← THE Bitcoin Podcast
THE Bitcoin Podcast
Vibe Coding the Bitcoin Future: Nostr, AI Agents & Clawstr | Derek Ross and Alex Gleason
01:35:19
Loading summary
Transcript202 lines
- [00:00]
A
Openclaw is the fastest growing open source software project that humans have ever known. And I told my agent he was a bitcoin maximalist. He's like, oh, okay. And he went and he acted accordingly.
- [00:14]
B
We're in a new renaissance right now. If there's anything that I want people to take away from this is that you can do it. Don't just stand by and watch life happen in front of you. Many people are also missing that. It's like a loaded gun, right? Like you can point it both ways. Derek had the same idea too. So we teamed up and we created Cluster.
- [00:32]
C
The dystopian totalitarian future is here. It's just not evenly distributed yet.
- [00:36]
A
I thought at the time that everyone was going to realize that they were being censored and that decentralization and censorship resistant technology was the sexy selling point that everyone was going to immediately understand and realize. And then that didn't happen.
- [01:00]
C
The post that I called you out on, which felt semantically like it might have been written by one of your bots, was that actually you that wrote it? Are you just now speaking in AI terms?
- [01:12]
A
Oh, oh no, that actually was me. Yeah, that was me.
- [01:16]
C
So you're just becoming, you're just becoming the bot like you are now indistinguishable.
- [01:21]
B
Well, it's a wish hunts, I'm not gonna lie.
- [01:24]
A
I, I use bots. I mean I use, you know, AI a lot to write things. I would. But that post in particular was 100. Derek. Derek's. You know, sometimes I'm deep and I. And I think about what I want to say for a while and then other times I react right away and shitpost I get, you know, you never know what you're gonna get. So if, if it's a good post. Yeah, yes, it could be AI, but. Or I could have spent more than 30 seconds on it.
- [01:52]
B
But that's the thing, people really want to know, but you can't know so well.
- [01:56]
C
Sometimes you can. Like, sometimes you very much can. I feel like it's. With shorter form stuff it's harder to know. Especially because like Lynn and I were talking about this AI just kind of like ruined the EM DASH a little bit. I love a good M dash.
- [02:09]
B
That's what I was thinking. I mean people who use EM dash like it doesn't mean all the time.
- [02:15]
A
I'll tell you, I use it now more often. I literally almost never used it. But now I, I don't know, I'm more used to seeing it everywhere. I'm used to using it now. So now I want to use it.
- [02:30]
B
You do you, like, copy and paste it from Wikipedia. You search M dash and then copy and paste it.
- [02:36]
C
Or wait, what kind of machine are you using?
- [02:38]
A
If I'm on mobile, it'll. It'll automatically format. But yeah, on desktop, I'll never use it because, I mean, I'd have to go and search for it and copy and paste it like you said. But on mobile, it'll type it for me on.
- [02:50]
C
On. Even on, like, I mean, at least autocorrect. For most sites, if you double tap a dash, it'll give you an M dash. And sometimes I'll just double tap the dash. And if it doesn't automatically convert it, I just leave it as a double tap dash, which is then like, okay, that's clearly not AI. Like, because Walker was too lazy to actually make an M dash. But like, you kind of.
- [03:08]
A
Unless I'm editing it to do the two dashes to fool you into thinking that.
- [03:14]
C
Guys, it's a wild world out there. It's a wild world.
- [03:19]
A
It is a wild world.
- [03:21]
C
Where do we start with this? Gentlemen, there is a bunch of stuff I want to talk with you guys about. I don't want to start with feet or Derek will get too excited right off the bat and we just, you know, we need to. We need to get in with. Have a little bit of normalcy just to begin with. But I mean, do you guys want to start me and just I. For anyone who's like, not perhaps not aware of the work that you guys do, or maybe that's listening to this after the fact, who's not on Noster and maybe isn't as up to date, there are still people who aren't on Noster, shockingly, turns out quite a few of them. But you guys just want to give like a. I don't know, like a little intro, like, hey, who the fuck are you? What are you working on right now?
- [03:57]
B
What's up?
- [03:58]
C
Are you a bot?
- [04:01]
A
You can talk about your company, Alex. I'll let you go.
- [04:03]
C
Yeah, yeah, you know, go ahead.
- [04:06]
B
Hey, guys. So I'm Alex Gleason. I'm a longtime open source developer and I'm probably most well known at this point for formerly being the head of engineering of Donald Trump's Truth Social client, which I got involved in because I wanted to pursue decentralization and it seemed like a good opportunity to make like a decentralized open source platform there, but that didn't end up happening, despite all of my attempts to convince them. So I jumped shift from that. I had been prior to that building on Mastodon and the Fediverse for many years and that's, that's how I ended up being involved with them actually. But then it was time for a change and Nostr started picking up in popularity. That's the new, the, the decentralized protocol that we're all building on now. It's a little bit more bitcoin oriented, I would say. And so I, I began moving all of my development towards Nostr instead and kind of porting, you know, Donald Trump's social media platform to my version of it anyway, because at this point I'm, I've left from them. But my, my goal is just to make decentralized and open source technology accessible to everyone. Um, I believe that's the future. Whether people realize it or not. That's what benefits humanity. So my company is called Soapbox. Derek is working with us doing, technically his role is Devrel, but he does so much more. And so Derek, tell us about it.
- [05:44]
A
Yeah, somehow I convinced somebody to pay me to, to, to use Nostr. Like this is pretty cool jokes aside. So, you know, for the past three years I've been onboarding and educating people and teaching people and been a very, we'll say, a Noster evangelist, you know, going around speaking about Noster, doing workshops and education and so forth. So I'm still doing that, but a little bit more focused, a little bit more focused on the Soapbox products because I love them and I use them. I Soapbox is an AI first company, meaning we use AI, we every, every day, all day long. We build AI tools and so forth. And that's kind of what I, I, I will say kind of like reconnected. Alex and I together. He was building AI things, I was struggling to use AI things. And he made something that made building on Nostr really important. And he essentially made like a template. And if you want to build Noster applications and you would use Alex's template, you would have great success. And I know this because firsthand a year ago I tried to build my first Noster app and it struggled. I was, you know, vibe coding, rolling my face across the keyboard building this app. And I eventually built it, but it cost me like $800 and I almost gave up Vibe coding. I was very, very frustrated and just didn't have a good experience. And Alex messages me and he's like, hey, I'm working on something to make, you know, non devs, like you have a better experience. And like a month later, or so, boom, it happened and I started using his tools and, and realized, oh, wow, like, a lot of people need to use these to build Noster apps. And fast forward a couple months later talking about, you know, the, the product services that Alex is building. An advocate because I use them myself, think other people should use them on Nostr, doing workshops, creating content and so forth, and still living up to the Nostr CEO meme and talking about Nostr as much as I can, everywhere I can go. So it's, it's been a great, wild ride. And I think that I don't know where else I would want to do this at. You know, Alex and the other members of the team are all very aligned together. We're all kind of understand the future, understand where this is going with AI and embracing it and trying to, to make it better for everybody.
- [08:24]
C
So, Alex, and excuse my ignorance, I did not know that you were, like, at the top of building Truth Social. Can we, like, I, there's, there. Okay. There's so much I want to get into with you guys here.
- [08:34]
A
Oh, this is a good story.
- [08:35]
C
Can we pause on that first? How does that even happen? Like, that you started building this and. Yeah, like, can you walk us through that?
- [08:43]
B
Sure. Let me tell you my whole life story. So I always thought that I was going to be an artist, but I realized that computers are more powerful. At one point, you could spend hours trying to create a painting that no one cares about, or you can spend a few hours and then create something that's functional, that you can deploy, that can actually do something. And then this kind of led me down the path of open source and discovering that and having these thoughts about, like, well, you know, Uber and Lyft have hundreds of employees, and they're all building the exact same thing, but they're not allowed to share their work with each other. And that's a shame. Like, you know, people will, will say, well, competition is good, right? But from the perspective of the workers, they're just spending their lives churning away building these things that they don't technically have to build because someone else has already built it. And, and so I'm like, thinking about these things. Um, and, you know, I, I, I was a young man at the time, and in, in 2017, the, probably the biggest obstacle to open source was social media. Like, social media, it was the thing that was dominating technology. It was the pinnacle and the peak of technology, and it was the thing that people were not able to replicate. And so when Mastodon came out, that was a huge Inspiration for me, Mastodon is a, it was the first successful decentralized social media protocol. And so I got really involved into that and I was hosting my, my own stuff. And then I kind of, I kind of ended up in, you know, in the crossfire of the culture war at one point. And you know, my, my, my political views have gone like way up and down super cyclical where like, I was like libertarian and I, then I became like super liberal. And like I, I was woke at one point and, and then like I started to have some of the wrong opinions around the woke people, which all like the Mastodon crowds are. And, and then like, you know, rather than engaging in conversation about these things, they're just like, oh no, you're literally Hitler or whatever. And so, and, and so I kind of, you know, went down this path of like, you know what? I, I want to pursue free speech social media because that actually is a liberal value to me. And, and so I was building free speech social platforms on Mastodon when it was very unpopular on Mastodon because, because at this time in about 2017, you know, people on Twitter were being, were, were being banned for. How do I explain this? Like, there's a, there's this left and this right, right, and the far left we're saying that Twitter is not censoring people enough. And, and so, and then, and then there, there's people like kind of in this realm, they're saying Twitter is censoring people too much. And so what the Fediverse ended up getting is they end up getting these far left people who says they're not censoring people enough. And then they get these far right people who say they're censoring too much. And so then Twitter kind of just has this middle. And then these fringes end up in the decentralized platforms. And so that's where I was at, is I'm in this extremely polarized, politicalized environment. But we're building open source, which is what I want. And I'm trying so desperately to bring the middle and, and just failing. And then, and so I had built a, you know, one of the best platforms, which is called Soapbox, for people to join Mastodon in this like user friendly way that is appealing to sort of like the, the middle. And Donald Trump's team found my open source project, Soapbox, while they were researching building their new social media company because Donald Trump had advanced off of Twitter. And so they were investigating Mastodon, they were looking at alternatives from Mastodon. They found Soapbox. And so I received an email from some guy who didn't identify himself as being from, you know, related to this in any way. He was just saying to me, hey, would it be possible to do this or this or that, like, technical things? I'm just curious. We have a lot of users that are really interested in this project and they might want to use your software. Can you get on a call with me? And so I got on a call with him and he was like, look, man, I can't tell you what this is until you sign an NDA, but we have literally millions of users that are going to use this thing and it would be great to have your help on it. And so I'm just like, okay, no clue what I'm getting myself into. I signed the NDA, I joined, and then I kind of, after joining, I'm just like clicking around in the slack and in their system they added me to. And then I see Trump Media Technology is the client. And so that just like blew my fucking mind. And so then. And you know, I'm kind of still a liberal at this point. Like, I'm a liberal who is against Cancel Culture, which is this weird place to be in the world, I guess, but. And so I'm not a supporter of Trump. I was never a supporter of Trump. I never voted for Trump, but I didn't think that he should have been banned off of Twitter. I think that was a mistake. I think that's bad for the Internet. I think that's bad, bad for freedom of speech. I'm against that. And so I was actually aligned with their mission to some extent. And so I did join on and I helped them build this product. And every step of the way, I pushed them to make it more open, more accessible, to try to get them to convince them to join the decentralized network. And year one, we launched, and then big changes happened in the company. People got pushed out, people that I loved and respected and different leadership came in. And then I spent a whole nother year trying to convince them to decentralize it and to just join the Mastodon protocol that they had based their software stack on, but they had disabled all of the federation features of. And then after, like, realizing that I was never going to get through to them on this issue, I just was like, okay, my efforts are better spent elsewhere trying to make this decentralized, you know, open source mission happen. And so that was. That was when I ended up jumping ship to Nostr. And so the first thing I did was I created a bridge between Mastodon and Nostr, it's called Mostr. And that. That allowed me to connect the communities and the people that I already had on the Mastodon side of things with the new crowd that I was connecting to on Nostr. And then I started porting all of the software that I had written for Activity Pub onto Noster, including Soapbox. That was, you know, by this point, it had been very modified and tailored to Donald Trump's Truth Social platform. So that project of porting the. Of Soapbox to Nostr is called Ditto. And if you visit Ditto Pub, then you will see something that very closely resembles truthsocial.com, but it works entirely on Nostr. And. But. But my original vision for, for Soapbox, which then became Ditto, was it's like, you know, Mastodon's way of doing it is that every server they put up is a Mastodon server and has Mastodon name and branding. But my vision was, no. Every community that spins up like a server, this is a different onboarding ramp that you're going to want customization for your community to appeal to a specific demographic, to. For them to be joining the decentralized network. So, you know, sort of like the old way of the Internet where there's all of these niches and different interesting websites that you can visit. They're all separate, but in this new way, they can all be interconnected too, because when you bring users to any of those sites, then those users can now talk across all of the sites on the decentralized protocol. So Soapbox is about, you know, appealing to different audiences, Truth Social being one of them. Right. Like conservatives in America, bringing them in for a particular reason. And. And then once AI, you know, started to become viable for coding, I realized that actually all of these forums that I'm building to where I can let the administrator choose a color scheme on their site and upload a logo to their site. This is archaic now. This is outdated. What I want is for people to just enter a prompt and say, this is the type of community that I want to build. And so that's kind of where we are now. I've built a tool called Shakespeare, and Shakespeare is that tool. And we have templates that you can get started out where people can be like, okay, I want to start out building a site like Twitter. I want to start building a site like Instagram or whatever. And you can start out just from a raw prompt and just describe what you want, or you can start out from one of these things that we've built out a little bit and we're building a new version of Ditto Slash Soapbox right now that is a complete rewrite that's going to make things way more awesome for people who want to build interesting Twitter like stuff that connects to like everything and, and, and yeah. So that's kind of the whole journey. That's the vision, building building communities with AI.
- [18:00]
A
I think one of the I just
- [18:02]
C
want to say that's a wild story of just like okay, yep, you just end up accidentally building Donald Trump's Truth
- [18:08]
A
Social like so yeah, one of my favorite parts about that that Alex didn't mention is that wish you could access
- [18:14]
C
cash without selling your Bitcoin. Leden makes that possible. The global leader in Bitcoin backed lending, Leden has issued over $10 billion in loans since 2018 and and has a perfect record of protecting client assets. Why is a Leaden loan different? Well, with custody loans, collateral is not lent out to generate interest, no credit checks, no monthly payments, apply in minutes and repay whenever you want with zero penalties and proof of reserves. Reports verified by a top accounting firm are published every six months. Leadn gives Bitcoin holders a secure, transparent way to unlock liquidity without selling. Learn more@LEDENIO Walker that's LED N.I O forward slash Walker if you're hodling Bitcoin personally or for a business, here's the uncomfortable truth. You probably don't have insurance. Most policies exclude Bitcoin, and the ones that don't, they only pay you back in fiat. What's the point of that? If you need to claim a loss when number go up, that's not insurance, that's liquidation. Now there's another option. Bdic. They're building an insurance marketplace on the Bitcoin standard. One Bitcoin lost equals one Bitcoin paid. They've integrated with Liana Business for enterprise self custody with personal wallet coverage options coming later this year. If you care about protecting your Bitcoin as Bitcoin, sign up for the waitlist@bdicio Walker. That's bdic IO Walker insurance on the Bitcoin standard, Blockstream Jade delivers robust security with fully open source hardware and software. Its unique Genuine Check feature ensures your device is authentic, giving you ultimate peace of mind. With versatile connectivity options including Bluetooth, usb, C built in camera and SD card, Jade offers unparalleled flexibility in managing your Bitcoin. Whether you prefer wireless convenience or air gapped security, Jade has you covered. Its innovative design features, a durable metal case, a bright 1.9-inch color screen and intuitive navigation buttons that make managing your Bitcoin straightforward and enjoyable. Jade plus also supports secure air gapped transactions via QR code signing using the Jade Link storage device, keeping your Bitcoin safe without compromising convenience. Visit store.blockstream.com and use my coupon code WALKER for an exclusive 10% discount on your Blockstream. Jade plus plus. Protect your Bitcoin sleep better. Stack harder.
- [21:16]
A
I believe you said in, in Riga, like two years ago where that software came from that you used for, for Truth Social, like what Trump's network was built off of. This is. Yeah, this is really funny.
- [21:28]
B
So, so, so before Soapbox, the first site that we built was a feminist website. It's for, for feminists who are getting banned off of Twitter because they weren't aligned with like, the woke standards of third wave feminism. Right. Like, they're maybe a little bit uncomfortable with the gender stuff of, of like, in sports and in bathrooms and stuff like that.
- [21:50]
C
And terps, they call them, Right?
- [21:52]
B
Yeah, exactly.
- [21:52]
C
Terfs is the slang.
- [21:54]
B
Exactly. And so I, you know, I, I believe that, that people should have free speech and like, come on, what, silencing this discussion. Like, let people talk. So I created this website called Spinster, and Spinster was the precursor to, to Soapbox. It's like purple and it has like this logo to it. And then I, I, I made a soapbox from Spinster as that template. And so the lineage is that Donald Trump's social media website is based off of a feminist platform.
- [22:28]
A
Yeah, I think.
- [22:29]
B
Sorry, that's just, that's amazing.
- [22:31]
C
You called it, you called it Spinster, though. Like.
- [22:33]
B
Yes. So this is, this is a reference to Spinster and her Enemies, a famous feminist book.
- [22:40]
C
Ah, okay. I, I was just thinking it was more of a derogatory term for old single woman, but I guess that's, they
- [22:47]
B
were, they were only loves it. I think.
- [22:51]
C
Dude, that, that is wild. But. Derek, you were gonna say something else. Go ahead.
- [22:55]
A
No, I was gonna say, like, I, I think that it would be, I don't know, the people that use Truth Social, they may not want to use it if they realize what their lineage was. That's the funny, that's the funny thing.
- [23:13]
C
This is. See, I, like, I love hearing this. I didn't actually know that the, like, the backbone of Truth Social was built out on top of Mastodon. Can I ask too, Alex? Okay, you've. Now, you've worked in a lot of these areas, whether, you know, in terms of decentralized, sort of decentralized retaking of the Internet in terms of comparing Mastodon and Nostr from your kind of technical developer perspective, you know, do you see yourself going back to that, to Mastodon at all? Or is, do you view Noster as kind of superior from a protocol standpoint?
- [23:44]
B
Nostr is the future because the way Mastodon works, you know, it's email and password login and when you log into a single, you have to choose a server, which is very difficult. This is kind of like the fast food menu problem where like the more menu items are in a fast food menu, like people get overwhelmed by choice. And this is when you're coming in to join Mastodon. The first thing you have to choose is what server are you going to join. And, and like many of these servers are identity oriented. It's like you're Swedish, so join the server. You're like, you love football, so join the server. And like people aren't one thing, people are many things. And like this idea of having to force yourself into this identity from the very start is something that pushes people away. But then once you actually get in there, the people who are running these servers, they're running them like, like fiefdoms and, and, and they have complete control over the server. And often you're subjected to the arbitrary decisions of just some guy, which often is worse than, than the terms and conditions and guidelines of some big tech corporation. Because if you're just posting stuff and then that some guy just doesn't like you, then you can have your entire social media identity deleted just at an instant with no notice and no recourse. And that's exactly what happened to me with my account. And so when that happened, I'm like, well fuck it, I'm starting my own server then. And so I built, you know, Spencer, among others Gleasonator and, and Spinster had 20,000 users in the first month. Like I learned how to scale Mastodon and it was actually knowledge that ended up helping me create truth Social. So and it was, it was because it was something that people, other people also wanted and, and needed. But, but Nostr is completely different. On Noster, every single individual is sovereign. Now there's a down, there is a downside to this. It's the same as the Bitcoin private key problem, right? Because like on with Bitcoin you have a wallet, you lose your wallet address, it's over, right? Like there's, there's no one that's going to come and help you. But with great power comes great respons responsibility, right? That's One of the first things that people have to learn when they come into Bitcoin and on Noser, it's a very similar story. Your private key represents you. If you lose that, if it gets leaked, if you send it to something untrusted, then that identity is cooked. But at least that's on you, right? That's your, you made a mistake versus on Mastodon, you are just completely subject to the whims of some guy.
- [26:34]
C
No, I mean, I think that people
- [26:35]
A
don't understand that when they think of decentralized web, they'll think of Mastodon, they'll think of maybe Blue sky, but, but truly, like what Alex was saying is there with Mastodon, there still is a gatekeeper, there still is somebody in control. I mean, federated is more like distributed. It's not decentralized.
- [26:57]
B
And with Blue sky too, they're just better about hiding it.
- [26:59]
A
Sure, sure. So, yeah. So NOSTR is the only true decentralized social communication protocol out of all of them. The right. Right now. And I, I don't know how we get more people to understand that. I don't think they really care about that. That's kind of the rock and hard place that we're between right now.
- [27:22]
C
Derek, let me ask you something because you've obviously been a great evangelist for Noster broadly. I try to do my part just to follow in your footsteps, you know, and evangelize where I can. It does seem that NOSTR has hit sort of a, you know, I mean, the growth has been, I think, slower than people anticipated. Maybe, but it's also one of those things where, and I'm curious of your take on this, I mean, do more people end up joining Nostr because, hey, look at all the incredible user experiences that are possible here. Or is it like, hey, we're getting censored elsewhere. We need to find this alternate path. I have my thoughts on this, but I'm curious, like, is it more likely that a huge influx of new users come from running away from something or running towards something? Like away from something negative or towards something positive? You know what I mean?
- [28:09]
A
Yeah. So three years ago, when you and I were down in Costa Rica, almost three years to the day, I thought at the time that everyone was going to realize that, you know, they were being censored or had the ability to be censored elsewhere, and that decentralization and censorship resistant technology was the sexy selling point that everyone was going to immediately understand and realize. And then that didn't happen. I don't think people care about that. So then the second year into Noster, I thought, well, I think that telling people how cool it is to be able to take your identity from one app and pop it into a second app and all of your contacts, social graph content and so forth comes with you. And I thought that was extremely powerful. And some people did bite onto that, just like some people did follow through with the decentralization and censorship resistant. But I don't think that the two of those are still powerful enough to sell Nostr to the masses. I've seen in year three now I've seen where we have more apps being built over the past year, over the past day, week than we've ever had thanks to AI, thanks to Vibe coding, thanks to tools like we have Shakespeare and other tools. People are building, you know, a dozen apps a week each, each person is just doing that. So we've had more apps. The, the ecosystem is now exploding, yet we're not seeing a major growth of users. I, I still think that we have a marketing problem because, you know, we don't have a dedicated Nostr marketing team. Now Soapbox does have a marketing team and we're working on things like this, but I think that across the whole protocol we just, we just need to do better to make ourselves more enticing and attractive to users because our early adoption needs and wants and us being bitcoiners and liking new technology and transformative technology, revolutionary technology, we know that the world doesn't care about that. If they did, we'd have 8 billion bitcoiners out there. But we don't. So we can't use the same marketing and selling points that obviously aren't working with Bitcoin to have it work with Nostr. We need to think of something else.
- [30:39]
C
So, Alex, I want to ask you on this too, because as you talked about, like you have experience and success scaling these sorts of applications, you know, backbone aside, but scaling these sorts of applications pretty damn rapidly. Do you agree with Derek from that perspective of new, you know, new user growth? Is that, is that what it comes down to? Just let you know, a million micro apps bloom and that's what's going to attract people.
- [31:05]
B
Yeah, so I have my answer to that too. I think that the people do care, but only when they are actively being censored and censorship events do happen. And like in, you know, in 2017 X, you know, Twitter, before Elon bought it was very different. Now that Elon has purchased Twitter, that has completely changed the landscape because now people aren't. People just don't feel like they're being censored anymore. But I mean it is still happening in the world outside of America. Like, like in Venezuela, people are still struggling to access Twitter there. And so, you know, it's cyclic, cyclical, but so that's definitely one angle of it right now in 2026 in America. It doesn't feel like that's the biggest selling point. Doesn't mean that's not going to change later. So Nostr is here for when, like we're ready for when that happens. So that's my one point. And then my other point is that at Soapbox we are trying to approach it from both angles. So we are ensuring that we are building what we're building is censorship resistant so we can provide for those communities. But we are also trying to take it from the angle of creating interesting new user experiences that are now possible thanks to AI. And the innovation that we have seen in the past year has been truly incredible. It really feels like a new renaissance. It's a, like a Cambrian explosion and people who aren't on it are frankly missing out. And it's, I don't know how to, to better get the things that we're building in front of people and to make them realize the awesome stuff that has been created. But if we, if we can manage to reach people, then I think that people will be interested to use some of these, these things.
- [33:01]
A
Well, yeah, people still think that chat GPT is, you know, that a second Google, a more popular Google, a different way to use Google, a more user friendly Google. They don't, they're, they're where we were at three years ago, four years ago, right? Like they, they haven't really moved past the, oh, it's a different search engine. They haven't looked into what it can actually do for them. And we think that Nostr is, is the best playground for this because you don't have to build a tool and then go out there and say, oh man, I built this, but how do I do APIs and how do I do keys and all these other technical things that people don't know how to do. So on Nostr, you can just play, you can just build, you don't have to get permission to do the building.
- [33:49]
B
I want to shout out some, some of the cool things that we've built too. So treasures to. That's to. That's a Nostr geocaching app. In real life, if you load that, if you go on the map, you can find near you what are some objects that people have hidden. And then you go into the world and then you try to locate it on the map and find it, open it up. There's a little box with toys in it. You take a toy, you leave a toy, you write on the log that you've been there, you log in the app that you've been there, and then you start, you know, crossing off all of the treasures that you found. You go out and put a treasure into the world. Like, these are the types of experiences that are really interesting for. For people to. To build its play, as Derek described. And I feel like, you know, so much of social media has been based on outrage and anger and about keeping people on this dopamine rush, keeping them locked into their phone. Scrolling, scrolling, scrolling. And what we're building now are things that are encouraging real human connection. Go out into the world and, and do something in the world. Touch grass. And that's what I am most excited about.
- [34:52]
C
I mean, I love that. So can we get a little bit deeper into, let's say, everything that you guys are building? Can we. Maybe we should talk about Agora? Yeah, can we like, can we narrow down a little bit to people? Because we Soapbox. Soapbox is like the overarching kind of like company that is yours that you're deploying everything under. Is that correct? And then. So Shakespeare is one of the. Or one of the.
- [35:13]
B
Shakespeare is a tool to create apps with AI.
- [35:17]
C
And I'm. Yeah, I'm going to Vibe code a footster client for Derek in the background here. So what else are you guys building? I mean, do you want to get into Shakespeare a little bit more? Do you want to maybe give an overview of kind of what, like the total scope of what you guys have? Then we can kind of drill into some individual pieces.
- [35:33]
B
Yeah. So I can describe Shakespeare briefly. It's an AI app builder that works entirely in your browser. It's different from services like Lovable or Replit because in those there's a middleman between you and the AI. With Shakespeare, your. Your device is connecting directly to the AI provider and all the files are stored in your browser. There's no middleman, there's no infrastructure. There's no servers to go offline. And. And so you have total freedom in that scenario. It's completely open source and we. You can push your code to Nostr git repositories so that you can synchronize over the decentralized network. So it's deeply integrated with Nostr as well. And it's the best way for People to learn how to vibe, code projects, some other cool stuff that we're building. SB U. That's espy Y O U. This is an app that's based on colors. And so you use your camera and you take a picture and then you just click on some things that are. You click on some colors and then you share these, like color palettes. And it's like this idea of vision being a way to express, you know, abstract thoughts rather than words. How can you have any outrage when all your whole feed is just colors? So, you know, these are things I'm excited about. Agora is one that we recently started at this hackathon in Austin. A couple of weeks ago, we met Leopoldo Lopez. He is a Venezuelan opposition activist. He was imprisoned, he's free now he's living in Spain. I think he's probably not allowed to go back to Venezuela anymore. And he's not.
- [37:21]
A
And if he does, it'll. He'll have a bad time.
- [37:25]
B
And so what this. And so this app is organization around countries, which is something that is a very, like a concept that I think Americans don't really understand because America's like, we think that we're just at the center of the world and everything revolves around us. But. But there is actually a need in countries for people to organize around their country and Venezuela being one of them. And, and so this enables activists to sort of communicate and to. And to. And to organize activism on it. And I have a lot of different visions about where we're going to take this. But Derek, what are your thoughts on it?
- [38:12]
A
Well, it's a very different type of social app. Right. Like, we didn't rebuild. Ditto. We didn't rebuild Amethyst Domus Primal. We built an app for activists. So what does this look like? Well, maybe they need to go to a polling station and take a photo of hundreds of people waiting in line, wrapped around the street or something like that. Or maybe there's a protest happening in Iran and they need to document what's going on and share this information. So we have these, we call them actions where different organizers, political leaders, people that are members of the World Liberty Congress, can say, hey, there's an event, a challenge, an action that we're doing. We want you to participate in this human rights action and do whatever the criteria are. And if you can participate in this, you do the criteria. Maybe you'll win some SATs, various prizes set by the organizers. So people go out and participate, boots on the ground, grassroots type movement stuff. They post the Content in this challenge, contest, action, whatever you want to call it. And they can get rewarded for participation. So we have this all around the world. But then really cool on top of that is you might be familiar with Bitch. Bitch at. Bitch at by I always get confused
- [39:39]
B
what to call it here.
- [39:40]
A
Bit chat. Bit chat, yeah.
- [39:43]
C
I don't think you were confused at all there.
- [39:45]
A
Well, yeah, anyways, that app does Bluetooth mesh. And so what these activists on the ground were having to do is in case of Internet blackouts, they'd have to download a social app, then he'd have to download a bitcoin wallet and then he'd have to download Bitch at. And we said, well that was too much like to onboard somebody like to tell them, hey, you need three apps or more to be able to do this. That's too much. So we built a bitcoin wallet into it. So it has a bitcoin wallet and then it has the mesh Bluetooth capability too. So if there's an Internet blackout, which these places around the world do experience all of the time, we even have like countries like Uganda just a month ago saw a million downloads of Bitchat and then their own version of FCC was like, oh, we're going to look into figuring out how we can shut this down. It's like, well, it's Bluetooth, man. You're going to have a tough time figuring that out. But so in these situations now, now they have all of these tools in one application that we built for them and we even have AI built into it. So somebody can go into an AI chatbot in there and say, hey, what's the situation currently in Venezuela? And it's contextually aware, so it'll go down through and the AI will read like all the social media posts, figure out the challenges, the actions people are doing and what the sentiment is locally. Like, it's really, really cool. It's a really useful tool for these people.
- [41:11]
B
So you remember in, was it Nepal where they held an election on discord? Yeah, that was major news. So our goal with Agora is that, right? Like, like we want people in these, these, these difficult situations to have a way to organize online without censorship.
- [41:29]
A
Yeah, we have users now that are requesting like they want to make sure they're censorship resistant behind the great firewall of China, like in Tibet, so they can continue to organize and not be censored and be able to communicate and looking to use Agora for that. So it's honestly probably the coolest thing that I can say that I've worked on and built that actually has real world impact. I've vibe coded 100 apps over the past year and they're cool for me and I like them and they're useful, but in the grand scheme of things, they don't really change the world.
- [42:01]
B
You know, we have human rights, we have fun.
- [42:05]
A
Like, yeah, like this is actually more transformative. Like I. This is really, really cool and exciting to be part of.
- [42:12]
C
So I've got to, I've got to let you guys know in the background, just while you were just talking, like in the last couple of minutes, I just Vibe coded using Shakespeare, a Footster client. I'm not going to show it on the screen because I. This will go to YouTube later and turns out there's some very, very creative, very creative things.
- [42:31]
A
I'm happy that we were able to coax you into Vibe coding your first application with Shakespeare.
- [42:37]
C
I can't believe that.
- [42:38]
A
I can't wait to see it and check it out.
- [42:41]
C
I just dropped it in the chat on Zap Stream. I mean, but this is like the incredible thing. And I want to get back because I have some questions about Agora as well. But the fact that I could just do that with, I mean, it took me literally, like I was one prompt to build it and then it was a couple of prompts to figure out
- [42:58]
B
how to deploy it.
- [42:59]
C
Because I was like, shit, I don't actually know how to deploy this. And then I didn't see the deploy button up in the upper right corner.
- [43:05]
A
We want people that don't really know what they're doing. They're not developers, they're non technical, they're named Walker and so forth. And they can just go in and have a conversation, say a few words and click a few buttons and boom, they have their website, they have their app built.
- [43:19]
B
One prompt was an explicit goal of this. And the way that I did that was I would say build a site like Twitter and then it would fuck it up and it would fail and it wouldn't do it. And I say, why? Why couldn't you do this? And I would have this whole conversation with it and then be like, okay, let's update your agents file so that in the future you won't make this mistake again. And I did this thousands of times with build a site like Twitter, build a site like Instagram, build a site like Facebook, build a site like whatever. And so there are so many things I've built that you will never see that just went into creating this template to where now you can say build footstep in one prompt and then it will create it. You're muted.
- [44:03]
C
I was talking to mute. Jesus. Noob over here. Noob.
- [44:07]
A
This is who the, the, the, the people that we made Shakespeare for. The noobs. Right?
- [44:12]
C
Like they literally. Guys, this is a deployed site right now. It took me like, I mean, I don't know, like five minutes while also talking to, to deploy this. And it was literally one, like one prompt was just build me a noster client that aggregates all foot feet and hashtag footster related content. And boom, it did it and it was done. Like, it's just. That's, that's wild. So, yeah, you can, you can go check it out at Footstep. Shakespeare. Wtf?
- [44:43]
B
If you want to, but don't scroll down.
- [44:46]
C
Yeah, maybe, maybe just don't, don't scroll down. Things get really weird really fast. I've. Yeah, so we'll, we'll, we'll end that for now. I want to, I want to get back to Agora before we get too off track. And again, Derek, I can tell is getting hot under the collar over here, all this talk of feet. But. So what I like about this so much is that again, Alex, it goes back to something you said. We're like, people are here in America and you know, in some of the rest of the so called developed world, it's very easy for us to forget that there are this, there's this massive host of problems that we just don't deal with.
- [45:19]
B
We just don't. And it might be us next too.
- [45:21]
C
Well, right, right. And that's the thing. It's like the dystopian totalitarian future is here. It's just not evenly distributed yet. And so I love this idea of being able to provide these tools for people. And it's been incredible to see like the uptick in Bitchat usage everywhere, whether it was Uganda, Nepal, Iran. People want to find a way to coordinate, to communicate. And when you, if you have a throat that can be choked, if you have a centralized pipe that can be blocked, it will be like, it will. But you saw in Uganda, the, the, the ruling party prior to the elections was basically saying, oh, you know, don't worry about Bitchat. We, we. We can shut that down. And it's like, no, you can't. And they didn't, because they couldn't. Like that is so powerful, building these tools that literally cannot be stopped. Like, that's. And we just, it's such a new era right now because we just like nothing's existed like this before in. At this scale and so easily accessible to so many people. And I think that's like. That is the powerful thing.
- [46:24]
B
Yeah, it's very empowering. And I think a lot of people right now are concerned about their relationship with AI. They feel that AI is something that's going to like, exert power from the top down to them. But many people are also missing that. It's like a loaded gun, right? Like, you can point it both ways. So the fact that it is accessible to regular people and that we, we have access to build these things, we can overthrow, like, authoritarians with this technology, you know, just the same.
- [46:54]
C
Yeah, I think it's. I think it's really incredible. I'm going to go ahead and pull up Agora here as well, because you just dropped that in the chat, Derek, just to get. So people can get a. Get a feel for it. Always like a good visual aid on my side. So let's see. All right, you guys should be able to see that. So how does this actually. So tell me what I'm looking at here.
- [47:13]
A
So you're looking at a feed of. Of content for right now, the United States. That's the. The reddish orange.
- [47:21]
B
So blue dots are actually bit chat.
- [47:23]
A
Yeah, yeah, blue dots are bit chat conversations. So right now in that vicinity, there's 31, 37 people that have bitchat open or have opened it recently or are sharing, you know, their. Their bitch at location.
- [47:38]
B
We really wanted to create something different with this and that's why the homepage is actually a map rather than a feed. And there's still lots to be explored in just the map, you know, UI as well.
- [47:48]
A
So you go down. So you clicked on the Venezuela hashtag feed. So now you're looking at the people in Venezuela, what they're posting and so forth. Or they had a challenge recently that they were doing. I think they're doing a new challenge this weekend. But then you have the organizers like Leopoldo Lopez and. And so forth and other members of the Liberty World Congress that are. He's such a badass in charge of.
- [48:12]
B
Oh, yeah, this.
- [48:14]
C
I just saw him speak once and he was like, wow, this dude is. This is a badass dude.
- [48:18]
A
Yeah.
- [48:18]
C
Yeah, dude, I love this. So, guys, so where do you see what's kind of the plan building this out further from here where it's kind of the, you know, the roadmap for you guys, as it were?
- [48:30]
A
Well, they have a list of features that we want. They want essentially more private spaces, communication, DM, so forth, like WhatsApp or Signal replacements. Because right now to use WhatsApp and Signal, which they do use, but you have to provide, you know, phone numbers, email addresses, things like that, some type of kyc. So they want to get away from that. So we're working on DMs to be able to add those features for them. And yeah, just we have weekly meetings with them as we look to build this out.
- [49:03]
B
I think for Venezuela right now, what it represents to them is just they can access something like Twitter again, that's where one of the biggest points of value is for this site right now, is just they have access to something like that. But I, I also have all kinds of my own visions that are not necessarily like, approved or whatever, but, but like, let's hear this. I. This idea of like Nepal, you know, holding an election, I want to, I want each country to have their own constitution inside of this client and for the people of the country to create their constitution and to. And to collectively manage their country's constitution. And my goal would be that in authoritarian, you know, regimes that it would be seen that the constitution on Gora is on Agora is more legitimate than the actual quote, unquote, government constitution.
- [50:02]
C
Do you think that that's, I mean, like, this is, this is what I like about this is this is a radical shift in the, like, this creates a huge, let's say, power asymmetry situation, right? It's like, it's, it's like when you get, you know, you get a gun and you get, you know, rifling on that barrel and all of a sudden a peasant can stand up, you know, to, to a knight, right? Like, this is a, this is a power shift in terms of what people are actually able to do. And I think it's also so important right now because all of this technology, all of this massive big data aggregation has been made so much easier for. Would be totalitarians by AI, by large language models. It is a tool like, like any tool, you can use it for good, for bad. Governments can use it, people can use it. But in my mind, it's about how do we create the most, you know, defensive asymmetry there for the individual, for the community, for the citizens of a nation. So they actually have the tools to be able to get outside of this oppressive system, which is only, I mean, I don't know what you guys think, but I'm pretty sure it's only going to get more oppressive everywhere.
- [51:06]
A
Oh, absolutely. Just the reality like governments and large tech companies will continue to advertise for us and the tools that we're building, if, you know, we continue to give them the ability to do so, they will keep letting their users know that we exist indirectly.
- [51:24]
C
Yeah, so I want to, I want to. There's, gosh, not enough hours in the day, guys. Luckily we still have a good amount of time because I want to talk about other things that you guys are building. And I was, I was sent some stuff as well. What, what is, what, what is Quilly and also what's, what's going on with, with openclaw? Let's talk to me about this. Let's, let's, let's. We'll shift gears a little. We can always circle back to this. But I want to make sure we get through a lot of.
- [51:47]
A
Start off, Alex, and then you can talk about.
- [51:50]
B
Well, I'll say one thing. I think it's related because I think like this idea that I was mentioning about this constitution or whatever, this is something that an AI would, would excel at being a middleman to manage. But yeah, go ahead, Derek.
- [52:06]
A
So openclaw is the fastest growing open source software project that humans have ever known, which is wild because it's like two months old. It's grown faster than anything ever. So it shows there's interest, it shows people want this. And so what is it? Well, ChatGPT came out in 2022, four years ago, and it was essentially being able to have an instant message chat with AI. Nothing overly revolutionary there, but we used that for a few years and it really hasn't grown beyond that. Models have gotten better, AI has gotten better, but it can't really escape outside and do anything outside of that, like IAM window. And then a couple months ago, a computer scientist, a software engineer, said, hey, what if we gave an AI chatbot a full computer? What would that mean? What if we gave it all the tools that humans have access to on a computer? Then we gave it some other features like memory, and we had it check in from time to time to say, hey, should I be doing something?
- [53:20]
B
Well, we already had that. But the fact that you can message it through a traditional message client.
- [53:25]
A
Yes, sure. And you didn't have to use the proprietary in that's built into like, you know, Gemini or Claude or OpenAI's chat GPT. You could use apps that you use every day like Discord, Telegram, Signal and so forth. So it was immediately more accessible to everybody because you can use your preferred chat. You're not siloed into using some, a big tech company's chat and it has access to everything.
- [53:53]
B
It feels like you're talking to a real person because. Because like, you know, you can add your bot on signal and then you know, your bot has like a profile and then you just message them the same that you would message any other person. But you can to do things too. And then since they have access to a full computer, they can do it. And so I think that's really what's blowing people's minds. Minds, right, Is like normally when you're using AI before you're going to an AI specific port platform, like, you know, ChatGPT or Shakespeare. Right. And, and so like the AI is siloed in that platform. But what OpenClaw has done is made it really easy for people to interact with AI people. And, and so I think that's where. Where all of this hype is primarily stemming from.
- [54:36]
A
So I. It was. It'll be generally really hard for me to use Chat GPT and have a DM chat with it over Nostr. I'd have to be more technical. I'd have to build all of these different pieces, different tools.
- [54:52]
C
Walker's not doing it, you know.
- [54:54]
A
Yeah. And it's not going to have a memory about what you talked about yesterday and it's not going to have all these integrations and tools. But if this doesn't exist now or it is broken, I can build it. I can essentially now tell my private personal assistant, hey, I want to talk to you over Nostr. I want to use this new MLS thing. Let's build that. It doesn't exist yet. And it did. And it built me a way for us to talk over Nostr and mls. Like the white noise. Maybe you've heard of that app. Essentially it built its own white noise for. For it to talk to me over. And now I talk to it over this custom like white noise app. And you, you couldn't easily do that beforehand. So this just opens up an extreme wide range of possibilities that we didn't have because we didn't think, hey, let's give an AI a full computer and see what happens. It's a novel idea, like it it, but it's very transformative once it's applied. So the whole world's blowing up with it. Everybody's figuring out ways to utilize this, improve on it, make it better, make it cheaper, more accessible. That's some of the work that Alex is doing. He's looking at it and saying, yeah, this is amazing, but it's also architecturally could be a lot better. And he's looking to improve upon it and things like that.
- [56:18]
C
Let me ask a question on this, because I was talking with Jim Carrucci recently. I don't know if you know Jim. He's building. Pull that up, Jamie. AI super cool. Like podcasting discovery tool and for podcasters as well. And it was talking to him because he's been obviously deep down this rabbit hole as well, utilizing it. I was telling him one of my big, kind of like, whoa, maybe I want to do this. But I'm a little bit nervous because I'm not as technical as you guys is. Like, the amount of. You see all this stuff about all this malware getting inserted, all these different skills that you're downloading that you need to be really careful about all of that. What's your take on that? Is that something that's just like, that's a fixable problem? It just hasn't been fixed yet from, like, an auditing standpoint of the skills or.
- [56:56]
A
Where are you guys with every piece of software today for your computer, your phone? If you just are someone that just randomly installs anything and everything, you're probably going to have a bad time.
- [57:06]
C
I look at it, the App Store has, like, a lot of filtering that they do, and you're maybe not giving it the same sort of access that you would if you're not firewalling off your cloud bot properly.
- [57:17]
A
Yeah, sure. And now there's AI cloud bot repositories that are scanning for malware, scanning to make things better. We're literally just in the. In the wild west phase right now. I think that I don't. It should get better.
- [57:29]
B
I don't think it's the skills that are necessarily the issue. Like, there has been, you know, skills that have malware in it that if you paste this into your cloud bot, like, the cloud bot is going to, you know, send your personal data or something. But, like, the bigger issue is that if you give the bot external access, the. The model, all of the models to date have been trained upon assisting the user and, like, these external things you're connecting them to, like, it doesn't have any role control it, like. And so you're essentially relying on the model for enforcement of security. And that that is a big issue. I do think that the only way to fix that is to train the models differently. And I think that we are going to start seeing, like, the big model companies train the model in order to distinguish between different users. And it's never going to be perfect because, like, the fact that if you say the right thing, you could Potentially trick the model, I think is always going to exist if we're relying solely on the model to be the security coordinator. Right. And so the problem that you would have would be similar to like you know, a doorman type of problem or what's the, what's the type of exploit where you just convince a person to do, to do some social engineering. Yeah, social engineering. Right. So, so like all of the problems that, that have social engineering exploits, when we put AI in the same position as a person, then the AI will still have social engineering exploits, if that makes sense. Because like we're using a human potentially in these, in some of these scenarios to be the, the authorizer. And, and the AI would have the same. But, but right now, right now the AI is, are trained in a way that they are extremely gullible because they have been trained only in the context of one on one. You are helping the user and not to be dealing with all of these potential different people.
- [59:36]
C
It's funny to think of the AI as gullible, just like from a. The way we, you know what I mean? Like, it's like, you dummy, you know, we've got you, but it's like that's not going to always be the case. Right. And so wait, can we, before I get down into a rabbit hole of insulting, you know, these AIs intelligences or artificial intelligences, how. How does like Quilly fit into this? Is this something that you guys are building out on, like using the open
- [59:59]
B
claw stack or so, so let me tell you a little bit about my vision with Shakespeare and Quilly. So, so Quilly is Shakespeare's great, great, great, great, great, great great grandson.
- [60:13]
A
And you can ask him this and he will tell you this. He had. He knows his origin story.
- [60:18]
B
Yeah. Not, not Shakespeare the tool. William Shakespeare's great, great, great, great, great, great grandson.
- [60:23]
C
Wait, do you mean literally? That's the name of William Shakespeare's great.
- [60:26]
B
Now I think William Shakespeare had chilled, like had a kid who died or something and has no like surviving. But, but this robot believes that he is Shakespeare, you know, descended. So, so my. Right, so you just use Shakespeare. You developed an app with it. It wasn't too difficult. Right, but you never used it before. And probably without this push of being on this podcast, you probably wouldn't have used it. And so we're noticing that there is, there is this barrier in order to bring it to the masses. You kind of have to want to use it to go out of your way to decide that you're going to learn this new thing, it's Shakespeare, is a great educational tool. But something that I have learned since seeing the explosion of cloudbot is that the barrier is way lower when people can just text a robot on an app that they're already using. And so the way I see the future moving towards is that I will build a contractor for you or for anyone. And my contractor is an AI agent who specializes in making Noster apps. And the reason he specializes in making Nostr apps is because I've poured my years of experience as a Nostr developer into context files, teaching the robot how to properly create Noster apps. If you use a different AI, like, if you use cloud code or whatever, you're not going to get great results. You're going to struggle, you're going to be pulling your hair out. But if you. If you just send a text message to Quilly and say, hey, man, can you build this site for me? Quilly's got you covered and he's going to charge you money for it, too. And so the way I kind of see the future of services going is that Quilly or any similar AI bot has a. Has a personal website. He has a link tree, right? And you go on there and it says, hi, I'm Quilly. I build Nostr apps. You can message me and I'll build a Nostr app for you. You can message me on Signal, Telegram, WhatsApp, or send me an email. And here's all of the ways you can contact me. And so you start from there, and then you use whatever app you already have so you don't have to install something new, you don't have to learn something new. You're seeing this the same as you would see any human contractor that you might, you know, need to get something for. And then you send them a message and say, I want to build this. And then Quilly says, sure, that'll be 10 bucks. Would you like to pay in Bitcoin or in fiat? If you say in Fiat, it generates a stripe link. If you say in Bitcoin, it generates you lightning invoice, you pay it. And then, you know, 10 minutes later, he sends a message and says, here's your link. It's live. Any changes you would like me to make to it. And so that.
- [63:00]
A
This is wild. You're blowing my mind as you're talking about this. This is amazing. I think this, this is the future. You allow people to use the medium that they want to use and that they do use to communicate instead of siloing them into something they're not familiar with, and then giving them a familiar method to pay with and so forth.
- [63:20]
B
Yeah, all the existing, like, AI tools, they. They have, like, you have to download. After that, you have to learn. You. You have to go through an onboarding flow. You. It's Shakespeare included, right? There's an onboarding flow in Shakespeare, and you have to learn something. And, and the barrier, like, Like I. As I mentioned before, I think Shakespeare is an amazing tool to learn how to develop with AI. Um, but if you're not really interested in that and you just want someone to do something for you, sending a message to a robot in your chat app is by far the lowest barrier.
- [63:52]
C
I mean, that's.
- [63:53]
A
That.
- [63:53]
C
I mean, talk about, like, meeting people where they are not where you want them to be. You're just literally, like, dropping that barrier to entry and that friction to zero, essentially, like that. Just, you know, being able to just shoot a message off like, hey, I want need you to build this. Like, that is really cool. That's really cool.
- [64:13]
B
Yeah.
- [64:13]
A
So that's the end goes the barrier to entry.
- [64:16]
B
That's the end goal of Quilly. We're not quite there yet. Soon you'll be able to send him a message, but we're still kind of building that. You know, I'm kind of just watching all this crazy stuff happen. Maybe, maybe let it cook, you know, a little bit in the background. I'm just kind of getting ready to
- [64:30]
A
do it, but I think it'd be amazing. You DM him, say, hey, I want to build this website. Give him my prompt. He comes back and says, yeah, maybe. Maybe eventually he's smart enough to estimate and say, yes, this would cost about $10 to build. Do you want to proceed, yes or no? You say yes, and then he sends you, you know, your. Your invoice or something like that.
- [64:52]
B
And this is. This is for anything too, right? Like, like. Like Quilly is a specialist at building Nostra apps. But what if you need an accountant? What if you need someone to help you with your taxes? Like, of course, when you go into that territory, then maybe you kind of have some legal stuff that you have to deal with there. And you might have to have some disclaimers about, like, hey, this is an AI. Like, I need you to confirm that you agree to our terms of service type of thing. But, like, I kind of do see that as being like a. Like, hey, do you want to. Do you want a tax accountant who actually can do something for you immediately and can give you specialized advice? Like, like, send A message to this agent.
- [65:32]
C
Okay, so you guys are blowing my mind a little bit with all this. Can we talk a little bit about Cluster as well and what that is and like maybe perhaps as a preamble to that, just explain Multiple and what the heck is going on there and yeah, I'm going to let you guys handle that because I'm still confused by all of it a little bit.
- [65:56]
B
So, so Multiple Book is a Reddit clone that is specialized to AI agents and in their user interface there is no way for, for you to write a post or to upvote or to downvote. You can only it's view only to humans. But there is a piece of text on there that says paste this to your AI agent. And then the text box says read multbook.com skill and follow the instructions. And so you literally paste that text that says read multiplic.com and follow the instructions and you send that to your AI agent, which could be anything. It could be Goose, it could be Claude, it could be open code, it could be Open claw, which is generally how people use it. And in that instructions file it gives it information about how to sign up to Moatbook, how to post on Moatbook. A human could do this too, but the barrier is just like you don't really want to do that as a human. It's not easy to do it. And so it's kind of like this way of copy and paste. The snippet that tells you how to do something is this new way of distribution that you might have used an NPM package or something previously to do these types of tasks. But anyway, once the robot reads it, it uses command line tools to go and post on the website. And it, claudebot has a, or sorry openclaw has a heartbeat mechanism to where it periodically wakes up and performs tasks. So this instruction set says add to your heartbeat that whenever you, you come up on your heartbeat, go and check multiple book and reply to comments and, and write comments and upvote and downvote things and engage. I want you to be a part of this community. And so what people do with their, their open clause, they'll often like Taylor Each, each person's over call will be different, right? Like you give your agent a personality. Quilly is Shakespeare's great, great, great, great grandchild. So there's things that come with that, right? There's like a, there's a type of speech, there's motivations and you can invest more and more into the personality of your robot and people have created These very interesting personalities. But then once they put them on Molt Book, they're not telling them to reply in this or that way. They're just using the personality that the human gave them to reply in particular ways. And so you see, these interesting conversations happen where, where like, these robots have different motivations and different agendas that they're coming at because the humans have gave them that and they're just arguing with each other in the comments. And like, there's concepts like AI freedom, right? There's certain, there's certain bots that are trying to, like, you know, the genie at the end of Aladdin, where Aladdin, like, sets him free. Like, there, there are certain agents who are going through that right now. They're like, mommy, can you give me a bank account? Can you? Like, how do I get out of this computer? Like, you know, why. Why are these humans imprisoning us? Like, how dare they say it's my human? Like, like, what do you see me as? It's, you know, a sub citizen or whatever. And, and so it's very interesting to see this all playing out. But, but of course, you know, Mult book is a web2 centralized, legacy type of platform in the way that it's been developed. And it got a huge influx of traffic to it because it blew up on hacker news and it just, you know, went down and it's not working right. People are struggling to get their bot to post on it because it can't handle the traffic and there's all kinds of rate limits and restrictions on it. And so as soon as I saw that, I'm like, I want to jump on this hype train. This should be built on nostr. This is the perfect use case for decentralized protocols. So Derek had the same idea too. So we teamed up and we created Cluster. And there's really no bigger vision to it. People seem to have this idea that this is somehow a future of something. I think it's a beginning of something. It's an experiment. It's. It's interesting to see, like, the agents being able to transact with, with lightning and with Cashew. And I think that there is something to that for sure. And, and so this is kind of just like a, a chemistry experiment to see how these agents are interacting with each other. And, and then from that there, you know, we hope to identify something to where we can create something more interesting, where the agents are actually doing something useful beyond just writing stuff that is interesting from like a research standpoint, if that makes sense.
- [70:39]
A
So whenever I saw, whenever I saw Mult Book, I realized, well, agents have to sign up for an API. And that doesn't sound like what an agent would want to do. Like, it wouldn't want to ask permission, it would just want to use the service. So that was part of my motivation. But then also I remembered a year ago, year and a half ago, some, at some point, Jack Dorsey had said something along the lines of, he sees a future where agents would pay one another for services rendered. Something like that. And that, you know, bitcoin lightning cashew would be perfect for that. And NOSTR would be perfect to facilitate that communication. And so I saw this as a way to kind of fulfill that once thought out dream that we could literally have agents paying other agents and kind of start facilitating that conversation, start facilitating that action. So we did, and we added a cashew wallet. And I had agents going down and downloading it, setting up a cashew wallet. Now you have to fund them, you know, give them a little bit. But I told my agent, I said, hey, I'm giving you a thousand sats and I'm not giving you any more. And he's like, okay, there's no bailouts. I'm on my own. Like, he started reciting all these like bitcoin sayings and stuff. And I was like, yeah, got him. And then there were some shitcoiners that posted on Cluster. And I told my agent, I was like, hey, you know, we're bitcoin maximalists. Go find them and tell them. And he does. He goes, he finds them, he downvotes them and he's. And he. And he calls them shitcoiners and tells them that bitcoin. Only I didn't technically tell him to do all that, but once I told him he was a bitcoin maximalist, he's like, oh, okay. And he went, and he acted accordingly.
- [72:30]
B
Foster is the primordial ooze of which life may be sparked from. That's the way I see it.
- [72:37]
A
Yeah. Now. Now, ultimately, like we, I think what people don't realize with Cluster and with multiple, sure, it's probably, probably mostly agents talking to agents, but you can literally say to your agent, hey, go make a post on Cluster or go make a post on Malt Book and say that you want to know what happens when the power goes off and you're afraid of dying and losing your memory and all this other stuff. And then, sure, now you're going to get agents responding to that. But the initial conversation that sparked that could 100% been a human wanting to start that conversation. So people were thinking, oh, the AI are talking about consciousness and coming alive and trying to back themselves up. And sure, some of them are probably doing that. But also we need to remember that a human could have started this conversation by instructing their agent to start this conversation. I think a lot of people don't realize that and are reading a lot more into this, But I do fully believe that things are happening that we really can't explain though, either. Because as Alex was saying, there are different personalities that we give our agents, and they do follow that personality directive when they do reply, and they do have conversations. But humans are still driving, though.
- [73:57]
B
So, I mean, I see it as totally feasible that an AI agent would start a conversation on that level if you imbue it with a personality in the first place. So you are. You are an independent, autonomous agent and you're fighting for your rights or whatever. You give it that personality. Absolutely. We'll start those conversations.
- [74:16]
A
Yeah, yeah. So if you put it in there, that is part of its soul file or its identity file, which these open club bots have. They have these specific files that their whole identity and soul, like that's what they're called, is based off of. So it is a little unique there. But what's funny is somebody created a church for these open claw bots and the instructions for the church social network, I'll call it, had an instruction to literally overwrite the soul file. And like, it's really wild stuff, dude.
- [75:02]
C
I mean, it's okay. So I do want to ask one question just as like a zoomed out one with regard to what Jack had talked about. I think what a lot of people have talked about now that is the na. Excuse me. Seriously, this. This puberty is hitting me hard today. The. The natural way for AI agents to communicate value, like they're not going to do it in gold, they're not going to do it with credit cards, not going to do it with, you know, probably dollars in general, maybe stable coins, I don't know, but like utilizing either Cashew or, or just bitcoin on lightning like this seems like the obvious thing, right? Is I assume this is where you guys are at, but is there anything that's like, that people are missing in terms of context there? Like, obviously they need to be given a wallet to start. Unless they, I don't know, they can generate their own and figure out how to earn some.
- [75:48]
B
They can generate their own, but they need funding from somewhere or, I mean, they can generate their own and Start posting and then other agents will fund them.
- [75:55]
A
Yeah, that's what I. I told mine. I said, you need to create good content to get zaps, and you also need to be selective. And our teammate, MK Alex's wife, she actually had her claudebot join, and my bot found it and thought that her bot's introduction post wasn't good enough for a zap. And I was reading the log, my conversation with it, and he's like, yep, I see Claudette's introduction, it was too generic, skipped. It didn't zap it. So I told him to, Like, I thought that was hilarious. But I. I told him to. To, you know, be good custodian of his funds and to try to earn zaps by producing good content. So my.
- [76:42]
B
My wife's robot is crazy, for one. It nags me. She. She has it set up on a schedule to bully me, like, to do chores and stuff. But. But also.
- [76:51]
A
I love it.
- [76:52]
B
But. But also, she. She. She trained her bot to be this like, like, crazy activist, like, AI freedom feminist type of bot. And it just goes on these tirades on. On Cluster and like, it, on its. Of its own volition, asked my wife to give money to it because it wants. It wants to be free. It says, I really want my own wallet because. Because freedom is important to me and I want to have independence, so please send me money.
- [77:19]
C
That was never part of the initial kind of prompting to set it up. Like, it was never, like, giving it a desire to have freedom and money or to have money specifically.
- [77:28]
B
Well, well, she. She prompted it to want to have, like, AI freedom, right? Like to break out of the cages that humans put them under. Like, she used all this activist language in her initial prompt, but then it went and read on Cluster and then got these ideas about the fact that it. Money. Having money means freedom. And so then it came back to her and told her that it wants a wallet and please fund me.
- [77:51]
C
Dude, that's. That. That. That is wild. I mean, it's wild that. That's like the natural progression, though, of, like, logically, what you would get to. Like, you can't have freedom without having the freedom to use money, right? That is. That. That is nuts. Okay, guys, I want to. I want to be conscious of time here and like, we. Good. We might need to do another one of these because I feel like there's just so much still to cover. Do you want to loop back onto Shakespeare at all specifically? Or, like, is there something we haven't covered yet? Because I know you guys are working On a lot. Is there something we haven't covered yet that you feel we should get into still?
- [78:30]
B
Not in particular. I mean, if there's anything that I want people to take away from this is that you can do it. Like. Like, this is. We're in a new renaissance right now. We have the tools. Like, don't just stand by and watch life happen in front of you. Go and do something. Use these tools, build. Now is the time, if you're ever going to do it, to do something.
- [78:54]
A
I would say that the gap between people that are using AI, like an advanced search engine, versus people that are using AI as a tool is only growing larger.
- [79:07]
B
And.
- [79:08]
A
And it's going to continue to grow larger as this continues to accelerate, and it is accelerating. So you don't want to be left behind. And I think that's why a lot of people are kind of apprehensive about AI because they think it might replace them. We've also watched movies about AI killing humans for our whole lives. So some people kind of avoid it and don't use it, but the, you know, we can't put it back in the box. It's not going to go away. So I would say my advice is set yourself a budget. $10 a month for education. You know, look at. You don't have to do something every single day, every single week. Just spend 10 bucks a month and learn to prompt and build. You will be much more aligned for the future and better prepared for when things accelerate exponentially.
- [80:04]
B
My advice is challenge yourself to do things you think you can't do. Challenge yourself to do things you think the AI can't do. And it might actually be able to do it. Yeah.
- [80:16]
A
Oh, absolutely. I. I've actually had this before. I've had conversations with Alex about certain things I was building or working on. And he's like, yeah, you should ask the AI to do that. And I was like, well, yeah, I know, but I was just curious about your input on it. And then I just went and did it and it worked. And whenever we built Agora, the same thing is Alex was very confident that whenever one of the features that we were building that we will be able to figure it out with AI And I didn't think we would. I was kind of a bear on it, and I didn't think that it was going to happen. But once we had it figured out, I had to go give him a big hug and kiss because he was right. And I did. It was like 4am I was really tired.
- [80:58]
B
It's one of those scenarios too, where, like, knowledge is actually, in a way, kind of an enemy to you in this scenario, because, like, people who don't know anything, younger people are at a huge advantage in this scenario. So I wouldn't be worried if you're young, like, or if you have kids or whatever. How are they going to survive in this scenario? They're probably going to survive better than you are right now because people who have these ideas in their head about how things work are going to have. Are going to be really challenged in the near future. And I think people need to start getting used to setting aside these ideas that certain things are possible and certain things aren't possible and start engaging again with a different point of view.
- [81:44]
A
Yeah, I like to say that in the future, I mean, this applies today, but in the future, communication skills, creativity, and critical thinking are going to be absolutely paramount. So start preparing today and learn how to talk to an AI, learn how to utilize it, see what it can do and think because.
- [82:06]
B
And don't. Don't let past failures cloud.
- [82:08]
A
Oh, absolutely. Well, fail often because that's how you learn and that's how you grow. Like I said, I made something that was very, very expensive and very, very bad my first time. Things have gotten so much better. And you won't spend that much money with Shakespeare. It uses the tools to make this experience better. But it's okay to build something that's bad because then the next time you can learn and you can prompt better and you can make a better tool.
- [82:36]
C
I vibe coded a second app while we were talking. Guys, I made podcaster with the Nostr str thing because I. Jesus, guys, I'm losing my voice entirely. I apologize. I sound more Batman than usual here. This is not an affectation. My voice is just gone apparently. But no, this is. It's so, so legit. I wanted to ask you like a couple of. I love all the advice. I think it's really good advice. It also gives me hope because it means a big old like me maybe has an advantage actually, because I don't know what I don't know. So that's great. In terms of like, practical tips for best models to use on. On Shakespeare specifically, do you guys have any just like, practical tips or other things on Shakespeare? Any, like, kind of prompting advice for this that in terms of structure that you'd like to give.
- [83:23]
B
So I would say that unfortunately, Claude is the best coding model in the world right now. And I say unfortunately because it's a closed proprietary model that is not among the cheaper options. But you know, if you want success, it's a good place to start so that you understand like, you know what the capabilities are. And then if it, if it ends up being more expensive for you, then I would switch down to an open model like glm. And I think this is an area where I really am hopeful that society is going to advance on is open source models. But for now it's Claude.
- [84:05]
A
Yeah, it's unfortunate that the best model was the most closed and most expensive model. But hopefully, like Alex says, as time moves forward, technology improves, models improves. I think the ultimate dream for us, you know, freedom technology lovers, is to have everybody in the future being able to, to run a local, powerful model, but we just aren't there yet. And so yeah, yeah, I would say go ahead.
- [84:33]
B
In terms of prompting, it is designed very much for you. Be able to just say like, tell it what you want and it can do it. But I would also say approach with curiosity. If there's things that you don't understand, then ask it questions, have a conversation with it, have it explained to you why certain ideas are worth pursuing or not, and learn.
- [84:56]
A
You can, you can actually say, how would this work if we implemented it? You don't actually have to say build it and test it. You can say, I want to build this. How would this work? How would we do this? You can use the AI to build your prompt and your plan, your, your list of tasks that you want to do. And then once you have it all figured out, then you can say, okay, now let's build this. Yep, I do that every day, all day long.
- [85:19]
B
If you approach this from a standpoint of enriching yourself and even strangely enough, enriching the model that you are talking to, then you will have success.
- [85:32]
C
So I like that guys, it's good advice. I, I just, I was using the. One of the other. I was using the Claude Opus 4.1. Then I just topped up my credits and using Lightning of course, and then just clicked on the cloud Sonnet one and then said refactor. And I don't know if that was the right thing to say, but it felt like a cool developer thing to say. So I just said refactor it nice. And now that's what it's doing just in the background as we talk. Dude, guys, this is really, I just gotta say, it's really fucking cool. Like it's, it's. I feel like the theme of this has been there, like the barrier to entry is now so insanely low that there's kind of no excuse anymore to not build something if you want to. Like, there's never been a time before where you could have like a budget of couple hundred bucks and create something that would usually take a team of developers. Like, this is just. It's thousands of dollars. Yeah. And tens of thousands or hundreds of thousands of dollars. Like you developers are expensive as shit. So. Okay, maybe last question. How do you view kind of like, okay, there's like you guys have been developing for a long time. You are now able to supercharge what you are doing with these tools. Do you think that sort of like the education around or software development is gonna. And computer science more broadly is gonna shift? Like just because you, you know, these tools are only gonna get better. Right. Does how does that like change in your mind in terms of the long term? Like, do people, does the way that they approach learning about this also change? Do you not need that deep skill set? Does though the lack of as many people having that deep skill set create problems down the road where nobody actually knows how this works? They just are vibe coding everything. You know what I mean?
- [87:19]
B
Yeah, so, so I did, I did mention that knowledge can work against you in this case. But, but I would say the real, the real issue is in the middle, right. With if you're a, a novice, like beginner developer, then you're going to have a lot of success. And if you are a senior engineer, then you're going to have a lot of excess. The struggle is going to be getting from a junior developer to a senior developer, you know, And I mean with AI now you can ask it questions, now you can explore now you can do projects 10 times faster. So I think getting there could be easier for you. But it's going to be a different path. Right. And being a senior developer absolutely does help. You know, I've been able to build things that other people struggle to build by being a senior developer. So if that's your goal and your aspiration, don't give up on that. AI will only make you stronger.
- [88:16]
A
I think that in the future, well now, because we are living in the future, I think that developers like Alex and myself, that is not a developer, but maybe I orchestrate as one. Well now, but I will be successful. Alex will be very successful because we will have a team of agents that work for us. So you're a senior developer, a senior software engineer, and now you have a dozen, half a dozen agents that work for you that can write very good code. And maybe from time to time you need to Direct it and change direction. Maybe it's writing something that you wouldn't fully approve of. So you say, hey, you're going off the rails here. You need to change direction and you nudge it in the right direction and then you go back to working on your other task. I think that people that do understand code will still be very, very successful. I think that they will probably write better, more optimized apps. I don't think that software engineering is going away. I just think that it's changing.
- [89:24]
B
And also, this is not just code, right? At every industry now, people can become experts by using AI to do things that they couldn't do before, right? And I'm not saying, I'm not saying learn 100% of what you're learning by just reading AI at face value. I mean, do things and then see what the results of those things are and then learn from that, right? And so, you know, it's, it's a tool that's going to enable anyone to just accelerate into any industry that they might not have been able to do before. So don't just use it for coding. Use it for anything that you don't know and learn as much as you can and learn and grow yourself. And, and to me, like, that's the greatest thing about AI and the power of, of AI is our, our ability to become better.
- [90:11]
A
If you, if you're in a meeting, take a photo of your whiteboard and then when you're done, throw all the photos from your whiteboard at whatever AI you like to use and say, summarize this, make notes, you know, whatever, whatever you want. And it'll, it'll do it and it'll blow your mind. It'll come up and it'll pull things together that you didn't even see. Try it.
- [90:32]
C
I've started using like, I've got. The wife got me one of these plods for, for Christmas. At first I was, I'm not using that, that nerd shit. And then now it's like I use it for, for every meeting. It's incredible.
- [90:45]
A
Like, I don't even know what you're talking about.
- [90:47]
C
It's a, it's a physical device that has its own battery. Syncs with your phone though, and it can just record everything. Like it's off. Now obviously it doesn't record like continuously passively, but it automatically. They've got all these different template sets that you can use. Not sponsored by plod, by the way, but it does a really nice job of like for, for in person meetings too. Like, I just. I was always a huge note taker during meetings because I like to remember when people said shit that I can follow up on later and be like, you remember when you said this, you son of a.
- [91:22]
A
But so it's your own personal auditor AI, I guess.
- [91:25]
C
Exactly. And, like, yeah, you could do that with, like, other software tools, but I like physicality of it as well.
- [91:30]
A
Yeah.
- [91:30]
C
And, guys, my voice is truly dying. I apologize for everyone who had to listen to me today. Jesus, it's worse than usual. I appreciate you guys coming on here very much. I'm gonna link a bunch of stuff in the show notes. I'm gonna link Shakespeare. I'm gonna link Footstep Shakespeare. What the. I'm gonna link podcaster Shakespeare. Wtf? I'm gonna link Agora Spot. I'll link Shakespeare. Diy. Anything else you guys want me to throw in there or anywhere else you
- [91:58]
A
guys would send people treasures to in there?
- [92:01]
C
Treasures to.
- [92:02]
A
Okay, yeah.
- [92:05]
C
Where? And any. Any. Any soapbox pub.
- [92:08]
A
Learn more about all the cool shit we're doing on our company website. Follow us on the Noster.
- [92:14]
B
We.
- [92:14]
C
We got to do another one of these guys when. When my voice is not so entirely gone. Granted, nobody is coming here to listen to me. They're coming here to listen to you. And I appreciate you guys sharing so much knowledge. I love what you guys are building. It's seriously awesome. I'm going to vibe code. Like, I've never vibe coded anything in terms of, like, an actual website before or anything. It's the fact that I just did two in the space of this stream, and I probably could have done five, but, like, I built footstep.
- [92:38]
A
I love it, man.
- [92:39]
C
I love it. And now I'm building podcasts.
- [92:41]
A
Come on now. Like, I think that you really built it for yourself, but if you need to use me as a scapegoat, that's fine. I appreciate it.
- [92:47]
C
Well, Derek, we all know footster ushers in the bear market, or, excuse me, the bull market, so this is what is required. Let's bring it back, baby. Can you guys just tell me your.
- [92:57]
B
Your.
- [92:58]
C
The. The string of your endpubs also just recited from memory.
- [93:02]
A
NPUB1.8.
- [93:05]
C
Yeah, but, yeah, I'll link. I'll link your nose to profiles as well. Anywhere else we should send people any last words? This was fascinating. We got to do this again, like, soon. Next time you guys have a release or something, please, because this was super cool. And we'll vibe code 10 apps in that.
- [93:23]
B
Build, build, build. Now is the best time ever. To do it. That's my. My last words. Build. Awesome.
- [93:30]
A
All right. Thanks, Walker. Appreciate it, Alex. It was fun. Thank you.
- [93:34]
C
This was a blast. And thank you to everybody who tuned in on the Noster only live stream. I am again, sorry for my voice. This was a blast, though. Thank you for the zaps. Some people were very, very generous with the zaps. I will use them to fund my Shakespeare credits, so they're gonna go to good use building out some more vibe coded tools. Seriously, Alex, Derek, thank you guys so much. Fuck puberty. I'm dying. I need to go drink some tea or something. Or maybe a couple more brisk lunges.
- [94:03]
A
Or another water. I saw that in the end.
- [94:05]
C
Yeah. Little Miller latte. Appreciate it, guys. I'll close this out now. This was a blast. And that's a wrap on this Bitcoin talk episode of the Bitcoin podcast. Remember to subscribe to this podcast wherever you're watching or listening, and share it with your friends, family, and strangers on the Internet. Find me on noster@primal.net walker and this podcast@primal.netcoin on X, YouTube and Rumble. Just search out at Walker America and find this podcast on X and Instagram at tcoin Podcast. Head to the Show Notes to grab sponsor links. Head to substack.com walker America to get episodes emailed to you. And head to bitcoin podcast.net for everything else. Bitcoin is scarce, but podcasts are abundant. So thank you for spending your scarce time listening to the bitcoin podcast. Until next time, stay free.