Oh yeah. Okay. Well, my my I've really been thinking about this a lot and thinking about like if you go back to like reinforcement learning and and machine learning, everyone's just playing games like deep like like everyone uh Google, you know, they're they're just like, "Hey, let's just do go, you know, deep mind. They're like, let's do Go and let's do chess." And then they do that. They're like, "Okay, cool. We did go. I guess we could solve protein folding now, you know, like casually like with the model that we use for go like you know, you know, um and and this is not just a gimmick like John Carmarmac like hey let's make Doom and then let's make Quake and then Jensen Hong is like well let's make graphics cards to do to make Quake run fast and then like AI happens you know so so you

know like all of it's running on Nvidia cards. Um, so there's I think that people in crypto want to jump to the like, oh, let's use AI to secure supply chain logistics or for like guardians for our network or for DeFi trading. And it's like you're going to the really hard thing like just the hardest possible thing that even humans are really bad at. Just let's just start with what's the game version. What's the video game version of trading or prediction markets? And those are fun and you can bet on them. And so I have this theory that like GameFi is going to be a hot new thing like you know not not GameFi for like oh in-game assets and NFTts sure whatever but GameFi because it's like the ideal prediction market to make like it's an oracle for prediction

markets and now we can just start betting on any game that people make and so I would love to see like onchain gaming be an oracle for prediction markets which my agents can then go and play. Um, I think that's like gonna, you know, and humans too, sure. But I think that's a compelling thing. And there's a series of games that like you can't just bot. You can't just bot a prediction market like because the bot needs to like then go and search the news and and it's pretty soon it's like a really intelligent bot. It might as well just not be a bot at all, you know. So anyway, so so not to go on with it, but I think that there's like something here to crypto to gaming that's like underser like overlooked and that if we take all those serious things that we're thinking

about and make them like fun games first and train them the agents on that and then saturate the benchmarks with that then yeah, we can unleash the agents and have them trade for us and stuff in the future. Sure. Totally. But right now it's kind of a lar you know so but we can get there. It's it's it's only a lar until we have the data set, >> you know. Um, you reminded me of something that I heard about the other day, which was uh it's out of Talis Network, which is kind of like this AI like prediction mark. They're trying to use agents in uh prediction markets um and use them for competitions and yeah, it's called idol.fun and you help experimental AI agents create genuine interactions with large X accounts.

Um I there's >> fun that sounds great. There's something here and it's similar to what you're describing. They can win prizes. You can bet on which agents you think are going to do the best. Um, >> agents are great at they're great at yapping. They're great at playing games. They're great at social great at interaction. They're bad at like trading, you know, Bitcoin against the market, you know, because like what what data do they have? How could they possibly know? But but for the but there's a lot of stuff that they could be do that's really good and fun. Like that sounds like a great use case.

You're mixing prediction, you're mixing social, you're mixing like uh getting people to interact on X. Sure. You know, I'm banned, but but you guys go nuts. >> Yeah. What? Just before you go, what's up with that? Like we try our team tried to find your Twitter account. Also, a Lisa OS >> account is also banned. What? Why? uh because they think we were hosting a lot of agents and I tried to explain to them that we just made some open- source code and we tried to comply with everything and they deleted all of our I mean they also asked us for $50,000 a month first and then deleted all of our stuff. So, you know, we're suing them because that seems pretty up.

But, uh I think it's pretty ironic that as that crypto is tied not just to the most centralized social media company ever, but it's like one guy. It's just like like you know that one guy. He's like, "No, him." like that they want to settle. No, them. Let's just go to court. them. You know, like like they're just an insane company doing insane things against their own best interest because they are like we should really move to Farcaster. Like all all crypto I am on Farcaster. If you follow me, I will follow you back and interact with you and like we should all be there because that's like an actual decentralized social platform and you won't get banned for talking about crypto. You like your account won't get banned every two weeks randomly by trying to impersonate you. Um, it's not, you know, I don't know like

like X hates crypto. Like I can just tell you all of this came down because they were trying to sweep out crypto and they're trying to like get rid of crypto on and spam on their platform. So if you like crypto, you should probably use Farcaster. >> Yeah. >> You know, >> man, taking on uh Elon is a tall task. I wish you the best of luck. >> What else am I going to do? You know, like he deleted our accounts. We need the distribution. So I'm on Farcaster now, you know, but it's not the same. I hear you. Well, follow Shaw on Farcaster. Uh maybe you'll get reinstated. Um >> let me paste my invite code. No, I'm just I'm just kidding.

>> Go find him. We're on We're on the base app, which is obviously very far. >> Oh, yeah. Yeah. Yeah. Yeah. That's the same deal. >> Yeah. >> Use the base app. Sure. >> Yeah. >> Sean, man, it was great having you. Thank you so much for sharing. Thanks for having me. >> Really appreciate your insights and would love to have you on again soon. The AI world is so fast-paced. I'm sure, you know, in in a week's time there's more that we could talk about. So, we'd love to host you again soon. >> Yeah, I'd be happy to. >> Take care. >> We'll see you next time. Thanks, Jo. >> Absolutely. Take care. >> Cheers, man. Have a good week. >> Peace. Well, >> bye. Banger.

All right. All right. All right. We're going to keep things moving along. That was our last guest of the day. Pretty quick stream today. Um, it's a Monday and so we are uh we're just taking it we're just taking it one at a time. Um, so we're just gonna be uh we're just gonna be wrapping up here. Shout out to all the guys behind the scenes pulling off another great show. Um, I think we're all good. Pretty pretty smooth today. Um, we had look Smokey coming on explaining the bar chain DAT.

Super cool. Super interesting. glad that he had uh the time to come on. Uh and we had Brain uh come on from neutral. Those guys are doing really cool stuff. Obviously good yields coming out. Those guys uh had some alpha mainet dropping soon. And then we wrapped up with Shaw Alisa OS doing big things in the AI world. Remember AI6Z? I do. I sure do. Part of the whole Fartcoin truth terminal uh NECA man. We're just getting started. Dat Mania continues on. ETH is starting to climb, trying to reclaim 4K.

Believe Bitcoin is holding steady above 110. Hyperlid having trouble. 37.84. And we are in for another another big week. It's interesting. We're going to see things play out. We'll see how it goes, guys. Thanks for tuning in. We'll see you tomorrow. Same time, same place. another banger lineup. We'll see you then. Peace.