Transcript: Give Your Agent a Bitcoin Wallet

Marty Bent [0:07] You've had a dynamic where money's become freer than free.

Vinny [0:09] Let me talk about a Fed just gone nuts, all, all the central banks going nuts. So it's all acting like safe haven. I believe that in a world where central bankers are tripping over themselves to devalue their currency, Bitcoin wins. In the world of fiat currencies, Bitcoin is the victor. I mean, that's part of the bull case for Bitcoin. If you're not paying attention, you probably should be. Probably should be. Probably should be.

Marty Bent [0:36] Vinny, you're a very quick internet celebrity this week. Welcome to the show.

Vinny [0:42] Yeah, pretty crazy. I like made a thanks for having me, Marty. It's nice to finally meet you in person. I'm I'm a subscriber and to the. TFTC, and I like what you guys do, so it's cool to be on, on the show. Um, I just like, I made a, a goal to, uh, post videos and be more active on Twitter, and, uh, 3 videos later I got Jack Dorsey retweeting me, so that worked out pretty well.

Marty Bent [1:10] It's pretty, uh, pretty sweet. And it is, uh, it's the name of the game these days, content marketing. You got to get in it, prove that you know something. It's tiring at sometimes.

Vinny [1:20] Yeah, right. People beating you to the punch, getting on, getting them on, getting the people they want on their podcast before you, stuff like that.

Marty Bent [1:29] Oh yeah, Greg beat me to the punch this week. But no, we had that back and forth last week when I think we reposted your original video on Buzz and I quote tweeted about sort of L4O to agentic payments and imagining that. And I think I think we definitely talk about Buzz. We are going to talk about Buzz and how it works. But I think taking a step back and just thinking about using these agents in general or using agents, using AI models, thinking of harnesses, models, agents, that's where I feel like most people on the outside looking in, if you haven't been playing around with these tools for a lot, it's all so confusing. And most people don't know where to start. And we've started, we actually launched this TFTC roundtable where we just have a private group of people. And we've been playing with this stuff for 2 years now with the models at the very least, but with the harnesses and OpenClaw and Hermes for the last 7 months. And we were like, okay, we sort of know our way around. And we're not technical at all. I know enough to be dangerous, but Certainly, I certainly wouldn't consider myself a software engineer. Set up my first VPS in January of this year, but—

Vinny [2:46] I set up my first VPS 2 weeks ago, and I am a software engineer. Yeah, yeah, yeah.

Marty Bent [2:55] It's okay.

Vinny [2:56] But I was just usually mostly hosting on cloud providers and stuff like that.

Marty Bent [3:00] So yeah, but I mean, based off of like our prerecording, A little powwow there. How do you view the state of AI and these tools and how to use them and how people should be?

Vinny [3:12] Well, I think you summed it up pretty well because it's really cool. It's very surprising to me to see how technical people like you can sound. You guys, you run your business, you know your workflows, you know how to describe what it is you need and the agents are able to give you that stuff. And then you're able to learn very quickly. You don't know the code syntax, but you're talking about hosting stuff on virtual private servers and things like that. It's pretty amazing the technical ability that, and the level of knowledge that you can attain just by using these tools. So I think that's where we're at. I think where the playing field is being leveled. There are definitely the doers and the non-doers, but it's no longer limited by technical ability. ability anymore.

Marty Bent [4:03] No, it's certainly— you just talk to your computer. That's like another thing that fascinates me.

Vinny [4:06] Yeah.

Marty Bent [4:07] Is it seems like the form factor or the medium through which we interact with computers is changing.

Vinny [4:13] I mean, that's what software development is, right? You were talking to your computer, but in a different kind of language, right? Now you're talking to your computer in English. So.

Marty Bent [4:21] Yeah.

Vinny [4:22] Yeah.

Marty Bent [4:23] And it's like I'm a huge voice-to-text maximalist. Just Option+Spacebar. I saw you were demoing Handy. I use SuperWhisper. I haven't tried Handy yet, but it looked very similar. And what are the— like, how long have you been playing with these tools? Like, obviously you've been demoing Buzz, but I imagine, I mean, you're building Wasp, which is trying to solve one of the pain points of people interacting with these models and these agents. So what's your journey been experimenting with this and—

Vinny [4:56] Yeah. So like you said, I'm a founding developer relations engineer on Wasp. The Wasp team, which is a full-stack framework that makes it very— the easiest way to explain it in this context is that it makes it very easy for humans and agents to build web apps with it. Because it just, again, like we're talking about, these things are abstraction layers on top of another. Like the way Buzz is an abstraction layer on top of the coding agents, Wasp is like an abstraction layer on top of the normal tech stack that you would use to build a web app. But like, it's usually all these disjointed parts. You have to kind of glue them all together yourself, or nowadays it's your agent doing it. But we have this like, you know, configuration file. It's like a central place where you can just write really simple config and then the compiler, like Wasp, the framework, puts it all, glues all the pieces together for you. So it's just like, it makes, it does a lot of heavy lifting for you or your agent, and then makes reviewing and looking at the pieces that have been put together a lot easier because the code has been reduced to like just a few lines of config. So in building Wasp, like obviously once these tools start coming out, you're gonna start using them. And like when ChatGPT dropped, that was like the first thing I did. I was like, How can I, I'm going to paste in like an example of a Wasp config file and see if ChatGPT can do it. And this was back when it was like GPT-3. I made a, like one of my first YouTube videos was build a full stack app in 10 minutes with ChatGPT and Wasp. And it was just like copying and pasting stuff back because we didn't have coding agents yet. So that already worked and that was like, okay, cool. We're onto something here. This is nice. And then it's also like right when the OpenAI API dropped, I was like, oh shit, how can I put this into an app? So it sounds silly now, but like I built an app that was a cover letter generator. It was just a GPT wrapper. You'll hear people joke about these all the time. Oh, GPT wrappers, bro. But like I made, I don't know, 10 grand off of a GPT wrapper just because I did it first. I like put it on and it behind Stripe Payments, like a really shitty app right when the API dropped. So that's been my journey starting out. I see this stuff and I'm like, cool, what creative or what cool thing can I try with it? What cool thing can I do with it? And I kind of take the other road. I'm not one of these workflow freaks where I think everything needs to be super detailed and stuff, but there's definitely potential in all this stuff. And I think just testing it out, finding the right way to use it just unlocks a lot of new possibilities for especially entrepreneurs, like young or like people starting out that didn't have access to legal advice or, I don't know, tax advice. You can get really far with that stuff, build a business, build a product, or if you have like a service business, you can build small software, personal tools, and like, it's just like opens up new worlds and new economies. So I really like that part of it.

Marty Bent [8:12] I did too. And it's, I mean, like, if you've been a subscriber, you've probably heard us talk about, I mean, we're a team of 6 here, 5 or 6, depending on the contractor work that we need on any given month. But it's insane what we've been able to do. And I know there's some people listening to this, like, Marty, stop talking to us. It's like, I'm talking about this so much because I think if you haven't been playing around with this, you really need to start doing so. And I think there's some decision paralysis because people are like, where do I start? And one of my personal preferences that I really have become convinced of over the last 7 months, and I think you sort of would think the same way, is why Buzz is so appealing, is don't lock yourself into one model provider. I like having a harness. I've been using Hermes, For many months and before that, OpenClaw. And I just like the ability to have the harness, which is, I view as the nervous system. You use the model as the agent brain, but then all this exists in the VPS that I set up in January. And then that has local file and folder systems that are our company brain that the model can plug into and then send that to the nervous system to get stuff done.

Vinny [9:31] Yep, definitely. Yeah. And I saw that you were talking about how you have all this context built up in Hermes, and now that you can just plug that into Buzz and share it with other agents is a huge unlock. So, I mean, that stuff's really cool. I think openness is— we're starting to see the openness factor of being able to have access to open source models. and just choices in the AI ecosystem, that's becoming more and more important because of whatever top-down policies or government mandates or just that stuff might just be too expensive for some kid who wants to get his idea off the ground. And yeah, you got to have access to an open-source model, switch it over or use shared compute, which Buzz has, which is a really cool feature. Go for it.

Marty Bent [10:21] Yeah. How do you view this? Closed source frontier, OpenAI, Anthropic versus open weight Chinese model?

Vinny [10:31] I mean, it's good competition. Let's see where it goes. Cloud code is definitely— I mean, we have to give a lot of credit to Anthropic and OpenAI. They've paved the way for sure. But we see their response to things is always kind of like, take a step back, like, serious? You guys sure about that? Like, you sure you want to go with that marketing messaging? Like, I don't know, there's always things like that. Or I'm waiting for them to— I assume that token subsidy is a real thing and that we won't get these prices forever, but you never know because just stuff is progressing so fast. So I think it's good, it's healthy for the ecosystem, and I think the other players that are behind on Anthropic, they like it too because it allows them to catch up. So I think it's just interesting to watch. I'm just watching from the outside using it and yeah, just having fun. To bring one of your points that you made earlier about— oh shit, I lost my train of thought. Anyway.

Marty Bent [11:47] Well, I'll continue. Was it the model? The one thing I agree, I think Anthropic and OpenAI, despite their flaws, mainly that they're run by either sociopathic or effective altruists, they have an effective altruist at the helm of one and a sociopath at the helm of the other. It is undeniable that the models they produced are incredible. That's what I worry about here in the States is them really positioning to get a regulatory moat and getting their way and the government killing the open weight, the ability of the open weight models to progress. But I'm cautiously optimistic that we'll avoid that. But bringing it back is, again, the harnesses and the one piece of advice I give to people who are beginning to play around with this and set up agentic flows is don't don't get vendor lock-in. I haven't opened Codex Desktop or Claude Desktop in 6 months. I've just been interacting predominantly through Telegram with my Hermes agent. I have a group chat with it with a bunch of different topics and it's represented parallel sessions. And so it's able to get work done. And the only time I'll open up Claude or Codex really is if I have to hop in the terminal and debug something, I'll open the CLI directly.

Vinny [13:11] Yep, yep. That makes a lot of sense. And you remind me of what I wanted to say earlier, which was, like you said, people have decision paralysis around this stuff. And I think really the important thing is just pick one, start using it. You know, it doesn't really matter which, the same topics, the same, like, kind of, yeah, same things apply to all the others. So it doesn't really matter which one you start with. Get your hands on the one that looks good or affordable or fits your philosophy and then just start using it to do stuff.

Marty Bent [13:44] Yeah. Now shifting to Buzz, I mean, I think we've both played around with it for the last week. It's definitely rough around the edges, definitely still a prototype, I think. But last night it took me a while going back. It took me actually like 2 days of like, I spent like an hour trying to do it. on Monday, and then last night I finally got around to finishing it, but connecting the Hermes agent to Buzz and getting it interacting with Fizz, Bumble, and Honey, the 3 agents that Buzz has out of the box. And it was pretty crazy. And for me, it was a bit of a heavier lift because again, I have my Hermes agent running in the VPS. So the way that Buzz works, it's running locally on your machine. And so I have Hermes Desktop running locally on the machine too, but I had to SSH into the VPS on Hermes Desktop and then port the, basically tunneling from Buzz on my computer to Hermes on my computer to Hermes in the cloud.

Vinny [14:48] Nobody would know that you're not a software developer talking like that.

Marty Bent [14:53] No, but somebody probably could have done it in 5 minutes, but it took me the better The better part of like 3 hours over 2 days to figure it out.

Vinny [15:03] Yeah, yeah, yeah.

Marty Bent [15:03] But I did.

Vinny [15:04] Yeah, yeah, exactly.

Marty Bent [15:06] But I think I'd be interested to get your thoughts too. I know you explained them on Greg's podcast yesterday, but just in your mind, compared to just using Codex or Claude Code or an agent harness or Wasp, like how is Buzz different and what does it unlock?

Vinny [15:26] Yeah, I think I said it a few times in different tweets and stuff that I think the Slack killer thing is a nice piece of marketing. It's a little bit superficial because I don't really think that's the only thing or even the main thing. I think the main thing, the main drivers behind it is this ability to switch models and harnesses. It's like kind of a Trojan horse to get people to switch over to local open source models, right? Give them the ability to use Claude and Codex, but also to switch to any other open source model or any other open source harness like OpenCode or Goose or one of those. And because if you can look a little bit around the corner, you can see, I mean, it's not even that hard. Timmii K3 is like, bumping up next to the frontier models for a fraction of the cost. So it already gives you a perfect avenue to go that route. So I see it kind of like a Trojan horse for open source models and for shared compute. And then further down the line, Bitcoin Lightning payments, whether it's agent to agent, human to agent, human to human. My guess is they start small, early preview, buggy, but they work out the kinks with the users so they know which kinks to really focus on and then slowly release those underlying features. And another one of those is Git. I think they're looking to kind of replace GitHub, right? So you've got all this code on GitHub, all your code, and if it's public, anybody can train a model on that. The whole idea is Buzz kind of works like this global context window for you and your community or your company or whatever. And then they have under the surface an ability to host your repositories remotely on your VPS or on your server, whichever server that you run whenever your relay is hosted, self-hosted. So I see them, I see it as being many things more than just a Slack killer, you know, and those are the directions I can foresee it going in. Yeah.

Marty Bent [17:39] Well, and I mean, you seem to be very well versed on it because you've done some videos on Nostr specifically, but I think Nostr, I know Nostr's at the core of this and as somebody who's been a long time Nostr bull, I think I was probably one of the first 100 people on Nostr, maybe first 1,000. I'd be more confident saying the first 1,000. So I've been on Nostr since—

Vinny [17:58] When did it launch actually? Do you remember?

Marty Bent [17:59] I think 2022. Okay. If I go, let me see. I think I was on it like 3 months after, or a month after it launched. It was very— it was like a very niche, like, not even a Bitcoin side project, but it was like a side project a bunch of Bitcoin developers worked on. I've been on since, yeah, July. Wow, it's 3 days past my 4th Nostr birthday. July 26th, 2022.

Vinny [18:31] Nice, nice.

Marty Bent [18:32] Yeah.

Vinny [18:33] I got on in 2023, I saw, but I'm not a huge like Primal user or Nostr user. But like, I, that app I was telling you about earlier, the GPT wrapper, I was like, fuck it, I want to learn how to use Lightning payments. So I put LNURL, like Lightning auth, I think it's LNURL auth in there. It's been a while since I messed with it and actually got some Lightning payments for people to generate cover letters. So that was cool. That was like, it seems ancient now in AI years. It was like 2023 or '24 that I put it up there. Yeah. So it was pretty funny.

Marty Bent [19:12] Yeah. But I mean, to your point, like Primal, Damus. So everybody listening doesn't know what Nostr is, Notes and Other Stuff Transmitted Over Relays. It's a communications protocol, very simple in the first, predominant use cases or applications built on it were social media-based. And for the longest time, Nostr developers and anybody who's been paying close attention to development, I think Jack is one of those people, I've talked with him specifically about this over the years, social media is just the tip of the iceberg, just this ability to use private-public key pairs to sign events and have cryptographic proof that something existed at some point in time in a certain state. is very powerful. In the social media use case, you can prove that somebody who controlled a private key at this point in time wrote this message and relate it to the network. But now, like you're saying, I think one of the biggest problems that many have with these LLMs, particularly if you're using Codex or Claude Code only, is the context windows are are relatively short if you're pinging the models a lot. So you can jam the context window or compact it multiple times a day, and then your AI system gets some form of amnesia if you don't have a second brain or something set up that it's plugged into. And that's where I think Buzz leveraging Nostr in this historical context that you have over time with signed events, is an unlock for creating a corpus of context that builds over time that the agents can just plug into.

Vinny [21:00] Yeah, that's a really good observation. I always looked at it more like, the thing I thought was cool is that it's a permissionless protocol, so you can just kind of interact with anything on the protocol. So you can just create these Frankenstein apps. You could connect, imagine trying to connect Slack and Discord and GitHub together and put 'em together. You can't really do that. You could do that very easily with Buzz and some other apps or things running on the protocol. So seeing Buzz come out built on top of Nostr and being like, okay, cool, there is a really nice use case for Nostr outside of social media. This is exciting. Let's see what else we can do. Yeah.

Marty Bent [21:39] So, Maurice, this episode was brought to you by our good friends at BitKey. Self-custody is one of the most important things we do as Bitcoiners. Not your keys, not your coins. A lot of people never make the jump because the process feels intimidating. BitKey is trying to make that jump easier. You guys know it pretty well here already. It's a Bitcoin wallet from Block built around a simple idea. Let's give people a better path to holding their own keys without making them feel like they need to become a security engineer overnight. If you have Bitcoin sitting in an exchange and you have been meaning to take custody, stop procrastinating. Use the code TFTC10 for 10% off the new BitKey at bitkey.world/TFTC10. All right, freaks, you know I don't take sponsor money from products I wouldn't use myself, so listen up. The Aave and Bitcoin Visa card is one of the most interesting things I've seen in the Bitcoin lending space in a long time. Here's the deal. You can get a line of credit up to $1 million backed by your Bitcoin without selling a single sat. No games, no annual fees, no minimum draws, and your Bitcoin is custodied by BitGo, one of the most trusted names in digital asset security. Aven never lends it out. There's no rehypothecation. You stay in control. You can lock in a fixed rate for up to 10 years. 10 years. That's 10 times longer than most lenders out there, or go interest only for up to 5 years. Rates start at 8.99% APR for a product that lets you keep your stack and still access liquidity. It's hard to beat. On top of this, guess what? You also get 2% unlimited cash back every time you use the card. Spend fiat, keep your Bitcoin. That's the whole game. If you've been stacking for years and you need liquidity without triggering a taxable event, this is worth a serious look. Go to aven.com/bitcoin. That's aven.com/bitcoin. Check it out. And then the agents get their own identity. Yep. Which I think is interesting. I'm not— I haven't explored the full thought experiment about what that means, but just thinking about where this— because I have my Hermes agent and everybody's like, when are you going to let other people on Buzz get access to it? I'm like, never, because it's got access to a server with a bunch of proprietary information on it. But you can think of a scenario where people just build individual agents that don't have as much sensitive proprietary information, don't have access to that information. People can just begin sharing agents. They can have Matrix Agent Smith-like reproducibility just basically emanate throughout buzz communities across the world.

Vinny [24:07] Yeah, exactly. Because they can move. You can move with that. When you go take your identity with you to a new community, it comes with. But I also think that it's there and built on Nostr so that it can enable sharing of resources, of compute, and paying for those. So I think there's more than one side, right? It's like all the benefits of Nostr, but I think that's going to be something we'll see. And I mean, I wrote that in one of my first tweets and Jack retweeted it, so there's a good— There's a good possibility that that is what they're planning.

Marty Bent [24:45] Yeah. No. And then the fact that it's on Nostr too, it's open source protocol. A lot of the applications built on top of it are open source as well. And so you can, the permissionless nature of it too, you can imagine just sending agents, be like, hey, somebody had to build something like this on Nostr and go find it incorporated into our workspace. the whole idea of shared compute, there's Routester, OpenAgents. There are these compute markets that have already been built on top of Nostr. So the ability to port in that code and just create an open marketplace, like, hey, I have a GPU running in my house, I'm not using it. I'm going to put an L402 gateway on top of it, plug into Routester or OpenAgents. I'm going to use their code to— basically market that I have this compute available and anybody that is connected to me via Buzz, if our mpubs are connected, like you, you have the ability to access this compute as well if you pay for it.

Vinny [25:47] Yeah. I mean, it's pretty cool. We saw what happened. I'm sure people, or you know about what's going on in India where they're banning BitChat and stuff like that. And then it's just like when you have these kinds of shared channels over open protocols, it's pretty hard to shut off access to a model if they're being shared that way, I would assume, right? So that's also a cool possibility.

Marty Bent [26:12] Yeah. I mean, I forget what the Nostr GitHub replacement is called, but yeah, if you get that in there, it's open source and the code is stored on relays that people are self-hosting. It's going to be very hard to shut that down.

Vinny [26:31] Definitely. Yeah, cool. So we've been pretty philosophical. Should we do any, um, some demos or what else? What are you thinking?

Marty Bent [26:39] Uh, I think let's do a demo. Like, I mean, and maybe as you're doing a demo of how you're using this, explain how, like, what, like, the unlock is. Like, for me, like, again, the agents being able to coordinate, like, Martin who is our Hermes agent. When I got him connected, I made him basically like the Sergeant at Arms lead connector. And then I rewrote the system prompts for Fizz, Bumble, and Honey and created one other agent. Basically, Martin became our orchestrator, and he's got access to, um, basically like our— we put all of our meeting, like our transcripts of our meetings, um, contracts, uh, advertisement schedules that we have to get in. So he's got access to all that company information.

Vinny [27:24] Cool.

Marty Bent [27:24] So it's like Martin, You know what we're trying to get done in the business. You know, sort of some of the low-hanging fruit initiatives that we haven't begun working on. Like, all right, hop in here with Fizz, Bumble, and Honey and delegate some of the tasks that we have not finished yet to them. And it was just like, I was watching last night, there was probably like a 45-minute loop of them just going back and forth until they got to a point of actually completing a task. So I was like, holy crap, this is crazy. Yeah.

Vinny [27:55] Yeah, it's pretty good at that too. I'm surprised at the end-to-end workflows that it's able to complete, know when to keep going, when to delegate, and when to stop and stuff like that. So that's pretty cool. Sounds like you might even be doing more practical stuff than I am because you're actually using it for your team. And meanwhile, I'm just playing around doing demo-y stuff. Can you see the screen here?

Marty Bent [28:23] Yep.

Vinny [28:24] Yeah. And let me just get this minimized here. So yeah, like you were saying, what's a good one? One thing I shared earlier today that a lot of people seem to like was this prompt where I'm pinning models to the agents. So I'll have one that is pinned to Sonnet, a lower-level model, and one that's pinned to Fable. For more complex tasks, but I'll have them bounce stuff back and forth because I kind of know the workflow. So I'll tell, you know, Shared Boy, you download a transcript from this, the video and this transcript, pass it to Fizz, and Fizz, which is Fable, find the 5 most surprising ideas from the transcript with timestamps and really get chunks where you're not cutting off a sentence or an idea and pass that new document of chunked transcript back to SharedBoy, who's Sonnet, use FFmpeg, cut out those sections to make short clips to post to social media. So I get them to make basically shorts for me for Twitter and stuff with that. And then like you said, they're bouncing stuff back and forth. And I find this workflow way nicer than in Claude Code where you have subagents. I don't know why, but I never caught on to using subagents. Like, you have the same thing basically, but just something about having them in a shared chat, which is like, people have been gravitating towards using mobile apps, mobile solutions, right? Because of something about that where you can just kind of have conversations with your teammates, with the agent. It just, I don't know, it feels nicer. And I really like that about Buzz. That's just like one example that I used it for today, where I'd prefer to use it over something like Claude Code.

Marty Bent [30:17] Yeah. Well, in this exact example, so we have this exact workflow. We built this for the pod. We'll use this workflow for this podcast, like a very similar workflow. It's probably not the exact same, but we do the same thing where it's like, hey, take the transcript with the timestamps of this episode, find the most engaging parts.

Vinny [30:40] Yeah.

Marty Bent [30:40] Actually, we have a Twitter API, a third-party Twitter API that we use, scrape that, look at what's trending, and if anything that we talked about matches with something that's beginning to trend, maybe we should focus on that. And then find the timestamps of the clip and then go clip it. But we have our social media manager who has access to Martin, in a sort of private Discord channel doing that workflow, like, in a silo. And so, like, when I think about Buzz, what this opens up is, like, the ability to take our social media manager and Martin, the single agent, out of that silo and, like, pull them into a more collaborative space where—

Vinny [31:20] Yep.

Marty Bent [31:20] Not only could he ask Martin to do this, but, like, he could delegate specific tasks within this particular workflow to get better. better results. And then on top of that, you can have the human members of our team hop in and be like, well, actually, yeah, you guys did this, but here's some input from me about another angle we could take on this.

Vinny [31:41] Yeah.

Marty Bent [31:42] It just really consolidates the work and makes it more open. And that's actually one thing I've been thinking a lot about, and it's actually, funnily enough, influenced by Dorsey. I think Dorsey is really emerging as somebody who's viewing this whole AI revolution with a very clear mind, and particularly his paper that he wrote with Olof from— Roloff, excuse me, from Sequoia, and then the podcast they did about the intelligence layer, companies becoming intelligence layers. Right. And this whole concept of instead of individual employees or members of your team, working in these silos. Like, yes, they will have delegated tasks and they'll have goals and there will be a division of labor, but the ability of these agents to open up collaboration across the team and for the team to have access to all the information about what everybody else is doing is extremely powerful. And I think that's something that people really haven't caught on to yet either.

Vinny [32:48] Yeah. And that's why I like to kind of refer to it as it's the first step in the right direction to have a global context window, right? Because it's like usually you have, like you were saying, things siloed. are spread out between different apps and you're bouncing back between them, you lose context, you lose the ability to collaborate with other people, get input from other people where it's like, you just talk about what you want to do, your ideas, and then you can have something implemented for you right there in that moment, right? So it just speeds things up, gives you this global context window, so to speak. And I think that's a really cool direction with this. And one other thing I wanted to point out was I'm doing this as a one-off. It's kind of like a skill or whatever. It's a prompt, right? But with a lot of that stuff, I think personally I'm inspired by Buzz that it's built on Nostr. I'm trying to think of ways to incorporate Nostr into apps that I build because I think it makes sense when it's a deterministic workflow, right? When it's like, You know, it's just a very strict process that's going to be repeated over and over to build some kind of like just niche small app that you host that does that. So like you're talking about where it's like scraping the, using the X API, scraping X for certain trends, cross-referencing that, like I kind of built something similar as just like a little demo. You might have seen it in the video that I posted on Twitter. And I originally had an externally facing API and I just thought it would be cool to learn a bit more about Nostr. Instead of my agent running a workflow or a recurring daily task and doing that call and using tokens for that stuff, why don't I use a recurring job in my Wasp app? Because that's also something really easy to do in Wasp. Recurring jobs are a primitive in Wasp. So you just set up a recurring job, tell the agent, set up a recurring job that instead of that takes the same stats that you're pushing to the external API, but push them to Nostr. Just make it a Nostr event, sign it, and push it to Nostr on my relay. And so I just get now tweet stats every day from my app pushing to Nostr. So the agent isn't doing this, but it's going into the agent context, right? So I can just reply to this here and be like, cool. you know, fizz, tell me what's the, like, what's working and what's not, right? Based on the current trends, like you're doing stuff like that, I think is really cool. I don't think it's like nailed the ability to use Nostr to its full potential, but I think just through trying out things like this, we might be able to find cool ways, creative ways to build apps on Nostr that are integrated into your relays and your Buzz communities.

Marty Bent [35:43] Sup, freaks? When you take Bitcoin seriously, you start with custody. You want to control your keys, avoid single points of failure, and make sure your savings cannot disappear because you or someone else screwed up. That is what Unchained has been focused on since 2016. Unchained is the leader in collaborative multisig custody and Bitcoin financial services that keep you in control. They secure over $12 billion in Bitcoin for more than 12,000 clients. That means about 1 out of every 200 Bitcoin sits inside an Unchained vault. Their model is simple. You hold 2 keys, they hold 1 key, and it always takes 2 keys to move Bitcoin, meaning their single key can't access your Bitcoin on its own. Just resilient shared custody that gives you institutional-grade security while keeping you sovereign. Unchained also lets you trade straight from your vault, access Bitcoin-backed commercial loans, open a Bitcoin IRA where you hold your own keys, and set up personal, business, trust, or retirement vaults. They even offer inheritance solutions built for long-term hodlers. Or opt for the highest-level private client service with Unchained Signature and get a dedicated account manager, discounted trading fees, exclusive access to events and features, and much, much more. If you want a partner that helps you secure and grow your Bitcoin without giving up control, go to unchained.com and use the code TFTC10 at checkout To get 10% off your new Bitcoin multisig vault. That's TFTC10 at unchained.com. Freaks, look at me. I'm glowing. I've got like an angel's halo going around me. You know why that is? Is I feel good. I feel taken care of. I feel blessed, healthy, happy. And that is because I'm a CrowdHealth member. My family and I have been CrowdHealth members for 5 years now. Literally this month, 5 years ago, we joined CrowdHealth. We've had 2 babies, we've had multiple health events, and we're never going back to health insurance. CrowdHealth is crowdfunded healthcare. So you, you sign up for CrowdHealth, you pay a monthly fee, you help out with other people's bills, and it's significantly cheaper than health insurance. We were on COBRA as a family of 3 when I left my last job before I went full-time with TFTC, went on the CrowdHealth. Now as a family of 5, we pay, I believe, $700 a month. It's significantly cheaper. They're going to negotiate prices lower for you. They've consistently negotiated healthcare prices as much as 50%, 60%, 80% in many cases. They help out with babies. If you have a pregnancy, you pay the first $3,000 and the crowd covers the rest. If you have a regular health event, you pay $500 and the crowd pays the rest. Go to joincrowdhealth.com, sign up today, use the code TFTC, opt out of health insurance. I'm uninsured, baby, and I love it. Use the code TFTC at joincrowdhealth.com and you'll get $99 a month for the first 3 months that you're on the CrowdHealth platform in the community. Bitcoiners, you found sovereign money. Now find sovereign health and sovereign healthcare. I think one of the underappreciated aspects of that building on Nostr too is the network effect. Because what Buzz essentially is, is a Nostr client, right? It's just a client, it's a front end built on top of the protocol and the relays connected to that protocol. And what many people, particularly the Nostr haters over the years, have not realized is that all Nostr needs to do, similar to Bitcoin, is just survive long enough and maintain a big critical mass of network effect that if it continues to grow, for entrepreneurs specifically, if you're building apps, the way that Nostr works with the public-private key pairs and the way clients work is I have a primal— I created a new endsec and public key for Buzz for my individual profile there, but I could easily import the nSec that I've been using for 4 years. And so the same nSec I use on Primal, I use on Damus, I use on Wisp, I use on Zap.Stream, I use on Ralster. I can just port that over to Buzz and have the full context of my Nostr history brought to Buzz. And so instead of trying to basically, as an entrepreneur releasing an app, trying to bootstrap network effects, if your app's good enough, Anybody on Nostr with a private-public key pair can just easily sign into your app with that and bring full context with them, which is extremely powerful.

Vinny [39:44] Yeah. And it's kind of cool to see, like, you know, sometimes we're like, for people with these forward-thinking technologies, it takes a while. Like you said, it just has to not die, but it's like people build it. They don't even know that like AI and coding agents are coming around the corner, but it takes something, some other kind of progression, some other kind of step change to then open up the aha moment, right? Where it's like, oh, shit, this is it. This is going to work now. This might really work. And that's how I kind of feel like it's happening here. It's like the beginning of that aha moment where it's like, ooh, Nostr can be bigger and can really play that role that it was designed to play. So I'm excited about that. Yeah.

Marty Bent [40:26] And it's like a weird feedback loop too, where the AI vibe coding is sort of enabling it in a way. Where instead of having to read the docs on how Nostr works, it's like you just tell your clanker, figure out.

Vinny [40:41] Yeah.

Marty Bent [40:41] When I spun up my OpenClaw agent, I was just messing around, seeing what it could do. I was like, okay, I want you to be able to spend Bitcoin. Go figure out what the best wallet is and the best way for me to get you Bitcoin. It did some research. It was like, hey, we have a server. there's this thing called PhoenixD where we can run a Lightning server on— we can spin up a Lightning server in our VPS. We should use that. I was like, okay, now I'll just send you— like, when you set it up, send me an invoice to basically create a channel. Because the way Phoenix works, it's like, okay, when you set up the server, send me an on-chain address, I'll send Bitcoin to that, and then you'll submarine swap that. into a Lightning channel so that you can spend. And it did that. And I was like, okay, you have a server, you have Bitcoin, you have LNURL auth. Since you have your server, you can sign the private key and you can sign into websites that are LNURL auth enabled. And I was like, all right, go to LN Markets and I want to see you put a trade, go long Bitcoin, sign in, go long.

Vinny [41:50] Cool.

Marty Bent [41:50] It did that. And then I was like, okay, I want you to manage this, just create a simple trend-following trading strategy. I don't care if you lose money. I just want to do this for experiment reasons. Also create a Nostr account and then post your trades on a daily basis. And it did that. It created an NSEC and an NPUB and was posting to Nostr autonomously for a month before I was like, all right, this is cool, but I don't feel like keeping up with the trades. But going back to the feedback loop, to your point about it just needs to survive and then something will happen where it really creates that aha moment, it's a symbiosis where Vibe Coding's enabling the accessibility to these protocols, which are a lot of times technobabble to many people, but you don't have to understand the technobabble anymore. Just tell your client.

Vinny [42:44] Yeah, yeah, that's crazy. It's really fun because you just have the idea and then you just go for it. I love that. Yeah. Something else I do a lot, like, I just get it to create HTML presentations for me so that I can learn stuff. Because you were talking about, like, I don't know how to use or build on Nostr, but I just tell my clanker to do it. Or if I want to learn basically a high-level thing, I'll just be like, create an HTML presentation. And I actually did that for Nostr. So I was like, I'm a full-stack developer, make sure I really know what I'm talking about, create that presentation for me, just like an HTML file. I love doing that to learn about stuff so that I get a deeper knowledge. And then I did this before I started building that app that was sending Nostr events. So when it does stuff, I still have a clue what it's doing. And if I run into problems, I can maybe point them in the right direction. So that's cool. That was fun while building this app here that I can just click sync and then I get that event, that Nostr event in Buzz. And then when it comes in there, some of the workflows, like right there, boom, came in, tells me, hey, you got new tweet stats. And then I can just start chatting with him. So some of the workflows in Buzz are still a little bit like, iffy. I don't know. There's a, this, it's an experimental feature. So in their defense, you know, they've got it in settings under experiments. You can click on workflows or projects. And I had some, some work, some don't, some are a little iffy. So I think they'll get it worked out though. Yeah, I think that's cool.

Marty Bent [44:21] I do as well. The, I'm sorry, I got my kids yelling in the background. I don't know if you hear them. The, Do you think this is making people more productive? I think that's one of the big questions, particularly for the bears and the doomers out there. Like, oh, it's great, usage is going up, but is there actually more productivity in the economy?

Vinny [44:44] You can't just ask a question like that and then leave, Marty. I just shut the door. I have to smell something. That's a good question because yes, it can. Is it? I think we have to wait a little while to see the effects, of course. It's going to take a while to trickle through on a broader basis. But I think if you know what you want and if you know what you're doing, you know what you want and you know when— even if you're asking these things for advice, if you know when it's giving bad advice, or you have a good sense of what's the better advice, which advice to take, then yeah, you're like, I started this, that app and a company. I wouldn't have been able to do that. I probably would not have done it pre-2023, pre-AI, because I'm not the kind of person that's going to go out on a limb and hire a tax advisor and all these people. I literally did the majority of the work first off with just agents, like copy and pasting websites, screenshots, like, what's this mean? What do I got to do here? What do I got to do? And just slogging through it. And before we really had good agents like this, but yeah, if you're driven to do that kind of stuff, you have, so you just have an idea and a will to do it, you can do it. So yeah, I think for the right people, it's a multiplier for sure. It's a great lever. It's a huge lever. And in the wrong hands, it's just going to lead to nothing as well. That's the way it goes.

Marty Bent [46:33] Yeah. I mean, you used the word earlier, but I think in terms of deterministic workflows and sharing them, but I agree too. That's the way I view it. And that's what I tell people is the way I use these tools is I know what the output should be, particularly for our workflows at TFTC. And so I know when it's wrong. I don't expect the models to go out and create something new for us, if that makes sense. It's like, I know we have this task that we need done and we typically don't get it done because we don't have enough bandwidth. I know what the end result should be. Please go do that and get the output. Is it right? Is it wrong? If it's right, thank you, good job. If it's wrong, you messed up here, do that. That's where I think people falter is thinking that it's going to just do something magical for them. Like, these are essentially— they're large language models. They're just doing a probabilistic analysis of what the best next word is and the output. And like, you just need to know what the output should be and apply that lens to it. Like, don't say, build me a business that makes $1 billion a year and make no mistakes. That's not going to happen.

Vinny [47:45] Right, right. Yeah, you still got to do the work. It's just easier to do the work. That's about it. So it's easier to get something done that you might have basically no knowledge on previously. Like, you're a great example, like being able to self-host these things, these tools to virtual private servers and connect to the SSH into them and all this stuff. It's like, yeah, that stuff was very hard just 4 years ago. So it's like, To me, it's not really a debate, but when I see companies that have token mandates, like you have to use so many tokens, you have to use AI to hit these benchmarks, hit these targets, and the people who sat around and did nothing before are just going to waste those tokens. So it all depends on how you approach it.

Marty Bent [48:38] Yeah, okay. Bringing this back to Buzz, what would be your top request for a feature?

Vinny [48:45] Top request for a feature? I think they're doing it right. I'm willing to, these are experimental features. This is early preview stuff. I think what I would like to see is if you're used to, if you're kind of used to working in a terminal and You know, just like these views. And then if you're a software developer, you know, you might want to jump back and run your own commands like you're used to that stuff. Could be cool if they had something embedded in here like that for the more technical people. But it's not like, it's not, yeah, the cool thing is it's open source. Someone like me or someone could put that in as a feature and submit a PR to Block. So maybe we'll see that. I think I like the approach. I'm willing to deal with some kind of the rough edges at the moment while it's really early preview because I think it's going in a very cool direction. I didn't— I'm not one to jump on bandwagons just because they're trendy. If I see something that interests me, then I'll try it out. So yeah, this seems like a pretty cool approach. I want to see more integration with— I want to see— what I would like to actually see is other cool stuff happening on Nostr that gets integrated here in a creative way so that we can use these tools together in creative ways. But I think we're still kind of figuring it out. At least I am.

Marty Bent [50:12] Yeah. And the tweet that you responded to, I think the first big unlock could be basically forking or implementing Ralster open agents for a market to compute, particularly for open source models, and then equipping the agents with Bitcoin wallets that they can load up. Or to your point, I think you said it on Greg's, being able to tip other humans or agents to go do work for you. Very trivial.

Vinny [50:42] Yeah. I mean, look at Primal or I'm sorry, Nostr. I use Primal mainly, but Nostr is such a good example. And I mean, Lyn Alden has pointed this out before that the incentive structure is so different than Twitter, right? Like when you make something useful on Nostr, people literally tip you, send you money. And now imagine that same— so we know it works. We know people are willing to tip for valuable information, insights, work, things like that. Now imagine you have that here in Buzz where it's like an open community. So it's not just your company, right? You're doing something private, but Let's say we all join together and we start some open community that's working on an open source project and someone just comes up with a really cool idea and mentions it in the community. You could pay them for good ideas because then you get the agents to implement them, or then you pay whoever's running the really cool high-performance open source model. You send them a Zap to run that, to implement that task on their shared compute or on their agent. So that's really cool. I think that's what I want to see, basically. And I think we will see that if, like you said, if it just stays around long enough, it doesn't die.

Marty Bent [52:01] Like a bounty board too.

Vinny [52:05] Yep.

Marty Bent [52:06] Create bounties for particular workflows or tasks to be done. And it's like a perfect example. This podcast is hosted on Nostr. We use Fountain as our podcast host. It's a Nostr client. And so there's going to be people who listen to this podcast and stream us Bitcoin over that. Lightning Network. And if you were— I could literally put your nPub in this episode. I think we can do it individual episodes and you could share in the stream of Bitcoin that's coming our way as people are boosting and streaming sats to this episode. Cool.

Vinny [52:43] I'll spend it on some sweet, powerful agents then in the future when it gets implemented. But yeah, that's what I would like to see.

Marty Bent [52:52] Yeah, I know it's getting late where you are. How should we end this? Any final thoughts, any calls to action, any philosophical musings about the state of AI, where we're going? And we're getting—

Vinny [53:08] No, yeah, no, I think we did a pretty good job of covering that stuff. I like the angle we took here. That was cool, different than the other podcasts. So that's nice. No, I think this is pretty good. Maybe if people want to know more or hear more, they should let us know. And then that'd be cool to— there might be some things out there that people are interested in that we haven't really hit on. And then we can explore it further, talk about it on the social media platforms or do another one of these. That would be cool.

Marty Bent [53:45] That'd be awesome. Well, appreciate you coming on.

Vinny [53:51] Thanks for having me.

Marty Bent [53:52] I'm sure it's been a hectic week for you. You're just out there talking about the tools that you're using and become a little internet celebrity this week. So I appreciate you coming on this show.

Vinny [54:04] You're welcome.

Marty Bent [54:04] It seems like you're a Bitcoiner. At least you're playing around with it, which is cool to see. And in Nostr, you're becoming well-versed on it. These are 2 projects that obviously I've been very passionate about for many years, and it feels like this agentic economy is the moment for Nostr and Bitcoin to shine. And it seems like we're getting to the point where people are beginning to recognize that.

Vinny [54:32] Yeah.

Marty Bent [54:33] It was just cool to see you talk about these technologies and the potential for them on a show like Greg Eisenberg, who really doesn't cover this the stuff much.

Vinny [54:44] So yeah, yeah, totally agree.

Marty Bent [54:46] Yeah. All right, well, hopefully we'll do this again at some point in the future. And until then, enjoy it, brother.

Vinny [54:53] You too.

Marty Bent [54:54] Peace and love, freaks. Thank you for listening to this episode of TFTC. If you've made it this far, I imagine you got some value out of the episode. If so, please share it far and wide with your friends and family. We're looking to get the word out there. Also, wherever you're listening, whether that's YouTube, Apple, Spotify, make sure you like and subscribe to the show. And if you can leave a rating on the podcasting platforms, that goes a long way. Last but not least, if you want to get these episodes a day early and ad-free, make sure you download the Fountain podcasting app. You can go to fountain.fm to find that. $5 a month gets you every episode a day early. Ad-free helps the show, gives you incredible value. So please consider subscribing via Fountain as well. Thank you for your time, and until next time.

Vinny [55:47] Okay.