OpenClaw and Building Fake Apps
#32

OpenClaw and Building Fake Apps

Justin:

Everyone wants progress and Pascal. You got two usernames.

Brian:

That's Pascal's AI agent.

Justin:

Oh, damn it. He's his Clodbot's here too.

Brian:

That's right.

Justin:

Is the Clodbot in the room with us right now? Welcome to the panel where two bootstrap founders talk about building better businesses and a better life. I'm Justin Jackson, cofounder of transistor.fm.

Brian:

And I'm Brian Castle. I'm the creator behind Builder Methods.

Justin:

Builder Methods. And now everybody's doing so many things. We're all the cofounders of all sorts of stuff, including and and the spawner of AI armies.

Brian:

I'm trying. I'm trying to raise my army here.

Justin:

You know? Okay. So what let's let's get it. Do we want to talk about your agent? So we could talk about your agent army.

Brian:

Like, what what are we like, what's on the agenda today? And then we'll do a quick catch up. So Yeah.

Justin:

Let's talk about the agenda. So Agent Army.

Brian:

Meaning like, you know, Open Claw, Clawbot.

Justin:

But you also had a screen that you shared in your community that I want to

Brian:

talk about? So I yeah. So I as of last night, I built a little management HQ app for managing my multiple agents that I have running on this Mac Mini through OpenClaw. But I'm I'm still exploring and figuring out that whole thing. We'll we'll get into all that.

Brian:

Okay. I'm trying to push through getting this Claude code course out. Sales have been pretty good for that. And I've I've been recording video lessons, and I'm gonna push the first batch of those out probably tomorrow. Couple other, like, business updates.

Brian:

I I went to Big Snow Tiny Conf last week. That was a that was a really good time. Oh, sweet. Took me twelve hours to drive up there and four hours to drive back.

Justin:

Oh, wow.

Brian:

Because it was I was driving driving through a blizzard.

Justin:

Well, at least you got some snow.

Brian:

Got a lot of snow, but it was also zero degrees up there.

Justin:

So that

Brian:

was fun. Oh, wow.

Justin:

So it was some cold.

Brian:

Yeah. Yeah. But it was a good time. Good good group. What about you?

Brian:

What did you do to highlight the last

Justin:

two got years? Some things on my brain.

Brian:

On your agenda. Yeah.

Justin:

I got about I got I want to talk about why I think it's good to build fake products. That's one thing on my head.

Brian:

Interesting.

Justin:

What I learned from Lars Lofgren, just published the first episode of the new marketing for developers podcast. An

Brian:

interesting person to hear from.

Justin:

Yeah. Members have access to the whole thing right now. I got to publish the public version for everyone else shortly. Very interesting.

Brian:

I can't wait to hear about this. Yeah.

Justin:

There I've got some big high level takeaways that I think I think everyone will enjoy. So and we're we're getting ready to do our founder retreat. John Buddha and I are leaving on the twenty fourth, going to Mexico for the first time. Oh. So it's it's that aisle it's an island off of Cancun.

Brian:

Oh, is that Cozumel?

Justin:

I think so. Nice. I'll I'll try to confirm as we go along here. Alright. Let's get into it.

Justin:

Because I wanna talk about this thing here. I'll I'll for the folks watching live. Hey, folks. By the way, you want to join us live? Go to panel what is our website?

Justin:

Panelpodcast.com.

Brian:

I always forget too. Is it ZuppPanel?

Justin:

It's Scrolls over bottom. Our mailing list, and I'll notify you when we go live.

Brian:

Even better, just I guess, you've been putting this on your YouTube channel too. Subscribe to Justin's YouTube, and you'll get it.

Justin:

Yeah. Just yeah. Get notified.

Brian:

Or just follow us on Twitter because we always try to tweet it when we're gonna go live.

Justin:

Follow us on Twitter or on Blue Sky. We don't have anything on LinkedIn right now. But so what Follow me

Brian:

on Twitter and follow Justin on Blue Sky.

Justin:

That's right. I I've I've still avoided coming back. Alright. So this is what you posted. You've got a list of eight agents here.

Justin:

Eight agents configured. You've got Bernard, who's a lead developer, Bink, who's a junior developer, Vale, who's a content marketer. These are awesome. Dude, there

Brian:

there are, like, backstories behind each of these two. I've got, like and and they have, like, personalities and and, like yeah.

Justin:

Okay. So and it this is related to your OpenClaw, MoltBook plot live

Brian:

on a Mac mini, these people.

Justin:

These people all live on a Mac mini using OpenClaw. Take me through it.

Brian:

Okay. So, like, as we sit here today, which is Thursday, February 5, I've been going deep in the rabbit hole ever since so Open Claw, like, exploded while I was at Big Snow, which was was it last week or the week before? Last week. And it was sort of just so so I didn't really have time to do to play with it then, but a couple of us were talking about it there. Yeah.

Brian:

As soon as I got home, I like, literally the next day, like, I ordered a Mac mini, which is not required. I know a lot of people are doing the VPS thing. I just wanted to get any excuse to to spend $600.

Justin:

Apparently, they sold out of them in San Francisco. Like, you couldn't buy one from Best Buy. That's what I heard.

Brian:

I was able to get it delivered the next day, at least at least as of two weeks ago here in the East Coast.

Justin:

It feels like we need a content warning here. Is I 'm seeing a lot of posts about OpenClaw that seem not real to me.

Brian:

There's a ton of everything. It's one of these things first of all, just before we get into my use of it, just in in the the story of OpenClaw is unbelievable. I mean, it's I I don't know if there's some tracking or record of this, but it was, like, the the spike of popularity on this GitHub. Yep. It was just a wall upward.

Brian:

Just Yes. Instant, which is, you know, it's it's crazy because it's like it's so popular, which that in itself makes it a little bit scary. Right? It's it's which means it's sort of like a big target for everything from good and bad and everything in between.

Justin:

We'll see if I can pull up the star history on this thing. Oh, yeah. Here we go.

Brian:

It's that that also means, like, it's also so new and raw, which means, like, security updates are not fully developed yet. So big, big warning flag for anyone playing with this stuff.

Justin:

Yeah. Here's the star history.

Brian:

Yeah.

Justin:

You want to see a hockey stick? There it is. That's that's the hockey stick right there.

Brian:

Yep.

Justin:

So here's what I understand. Let me know if I'm wrong. So my understanding of Open Claw is it's like running with scissors. It's it's allowing an autonomous agent or agents to run with scissors on a machine, often accessing Claude code, but it could be any model. And it basically takes away all the guardrails.

Justin:

And or you can take away a lot of the guardrails.

Brian:

I mean, at the basic level, you know how, like, when you run Claude code, you say, like, you you you Claude code are operating in this project directory for my project.

Justin:

Mhmm.

Brian:

It's essentially like you, OpenClaw, you're you're operating like that, except you're operating on my machine, like, at the machine level. So it there there are there are different configurations you can do, but as but out of the box with with no guardrails, you are basically giving it access to your entire machine. So, like, for like, rule number one, definitely just don't just run it on your own machine. Yeah. Like, the the the one where you do your work day to day or has access to your files.

Brian:

Like, just don't do that. Right? And I think everyone hopefully, everyone knows knows at least that. So then the next the next thought is like, okay, let's give it its own dedicated machine. Like, that's that's what I did.

Brian:

A lot of people went out and got a a Mac mini. Of course, even more people, this is also a valid approach, seems, is to is to get like a VPS account. Like Hetzner is a popular provider. It's $5 a month. You can get a server in the cloud.

Justin:

Yeah.

Brian:

I think Cloudflare, even like the next day, started offering like

Justin:

a Digital Ocean too. One click installs on all

Brian:

these So essentially, it's a box in the cloud where you can run Cloudbot. You you could also do things like Docker, like put it in a Docker container. So what I did was I I I looked into all those options, and I was like, just the the VPS thing, like, I know it's a lot cheaper than than a $600 Mac mini, but I don't know. Like, I I I still just like the old school idea of being able to, like, see my Mac Mini and, like, being able to remote into it and and manage it that way rather than, like, remote into a cloud server through the terminal. I don't know.

Brian:

Yeah. There's just a lot of little clunky things about that that I didn't like. And for I don't know. For me, like, $600 didn't seem like a big deal to to and also, like, if I end up using it as extensively as I plan to, although I'm not yet, but

Justin:

if Yeah.

Brian:

But if theoretic I do have a business strategy case for this that I'm thinking through. If if that comes to reality, then I'm definitely going to use more services than the $5 a month.

Justin:

That's right.

Brian:

Like like the memory and storage is definitely going to increase to $50 a month. And and all of a sudden, it's like, oh, you might as well just get a Mac mini. It would have been cheaper with a Mac mini.

Justin:

Yeah. You wanna you wanna give it some power. So just for folks who haven't heard any of these stories that are ringing around, again, some of these are fantasies, some of these are myths. But here's the one I heard Ben Ornstein tell, which is he he said his friend this is how I understand it. His friend set up Open Claw on a Mac Mini, gave it access to many things, basically unrestricted, talks to it via Telegram or WhatsApp, and says, I want you to book me a haircut.

Justin:

So it goes out and tries to book him a haircut on the web. It runs into a captcha, which automatically it can't continue. But this person had given it instructions to keep going until it solved the problem. So according again, this could be apocryphal. But according to the story, OpenClaw then decided to sign up for or use an Eleven Labs account, synthesize a voice, use the Twilio account that was already there or some sort of way of calling, called the barbershop, booked an appointment using this synthesized voice, and then reported back to the person.

Justin:

So a lot of the stories are of

Brian:

of Open Like, I hung out with Ben at Big Snow last week. Okay. And he was one of the him and, like, another guy, Charles, were, the Open Claw, like, hype guys. Like, they were into it. Like Okay.

Brian:

Okay. And so and so it was, like, hanging out with them and coming home. Was like, okay. I gotta actually look at this. Because at first, was, like, very skeptical.

Brian:

At first, I was not so skeptical of the security stuff. At at first, I was just like, I don't I don't really get what the use case is for this.

Justin:

Like,

Brian:

like, why is this more interesting than just running everything that I can do with my Claude code? Yeah. But it it took me about a week to realize, like, oh, this this for me, personally, this could solve some real challenges that I have in my business,

Justin:

I think,

Brian:

that are actually different from what I do with Cloud Code.

Justin:

So what have you done with it so far? So you've set up these

Brian:

I really nerd out on it. So, like, I'm not actually using it, like, for real work just yet because I've been so consumed in setting it up and learning about it and learning about the security configurations and how I and and planning out my workflows and setting up this multi agent setup, which I could talk about, that I've I've just been in, like, builder setup research learning mode. Yeah. I feel like I turned its corner last week, and and that was the tweet this morning. So I think I'm ready to start to really fire it up.

Brian:

But another big open question we should talk about is the costs of token usage. And the when you initially set up OpenCLaw, it's gonna ask you, do you wanna put in an API key for OpenAI or Anthropic or any of these other ones? Mhmm. Or do you wanna connect your Claude Max account? Mhmm.

Brian:

That that is available. Like, it's very easy to just off authenticate your Claude Max plan, your subscription plan, but that is against Claude's terms.

Justin:

This is what I've heard. This is what I've heard.

Brian:

And so but and, like, you know, the costs of just running tokens, especially if you're using Claude Opus, are just extremely expensive. Right? Like so I I was going really deep over the last week just figuring out, like, okay. If I'm gonna operate this the way that I want my team of agents to operate, I really need to figure out a strategy around, like, how am I gonna optimize the token usage? Because literally, just in the first, like, two days of tinkering with this and playing around with Telegram chats, just a few chats, like, figuring out configurations, like, I rung a a bill of, like, a $100 in, like, a few hours of chatting.

Brian:

Wow. You know? That like, clearly, that's just not gonna be sustainable. I I've landed on a couple strategies. I've I've looked at it from so many different angles.

Brian:

I like, at first, my my initial thought was like, maybe I'll just cancel my max plan and just go all API tokens everywhere. But that's clearly just gonna be way too expensive Yeah. Ongoing. So I think I am gonna stick with my max plan. And I'm and I'm trying to not break the rules.

Brian:

Like, I'm not gonna use my max plan Yeah. With OpenClaw. There there definitely are stories of claw of Anthropic shutting down, like banning accounts. Mhmm. I think as of actually January, they rolled out some sort of detection of, like, if if you are running a bot to operate a Cloud Code instance, like, can detect it and they can they can shut your account down.

Brian:

So I definitely don't want that to happen. Yeah. So but, you know, I do heavily use Opus 4.5 for a lot of my hard hardcore work, like coding and writing, and that's really important to me. So a lot of that, I'm still gonna use do myself. Like, I I have no plans to stop creating on my machine with Cloud Code with Opus.

Brian:

Like, that's gonna continue. When it comes to chatting with my agents through OpenClaw, I have them now all defaulting to a to a cheap model, like defaulting to, like, Haiku. I I might do one of the other ones.

Justin:

Okay.

Brian:

But then you can have them, like, delegate to sub agents. So you can have their sub agent use Opus. So it's like, I will spend So I I expect I'll be spending for both, like API tokens and Claude Max Plan. And, like, important activities that I want the agents that I want to delegate to agents, they can use Opus on, like, a selective basis. Mhmm.

Brian:

But day to day, like, chatting or low level, like, admin stuff doesn't need to use Opus. That can use cheaper models. But then I'm also trying to figure out, like, how can I how can I like because the the big benefit, the big reason why this is different from a Claude code or cursor or is the memory aspect? Mhmm. So this is this is where I think it could be a really big unlock for me in my business is that, like, theoretically, the idea is that I want all of my agents to have access to the same brain, the same memory.

Brian:

And that means, like, every project we ever work on, every every blog post, every YouTube I create, every code project, every commit that I make, every podcast that I record, every transcript of those podcasts. I want all of that flowing into this file system that all of my agents can read at any time. Yeah. No matter what. Right?

Brian:

And up until now, past year, I sort of have that, but there's still a lot of friction, especially between Claude on the web, which I use a ton for writing, for thought partner, strategic planning, creative thinking. I I constantly go to Claude to help me think through stuff. So I have pages and pages of strategic conversations and artifacts in Claude. Mhmm. And then and then I have Claude code, which is files on my system and pushed up to GitHub.

Brian:

But those are not necessarily connected. So I'm constantly passing back and forth. Here's an artifact that I talked about over here. Now let's code it over here. Here's the progress we made on the code over there.

Brian:

Let me bring those files into Cloud and talk about it over here. There's a lot of manual lifting back and forth. And my thought is, hopefully, I can figure out some system where all of that stuff gets logged and stored into markdown files that live in this memory system that all of my agents can access at any given time. And then I can start to have these different agents. If I need agents to work on new features or bug fixes in the background, I can delegate that.

Brian:

If I want them to help me draft social media posts and maybe even post them or or schedule them, they have access to all this content that they can draw from.

Justin:

Yeah. So this is really like the the idea is that this could be the true manifestation of this second brain stuff. Like my second brain really is here. What I'm confused about is, is the context window in OpenClaw actually bigger? And why is it bigger on an OpenCLaw machine as opposed to like, if you're using a model, doesn't it

Brian:

No. The model context windows are the same.

Justin:

Okay. So what what makes the memory better?

Brian:

It's a little bit different because, like, just how the tool works. Right? So in in, let's say, Claude Code, when you when you're in a session in Claude Code, and then you get and then you fill up your context window and you hit clear to clear it out, or or if you close your computer and come back the next day and you don't resume the same session, then it's then it doesn't remember anything Yeah. From yesterday. It has to relearn everything you you give it.

Brian:

Yeah. With this, like, you still have context window limits, but everything is getting logged into files that live on your Got file it.

Justin:

Got it.

Brian:

So the system prompts and this OpenCLOB system that gets installed when you install like, it's all designed around like, it has access to all these files on your file system. It literally like like, there's like a folder called memory. Yeah. And there's a folder called sessions

Justin:

in Yeah.

Brian:

For each agent. And those are JSON and markdown files that just log everything. And so agents can, like, when you start up a new session, I think in the system prompt, it has, like, what you talked about today and yesterday just right off the bat. Like, it's always there. And then it could it could pull from late, you know, further back.

Justin:

Yeah. Could you could you replicate some of this in Claude code? Like, I've tried doing this a little bit where I just have at the end of a workday, my typical process is to say, commit and push what we've done. And I want you to write project notes for today. Unfortunately, right now, it's it's storing those in Claude.

Justin:

Md. But I've I've wanted to take those out and just say, every day, I just want a journal of what we've done. So could you replicate some of this by just saying Yeah.

Brian:

Like that. That's what I'm sort of trying to set up right now.

Justin:

I'm trying

Brian:

to figure out the best logistics for that for me. And I've looked at it from so many different angles because the challenge is that, like, most of those project conversations, I want to be using Opus.

Justin:

Mhmm.

Brian:

Right? Which means I want to be using my Max Plan. Yeah. Which means I have to be doing it in Claude or Claude code. Up until now, I have only used Claude Code for coding, and I've only used Claude for writing and thinking.

Brian:

Mhmm. It's it's I haven't really figured out a good easy way to, like, automate taking my conversations and artifacts out of Claude and logging them or storing them in a file system other than, like, manually clicking it or remembering to do that every single time. Yeah. And so I'm I'm kicking around, like, the idea of, like, maybe I ditch using Claude and move all of my writing and creative and marketing stuff into Claude code, which is a possibility. But that interface is not it's great for coding, but it's not great for, like, non coding stuff.

Brian:

So I don't know. I'm still that's that's an open question. You know?

Justin:

Yeah. Because I think the other thing in my head is I wonder if we can bypass this whole open need for OpenClaw with just a new process inside of Claude code, which is in Claude code right now, I can give it exam I can give it permissions to access other directories, theoretically, and project files. So I could say, for example, right now, I'm in the app directory. But I could say, hey, I want you to go over to the marketing directory. I want us to link these projects.

Justin:

And I've got notes in a folder called memory on both project directories. I've got a folder called memory, and all of my journal entries are there. And then maybe you could you could move to the Claude CLI on your iPhone and then just interact with it just that way as well.

Brian:

Yeah. And I and I also have been thinking about that. It's like, do I really even need OpenClaw? And I've seen other people do this too, which is like like, okay. OpenClaw launched, it and it is a really interesting and fascinating idea for putting agents on, like, their own standalone machine, their own box.

Brian:

Yeah. But do you even need OpenClaw? Can you just run Cloud Code or or build your own whip up your own much simpler version of OpenClaw? I've seen some people doing that. Like Christian Jenko was was tweeting about that.

Brian:

Mhmm. I started to I started to explore that for, half a day last week. I was like, maybe I can just do something custom spun up. And then I was like, oh, well, there's all these, like, nice features that I don't wanna rebuild. Yeah.

Brian:

Especially the the memory system is is pretty well thought out. I don't know. Like, that that might be an evolution that, like, I do and that maybe many people will do over the next year is, like, OpenCLaw was the first of of many different similar concepts.

Justin:

Yeah. Well, in the chat right now, they're asking, what about co work? So how does Cowork

Brian:

That's that's another interesting one. Cowork is sort of similar. You can start to get to like, Cowork is like it does work on your Mac file system. I haven't really played with Cowork too much myself, but my understanding is, yeah. Like, that that could be a good option.

Brian:

I don't know that it has all the memory stuff the way that Open Claw is is set up. Like, I think it's still, like, task based and sort of self contained.

Justin:

Although you might be able to customize it with a plug in and get it to because some of this, it just feels like the it's a process related thing, which is again, when I'm working with Claude Code, basically, I work all day and I'm I'm committing changes throughout the day. But before I leave for the day, I want to commit and push. And then I want to update ClaudeMD if there's anything important. And then I also want to just have a separate journal entry for project notes. What did we do today?

Justin:

That's my ideal. But that could be for me a text expander snippet that I just use every day at the end of the day in the terminal. So I'm wondering if there's like if if for the for the Luddite normies out there like me, if if that's a better system.

Brian:

You know, like but, like, what you're talking about is, like, I spent a day or two over the past week just thinking through that. Like, what is my day to day how am I gonna change my day to day workflow so that I have more of my conversations and artifacts centralized in a file system? Because right now, they're still two separated. And, like, you know, I've I've also been a long time Obsidian notes user. I have, like, years of notes in there, which is a nice markdown editor, and it does save them as files in your file system.

Brian:

So that's nice, except it's lacking Claude. Right? Like, I can't have Claude help me write my my notes. It's not a conversational thing. It's just a note taker.

Brian:

So I so I'm still trying to figure out, like, how can I and I and then I I was playing around with the idea of having a Claude skill built into my Claude account so that when I'm having a conversation, like, trigger saving this conversation somewhere? And but then even that ran into some some limitation of, like, it can't like like, it could maybe save an artifact somewhere, but it can't take the whole conversation.

Justin:

Mhmm.

Brian:

I don't know. So there's a lot of, like, little logistical things. But, like, ultimately, at the end of the day, the vision is and so I I spend a lot of time figuring out the multi agent things. Like, by default, OpenClaw is a single agent. Right?

Brian:

Like, you're you're gonna have one one Telegram chat with one agent. Yeah. And I wanted to have multiple.

Justin:

Okay.

Brian:

So in that screenshot, I have something like six work agents and, like, two personal agents. Right? And the idea is that each of those has a different telegram chat

Justin:

with me.

Brian:

Each of them has a different yeah. I even built like, I had Claude come up with, like, personas for each of them, you know, like like the grumpy old developer, the the junior young guy, the, you know Yeah. Yeah. Yeah. I I use Gemini to to create some, like, silly, like, robot, like, inspired by, like, the Gorillaz band.

Brian:

You know? Yes. Anyway, so so I have I have those. And, like, the idea is that I can I can have, like, different chats? Like like, I have one called Vail.

Brian:

That's the content marketer.

Justin:

Mhmm.

Brian:

I wanna be able to have, like, different Telegram chats. And I also connected them to my Slack workspace so that I can, like, have marketing related chats and tasks with this Vail agent and development related related tasks with Bernard and, you know, strategy with Clover.

Justin:

I mean, I want this for sure. This kind of context context aware persona that I can chat with and that I can do work with. I also think the other the other thing that seems to be resonating about OpenClaw is just the fact that you can chat with it via Telegram. That interface layer actually solves a bunch of problems. Like, for example, everything that you dictate into Telegram now gets saved, and all of their responses get saved.

Justin:

So now you already have a a place to store conversations and refer back to them. And I even like the problem right now I have with all this stuff is it's Siri. It's like the the when I'm using my iPhone and I got my AirPods in, I the only person I can talk to out of the box is Siri. And I can't do anything with Siri except for compose a

Brian:

note. Yeah.

Justin:

I compose a note with Siri, and I get it to set timers and alarms. That's basically it.

Brian:

Yeah. Totally.

Justin:

And so I think the interface Yeah. Layer is what people are starting to I think they wanted a better interface that was better than Siri, better than what they were getting on their iPhone. Well, Telegram's easy. I can interact with Telegram all day. I can interact with their voice.

Justin:

I can interact I think part of the the attraction of this OpenClaw thing is just the fact that out of the box, you can connect it to WhatsApp and Telegram and Slack.

Brian:

For sure. You're right. It's the interface thing. Of course, like ChatGPT and Claude, like, the part of the breakthrough with those a couple years ago was the chat interface to AI. Right?

Brian:

Like, because, like, GPT two and one didn't have that chat GPT layer yet. So, like, that's why it didn't break through until, like, GPT three. But but like, so I think, like, ChatGPT and the and the Clawd's, like like, unlocked. Like, chat is the way to interact with AI. And then and then the models got a lot better.

Brian:

But I think now with OpenClaw, the next like, yes, it's the chat with Telegram or Slack or WhatsApp or whatever you want. Mhmm. But but then it's also the the other connections to everything else you do because it has access to your to whatever you can do on a files on on a computer system. Like like, that's the big challenge for me in my in my business is, like, there's so much, like, okay. I finished a I finished a newsletter that I that Claude helped me write, but then I copy and paste that into Kit.

Brian:

And then I format it and schedule it and, like or, you know, just a lot of little clicking and carrying this piece of content over to there and, like and scheduling it. And and I

Justin:

I think okay. This is I know we're talking long about this, but I want to this is actually something I I think was also attractive about all the mythology and legends and stories that came out of Open Claw is what's the challenge with anything? The challenge with hiring a personal assistant who's a human, the challenge with hiring an employee, the challenge with anything, the challenge with doing work yourself is that there's always you're gonna run into things. Right? So it's like like you said, a typical thing would be what's a very common one that AI just can't touch right now?

Justin:

Like every

Brian:

Video editing.

Justin:

Well, video editing. But here's a simple one that seems simple. Every quarter, my accountant says, Justin, can you log into your bank accounts and export all the PDFs and then send them to me so I can reconcile your accounts? That is difficult to hire somebody for. As soon as I hire a personal assistant for it of all, it's sensitive.

Justin:

But if you get over that hurdle, then you have two factor auth that often depends on me being close to my

Brian:

phone. Mhmm.

Justin:

Then it just goes problem after problem after problem. And I think one thing that appealed to people about this Open Claw thing is it's like, oh, this thing will keep going. If you instruct it to, it will keep going in the Ralph Wiggum loop until it's done. Like, it will just plow through barriers. And I can see why that was attractive to people.

Justin:

And then, yeah, eliminating all these points of friction. So so much of what we do in a day is all this connective tissue between Yes. Like, I compose something in Claude. I might even take it somewhere else and clean it up and write it. And then and then from there, I gotta copy and paste it into kids.

Brian:

Or and and also it's like for me, it also happens when I'm out. Like, I do a morning walk for thirty minutes, and I'm doing a voice recording of of my next YouTube video, like just the idea, just brain dumping the idea. Right? But, you know, I do wanna speak to something else too. And this is where it didn't click with me the the day that OpenClaw started to get popular.

Brian:

And part of it is because of, like, the marketing copy on the OpenClaw website and then what everyone what everyone started to talk about. Yeah. Everyone is looking at it like this is an agent that you're gonna give full unfettered access to your own email account and your own calendar, and it's gonna be like as you and do things as you and have access to all of you in your life. Like, to me, like, no. Like, that is not what I want.

Brian:

Yeah. And that's that's why at first I was like, I'm not interested in that. I have no desire to have an agent reading and replying to my emails for me. I don't have the problem of wanting an agent to book restaurant reservations for me. I'd rather pick the restaurant myself.

Brian:

Yeah. Like like, that that isn't a problem that I have. I'm not interested. That was my that was my initial reaction.

Justin:

Right? Yeah.

Brian:

What made it click for me was, like, think about this exactly like an employee or maybe maybe a team of employees. Yeah. So the way that I'm doing it is I I did set up this Mac Mini. It's it's it's its own user. It has no it's I I didn't log in to my Apple account on it or my iCloud or even my Dropbox account.

Brian:

It had it's not connected in any way from there that standpoint. I actually gave it its own email address.

Justin:

Mhmm.

Brian:

So I can invite that email to Was that just a

Justin:

Gmail address? Or like

Brian:

I used a Proton.

Justin:

Okay.

Brian:

Dot me. And so I, like, I used that to to set up, like, a GitHub account for it so that so I could see, like, commits from it. Mhmm. Yep. I can I can grant and revoke access through that?

Justin:

Did you give it its own credit card for spending?

Brian:

No. Like, I like, that's I mean, I don't know. Like, maybe at some point, I I but I haven't done that. Like, I haven't had the need to do that. Right?

Brian:

It's just like what So if you think about it, like, it's not It's definitely not the same. But when you hire an employee, you are inviting them to do things and have access to things. Yeah. And if they're an employee like in your company, you're probably gonna buy them a computer to work on. It's kind of simple.

Brian:

If you start to think about it like that, you start So for me, I think about it like my company is me and a part time video editor works for me. Mhmm. Other than that, like, I don't have look, Clarity Flow is separate. I have a full time developer who manages that. Yeah.

Brian:

But in builder methods, it's me and a video editor.

Justin:

Mhmm.

Brian:

And I don't have any developers other than me and and coding agents. I don't have a virtual assistant. I probably would if if this were a couple of years ago, I would have hired a virtual assistant by by now in my business for all those little tasks. But I'm convinced that a lot of that stuff can and should be processed by agents. And I haven't figured out all of the processes, but it's absolutely possible to to do that.

Brian:

Yeah. And then same thing with like content and marketing. Obviously, like, I'm I'm the content person, but like I I heavily use AI in my creative process. You know?

Justin:

Like

Brian:

like the idea of like outsourcing to writers is not a thing anymore. Mhmm. You know? And even, like, I I'm gonna I'm gonna do, like, ads funnels at some point. Like, I'm not gonna hire, like, a pay per click expert for that.

Brian:

I'm I'm gonna, like, have, like, AI prepare a strategy and a game plan, and maybe I'll, like, do the initial setup of it, but it it'll be a process that could be automated. You know? Yeah.

Justin:

This is just such an interesting it feels like when it comes to hiring, there's a few different jobs to be done. So the boss sometimes is hiring because it's like, I need someone to take this work off my plate. And this is work that I'm is coming to me and being directed at me or could just come to an email and is being directed an email. But somebody needs to go there and go, okay, the accountant needs this. So Mhmm.

Justin:

Right now that comes to me. And then I could theoretically delegate that to somebody else. Say, hey, this needs to be done. They're basically just taking it off my plate. And eventually, they could be the point person for that.

Justin:

Listen. I could tell the accountant, you just email this person. They will take it from there. So that's one form of work. And that's still, I think, you know, a big part of hiring.

Justin:

But I could see how I could see how AI you might hire it to do some of that work. Right? It's just like, hey, I just need this taken off my plate. The one thing that you're going to miss still is hiring someone who can be autonomously creative and proactive. Because it feels like the these AI agents are still very reactive.

Justin:

They're always just taking direction. Right? They're not Mhmm. They're not able to come up with their own ideas for things that you might do or opportunities or or whatever. So, yeah, I I mean, I don't you don't need to respond to that.

Justin:

I'm just thinking that off the top of my head.

Brian:

That's that's another big challenge that, like, I I run a creative business. It's a creator business. Right? Like, it it's mostly built on, like, me creating ideas for for the YouTube channel and courses and and stuff. I'm also thinking a lot about, like, how can I streamline what I do before I record a YouTube video, like the ideation process?

Brian:

I and and I can't and I don't want to rely on AI to come up with ideas. Like, there are people out there who will just scrape YouTube and just do, like, copycat videos. I'm not I'm not interested in that. Yeah. But I do wanna be able to take my raw ideas and not spend a whole day sitting here, like, staring at the screen, but have more of a process where I could have a chat with my creative agent Mhmm.

Brian:

With with Opus. Yeah. Right? I mean, that this is this is what I do with Opus now. It's like I start with a raw idea, an observation, an an insight, or something I wanna talk about, and I'll brain dump a a voice memo in it.

Brian:

And and then I have a a long back and forth of of creative, like, fleshing it out and then and then writing it and crafting it. You know? Yeah. But, like, that's still, like, too messy for me. I still wanna be able to, like like, I'm not publishing fast enough because I spend too much time in that creative loop.

Brian:

You know?

Justin:

Yeah. I I think what's attractive about OpenClaw is the same thing that attracts me to the idea of getting an executive assistant is there is just a bunch of stuff in my life that it would be great to have that first of all, it'd great to have someone else be able to delegate the work. That's one piece. But even better, a step above that is somebody who is taking who is taking responsibility for that thing, and is just doing it in an ongoing way. And here's another example I can think of.

Justin:

It's like like if if I put AI in in charge of spider control at my house, how would it do that? I don't I don't know. Maybe it would detect, like, number of screams per day or something like that. You know? Like Yeah.

Justin:

My wife screams, oh, that's probably a spider. So like Number of screams per day. But the truth is we hired a spider control guy. Right? And I had to like, it was like my wife is like, there's way too many spiders.

Justin:

We saw a bunch of black widows. Justin, do something about it. So then I take that. I call up a spider control guy. I say, I want you to take control of this problem.

Justin:

Yeah. So I had to do that initial step. He comes in. He scans the whole house. He does it.

Justin:

He sets up traps. And then he says and personally, this is like my love language. He says, and now I will come back and check every year and just see how we're doing. Perfect. Now, I'm not responsible for spider control anymore.

Justin:

This guy's just gonna show up on a regular basis. He's gonna check all the traps. He's gonna do everything he needs to do. And then he's gonna bill me. That's kind of ideal.

Justin:

That this is these are the real problems people are trying to solve. And AI might be part of the solution. But then I'm also thinking how the human element with all this stuff, how does it all interact? You know, like

Brian:

I mean, the way that I think about it also is like like, I don't I'm not really looking to delegate the ideation and creation of new projects Mhmm. To AI, and frankly, even to people. Like, I I am a solo founder of this business. I'm the creator of it. And anything that I create, a new product, a new video, a new course, a new anything, that's what I wanna be spending my time on.

Brian:

Like like, I chose to do this business because I like, yesterday, I spent the whole day creating, designing this app to run my agents. That that screenshot is is a whole app that that took me about it took me one full day to design it and build it. Mhmm. And, you know, it it tracks their tasks. It tracks the token costs.

Brian:

It tracks activity. It it does a lot of things in there, and it hooks into the file system. It's pretty cool. But, like, I there are people out there who are tweeting about, like, how they just task their open their OpenClaw agent to go build build a similar dashboard like that. Just go go do it.

Brian:

Figure it out. Do it. Like, to me, I can't I I I wanna be in the nitty gritty. Like, every UI choice in here was me directing it or crafting it in the way that I want it. Right?

Brian:

Mhmm. Especially if it's a tool that I'm gonna be using.

Justin:

Yeah.

Brian:

So the way that I think about agents is more like I want them to be the next phase. They they are the maintainers. Right? Like like, when I have a new feature for this or for builder methods or for whatever, like, I should be able to voice note it into a chat Yeah. And it turns into a spec.

Brian:

And if it's, like, like, small to medium sized, like, that's something that an agent can and should be just creating a PR for. You know? If it's a totally new redesign or a totally new concept, then I wanna be involved in that. Then I'm gonna do that myself in in Cloud Code or whatever. Yeah.

Brian:

But if but if it's like fixing a bug or or doing this or or, like, if it's a marketing process. Right? Like, if I have a like, there there should be some idea like, some creative stuff where it's like, it has access to all of my activity, all of my projects through these through the file system. Mhmm. And every week, like, it knows that I need to be generating new ideas.

Brian:

So maybe it surfaces like a list of, like, here are some interesting pull like, that I just pulled out of your recent GitHub commit or your recent podcast transcript. Like, these could be

Justin:

Or your books.

Brian:

Maybe they're Bookmarks. Bookmarks. Or, like, these could be, like, maybe something that I said at forty two minutes into the recent panel episode could become could become the basis of my next YouTube video. I don't know. But like, there there's so like, to me, the the big one of the big drivers was like, there's so much activity that happens between me and this computer and these podcasts and the code.

Brian:

And I'm not tweeting all the time. Not broadcasting everything that I'm doing. I'm broadcasting maybe 5% of what I'm doing. And so if I could figure out some systems to extract more publishable content, because I feel like there's a lot of good material. Sometimes I've done that, and it's like I start to do that, and then I'm like, this is so raw.

Brian:

Now I wanna spend an hour crafting it and refining it, and now I'm not actually getting any real work done. So if there's just some sort of process that's happening in the background, like watching and extracting ideas and turning it into at least pretty good drafts that I can approve.

Justin:

You know? Mhmm. Yeah. I I mean, I yeah. I I'm I'm curious to see how this this experiment I'm kind of waiting for everyone else to figure out this stuff.

Justin:

And then but the parts that appeal to me are, you know, the the Christian Genco, like, talk to my phone. And because right now, you know, I can't interact with Claude code on my phone at all. But theoretically, it could be doing all sorts of work for me while I'm just out doing other things. So yeah, I'm I'm curious about all of that. Let's let's switch it up.

Justin:

I want to talk a little bit about building fake products. So you asked me, hey, why why aren't we using that that podcast countdown timer? And I I feel like this was always true. But now the cycle times are so much shorter. The value of Build First has just increased dramatically.

Justin:

Build First allows you to create something in a couple of minutes and start using it and start basically interacting with reality right away. So the podcast timer countdown timer is a perfect example. In that app, we set up segments. We give it a timer. It counts down.

Justin:

As soon as I introduced that with you and I, it it added I noticed it added a little bit of anxiety. It's like, oh, man, I got time now counting down. So you start to feel a real interaction with the product right away. One of the things I've been talking about on this show forever is can we at Transistor figure out a way to do video podcast hosting? And it's just such a big hairy problem.

Justin:

It it feels like too much. And then a couple of weeks ago, said, I'm just going to start building it and see

Brian:

Hold my beer.

Justin:

How it feels. And not not because I want to build something that's production ready that we're going to just drop into the app. But because for this very reason, I want to try start to experience both from the technical implementation side, but even more as a user. What do I run into? What do I want?

Justin:

What do how does this feel? And in the past, you just could not do this. Just to get to the point I'm at right now, I built a full video podcasting app in a couple days, the first version. And just to get to this point with pretty talented engineers would have taken six months, a year, maybe even two years. It would have taken a long time.

Justin:

But I got I've got a little app running.

Brian:

What does this do? Like like, for those who aren't watching, they're you're they're listening. You you just started to show this thing that it looks like it's called AmpCast. I don't know if that's a temporary name.

Justin:

Yeah. That's just a temporary name. This is never I don't I don't think this will ever be a real product. This is just a prototype I've built.

Brian:

So we've got show a list of shows. Can click into one.

Justin:

You can connect it to Transistor and import an existing show. So let's say I import restaurant success here. And then I can connect it to YouTube. And it that works. I can I've got all the awesome connection to YouTube and everything working.

Brian:

What makes it a I'm curious, like, what does it make what makes it a video podcast?

Justin:

Well, this is this is what I'm this is what was so helpful is previously, I thought the the way to do this is we have to do everything. We have to allow people to upload the video to us. We have to encode it for like use HLS encoding, host it, etc. As I got into this, I started to realize that what people actually want is not that. What they want is let me see if I can if I can find this thing here.

Brian:

Like turning their YouTube videos into Basically,

Justin:

they want to upload a video, have it syndicated to YouTube and maybe Spotify video, and then create an audio only version that syndicates to all the audio platforms. Mhmm. It wasn't until I started working on this that this became very clear. Right? So we've been talking I think, you know, HLS maybe should exist for the nerds.

Justin:

But when I think about 90% of people I talk to who are doing podcasting, what do they want? They want to upload one video file. They want it to go to that to get get uploaded to Spotify and YouTube as a video episode. And then they want it transcoded and as an audio file and added to transistor. That's what 80% of people want.

Brian:

Totally.

Justin:

And so I didn't experience that. I didn't fully realize that. That was in my brain. But I didn't fully realize that, honestly, until I'll see if I can find it. But actually, when I was working through Design OS, I had it create a breadboard for the different flows.

Justin:

And once I saw the breadboard, I was like, wait a second. And once I'm talking to Claude Code, and it's like, yeah, like that HLS video component really adds a lot of complexity to this app. Mhmm.

Brian:

That's that's like very similar to my process with any new new app. It's like, alright. Like, here's my concept. I think I wanna do it this way. And then and then you run into things and have back and forth with Claude.

Justin:

Yes.

Brian:

So And then once you see the see the views, see it and like yeah.

Justin:

Well, and there's just so many things that you run into that until you're actually using it, usage has always been oxygen for product realizations, for actually understanding something. So here's another example.

Brian:

So Does it I was wondering, like, does it push out to like, if do would you upload the file to Transistor and then that actually uploads to YouTube and Spotify and

Justin:

In my little prototype, what this does is you just go here. I'll I'll I'll do one right now. So let's just make sure this is connected to YouTube. So I'm gonna I have to keep reawthing it because I'm still in test mode. But for the listener, all I've done is I've opened up a fake show called Super Awesome.

Justin:

I'm gonna click new episode. I'm gonna choose a video file. And it's starting to upload. That uploads to r two. I'm gonna say sample app for Brian.

Justin:

And the other thing I like about this, by the way, is it's allowed me to play with a simple interface again. I even hid a lot of the advanced settings in this expand collapse UI here. If I click publish episode, this is now uploading to YouTube. That's completed. Now it's extracting the audio.

Justin:

That is processed and completed. If we click here and view it on YouTube, still processing the video. But that

Brian:

would be

Justin:

live right away. And you click on transistor. Episode is live. The YouTube embed has automatically embedded. Oh, you haven't you can't see that.

Justin:

YouTube embed has automatically embedded in the share page. The episode is playable. It all happens in one step.

Brian:

Nice. I mean, that's that's beautiful.

Justin:

That's I mean And and here's the other benefit. I've been playing with this. I've been using it.

Brian:

Especially if you could you could schedule it.

Justin:

You can schedule it. I've already got that built in. The I'm now showing this to people. So my one buddy, he he does uploads all the sermons for his church. And he's like, dude, he's like, I wish I could just do this in one step.

Justin:

It's taking me so much time to first upload it to YouTube. Then I basically copy and paste the same title description, etcetera, over to a podcast host and then do it there. I just want it one stop shop. So I I can record a video for him. Eventually, he'll be able to use this.

Justin:

And now I'm gonna have a real person using it and will experience how does this work. Today, for the first time, I I had to we at Transistor right now, we don't have an uploader that can handle files bigger than a gig. So I had ClaudeCode build a file upload file uploader that can handle big files. Today, for the Marketing for Developers podcast, this Lars Lofgren interview was 15 gigabytes or something like that. It's like, okay.

Justin:

Well, let's let's see how this system handles a 15 gigabyte video. I'm also gonna be tracking the costs now on, you know, the different like, for hosting and for bandwidth and all that stuff. I can test all this stuff out before we implement it in our production product. And I think building fake products is going to People need to be doing this. This is this is the way to explore and truly feel and shape the entirety of the problem before you sit down and write a, you know, implementation plan for your existing app.

Justin:

This is like, let's figure out everything now. Even something as small as like you'll notice this one says private. This is a private podcast. And then I as soon as I start using this, because I wanted to publish this episode today, I'm like, wait a second. A private podcast is going to have to have a different setup than a public podcast.

Justin:

And I'm like, what would I want personally as a user? Well, I might just want it to be an unlisted playlist and unlisted video on YouTube. Can I do that? Yeah. So now, if it's a private podcast, it automatically creates a unlisted playlist on YouTube, publishes an unlisted I

Brian:

love it, man.

Justin:

I wouldn't have thought of that beforehand. And Yeah. All of this reminded me of this great story from FreshBooks. This is a story that Mike McDermott told me ages ago. And I don't know if a lot of people have heard about this.

Justin:

But basically, they decided they they had this problem back in who knows when this was? 2017. They had this problem where they had this mature product, but they wanted to do a full rewrite. And so they decided they went through a bunch of ideas. But then Mike said, what if we just created a new company and competed with ourselves?

Justin:

A new company could have its own name, brand, logos, website, articles of incorporation, user agreement, service staff. We could use that to figure out if we could build a product that is truly better than the one we're offering. They built for so FreshBooks v two or v three or whatever it was, they created an entirely different company. And then they built the product ground up from there, got real users using it, had its own pipeline and everything. And then only when it had proven itself with whatever it was, a 100 users, a 100 paying customers, did they switch it over and have it and have it as the main FreshBooks app?

Brian:

I didn't I didn't know that they did that. That's really cool. Yeah. I think really cool. I yeah.

Brian:

Like, this what you're describing, this whole this whole process of, like, you bring a new feature at least in a in a concept or a prototype form. Like, it it it dovetails exactly with what we were talking about with with Jordan on the on the recent episode over here a couple weeks ago. It's like, I'm I'm curious to know for you with with your team at Transistor, like, what is how does this gonna change the process of bringing new features into Transistor? Like, I I just this morning, I recorded my next YouTube video, and it's about the Claude integration with Slack. Mhmm.

Brian:

And and I at first, I was like, that's okay. It's just a little connector with Slack. That's kind of a nice to have. What's the big deal? And then I saw one of the one of the Claude team members post an article on on X about it and how they're using it in inside Claude.

Brian:

And I realized like, oh, there there are some real workflow breakthroughs here. And one what what you were just describing about like, okay, transistor is its own thing. It's it's it's well established code base. It's got this whole user base on it.

Justin:

Like Mhmm.

Brian:

You can't just, like, change change course or ship features, like, in a day like you can with a Vibe coded thing.

Justin:

Yeah.

Brian:

But but what they were doing what they're doing at Anthropic, apparently, is in Slack, you can like, with the Slack connector to Claude code

Justin:

Mhmm.

Brian:

Anyone on the team, marketers, product people, go to market, sales, Anyone can at message Claude and kick off a Claude code project. And then Claude code goes and and works on it and creates a PR to the actual code base.

Justin:

Mhmm.

Brian:

So like so that's an interesting product process workflow for teams. Right? So it's like instead of Vibe coding a prototype, which is very different and new and shiny and separate from what your actual code base is, like, anyone can pull off a branch and and vibe and essentially vibe code. Not even vibe code, but at least just start. Yeah.

Brian:

Like like, what what if we add a video uploading feature to our existing code base? What would that look like? And and could we achieve this? Is that even technically possible? Obviously, it'll depend on your on your own team's, like, strategy and workflow.

Brian:

But but you or anyone on your team could, like, essentially at message cloud and at least start a PR. This isn't gonna ship tomorrow, but at least it's something that it's like, oh, this is technically possible. And and, like, it it could or or we will run into these technical challenges of it. Like, now now we understand that. Then the engineering team can take a look at it.

Brian:

It doesn't even necessarily have to be a code review challenge. That could just be the v one of this or a draft version of this feature so that the engineer can see, oh, okay, I see what I see what the product team is going for here. Yeah. Let me rebuild this properly with so that it's ready for prime time. But, like, that's a much higher fidelity Yeah.

Brian:

Vision than, like, a wireframe or even a Vibe coated prototype that's separate. Like, if it's a PR off of your real code base, you know.

Justin:

I think we're gonna do both. I mean, we're still stepping into this carefully. Yeah. I think we're gonna do both. For this particular project, it made sense to do it separate because we needed a separate r two bucket.

Justin:

I don't wanna use what Transistor already has. We there's a bunch of things that I just wanted to keep separate because this is involves a lot more moving pieces.

Brian:

So you're so you're gonna rebrand to AmpCast.

Justin:

I mean, there is a world in which we go, you know what? We don't want to add this to transistor. We're gonna rebuild it as a separate thing. Maybe a video first podcasting app. That could be true.

Justin:

I I think also the being able to to have something that I can hand over to my friend who's working at this church, and have a limited subset of people actually using this workflow and just running into everything that could happen with use with usage, taking all those notes, and then using that to implement that in the main app. I think that's the other piece. And then the other thing is just being able to see think I've also connected it to the YouTube Analytics API and the Transistor Analytics API. And so now I have a a view of YouTube views versus audio downloads. Sweet.

Justin:

And so being able to do that in its own space and show it to people, show it to different users, friends of mine that I know are actual real podcasters, and eventually getting them to use it. I think there's just a real benefit to that. And Then using everything we learn to inform what we're doing with Transistor. The other reason to do it outside of your regular app is this also led me to to it's still a Rails app on the back end. But I said, on the front end, I don't care what you use.

Justin:

I you you can use whatever you think is best, Claude Code. Yeah. And Yep. You know, we we had when we built the transistor app, we had some strong opinions about how much JavaScript we were going to use, what kind of JavaScript libraries we're going use. Totally.

Justin:

I wanted to go and run with scissors

Brian:

Strip it all away.

Justin:

A little bit.

Brian:

Fresh start. I I mean, yeah, like, I really see a night and day experience between literally apps that I started the code base six months ago or earlier Mhmm. And apps that began within the past three to six months. Yep. Like, the ones that I started more recently are just so much easier to get into and iterate on because AI helped build it from day one.

Brian:

And like, and, you know, going into legacy code bases, even builder methods, which started earlier in the year last year

Justin:

Mhmm.

Brian:

There's a lot of my old opinions baked in there, which which does make it harder for agents to be able to pick up from from that point. Right? Because, like, it it has my hand coded opinions from day one that it has to work around, that it has to try to align itself to. And, yeah, like, my AgentOS sorta helps with that, but, like, it's still a challenge. But, like, it but something like like the app I built yesterday for the agent tracker, like the HQ thing, it like like, I just designed it.

Brian:

And I use Design OS for the design, but it but it it created React components that I'm that I, you know, brought into a Rails and Inertia and

Justin:

Well, here's the other thing I'm doing. Yeah. This is this is the other thing is, again, I thought, okay, I I built this amp casting. And I that was just me working with Cloud Code. Here's the vision.

Justin:

Here's the plan. We went through plan mode. We did all that stuff. But now in a separate Cloud Code session, I'm using dot design OS to re design and architect the same app again, but using a totally different approach. And Yeah.

Justin:

And now so design OS is helping me to really think through onboarding, helping me to really think through all these edge cases. And well, if you if you don't handle this at the beginning, even something like, at what point do you connect to YouTube? At first, I had it on the account level right when you get in. And then as I'm using the first version of my app, I'm like, actually, you want this to be you want to be able to connect to different accounts and channels per show. Okay.

Justin:

Yeah. Well, go into Design OS, and I'm like writing the spec from and we'll see. Maybe the first version I make is actually the better version. But when the cost of building a prototype is essentially just your time, why not do it multiple times? Then maybe the second version I build, I just get more realizations.

Justin:

It's like, oh my god. Like, now and I'm building all these assets. One thing I really like about Design OS, if you're not using it, I highly recommend it, is I didn't realize that you you create all these, like, interactive prototypes. So after a section's shaped, Design OS creates those, like, you can click through on those screens, but they're basically interactive prototypes. You know what I'm talking about.

Justin:

Right?

Brian:

Well, you can yeah. You can view the the designs that it creates.

Justin:

Yeah. I didn't realize that Design OS did that. I mean, I'm gonna bring I'm gonna bring all of those assets when we actually if and when we decide because that's the other nice thing about this is I can build all of this, get it out of my system.

Brian:

Actually, like, since we're talking talking about it, I can pull up the design OS that I did for that thing yesterday.

Justin:

Yeah. I wanna see that.

Brian:

Let me see if I can and I actually tried to record this as a video for for members, but then the the recording crapped out, and I couldn't use it. This is for what a little product that I call BMHQ. It's like Builder Methods HQ. It's it's only for me to use with my agents. And I started with, like, the the overview of it, and then I ended up building out these sections.

Brian:

It's like I like, it's gonna have dashboard. It's gonna have the agents, the activity, a usage tracker, the the cron jobs that it's running, the tasks with Kanban board and skills. Right? So then I went went through and got into the design stuff. I didn't I didn't spend a ton of time on that, but then in into the sections.

Brian:

So, like, here's the dashboard, and then you can view the actual dashboard.

Justin:

Yes.

Brian:

And it's got like a like a, you know, like a responsive thing there. And then it also has like dark mode that you can turn on.

Justin:

Like, this was a game changer just to be able to see things beforehand?

Brian:

Yeah. So, like like, this is like the the dashboard view, but then I sort of need to go back into so like if I go to agents, and then agents has like a list view and an agent detail view. So then if I go to the list view I ended up actually redesigning this a bit in my in my real one. So like the real one looks like this, and I I mapped a domain to it. So yeah.

Brian:

Like I I I sort of like went through the process. One thing like going through it yesterday, I it's been a few weeks since I actually used Design OS on anything. Mhmm. Here's the tasks one. So like if I go to the board view, I've got this like Kanban board where I can, you know, work with tasks.

Brian:

I can put that into a list

Justin:

Design OS just creates this just from your just from the interview.

Brian:

Yeah. Yeah. Exactly. I mean, to get to this first of all, I did the shell. Right?

Brian:

But to get to this, the key insight, I think, in any design process with AI now is, yeah, the interview gets me to this little mini spec right here, which is like user flows and UI requirements. But then I think the key thing is this sample data. The the data models are less important. What's what's a little bit more important is generating this sample data Mhmm. First before you design it.

Brian:

Yep. So have AI like, get a get a really raw, rough concept of like, okay, we're gonna do this Kanban board. It's gonna need to have tasks and assign them to agents. And if we were to generate some JSON data off of that, here's what that data would look like, here's the shape of it. And then once it has that, it's able to create a view that takes all of that into account and puts it into an interface.

Brian:

Because if you skip that step, then it's going to sort of dream up all these details, know, and then it's not gonna be quite right. Yeah. Yeah. But that like, going through it yesterday, I I did find like, now I'm thinking, like, at some point, I wanna do an update to Design OS because it it still took too long.

Justin:

It does take a long time. That I I I I did have a feeling of like, ugh. Like, I this took a while.

Brian:

Yeah. And, like, there's a lot of these steps where I'm just, like, even going through the spec and the sample data, like, it's important, but it's not even important that I review it. I just want it to be done and get right into the design part. And then the other part that's a little bit too heavy right now is the export. So when I get to the final export where I export out these React components and and the views, it's a little bit too prescriptive in terms of like like, yes, it gives you all the design React components ready to drop into your real app.

Justin:

Yep.

Brian:

But it's also including in that, like, all this, like, prescriptive, like, here's how you should model the database, and here's how the back end logic should work. And I and I don't want it to do any of that. I want the I want Claude code in the final app to be able to take these React components and just the high level feature requirements Mhmm. And then that agent to decide and help you help you figure out the actual data model.

Justin:

Yeah.

Brian:

So there's I think like the net there is gonna be another version of Design OS at some point that strips out. Just like I did with Agent OS version three, I stripped out a lot of the bloat. Yeah. I think going through it helped me realize that there's a lot of extra bloat that can be removed.

Justin:

Yeah. I'm a big fan. I think even just the just even having the just even having the process is really helpful.

Brian:

Yeah. I got a quick thing, but I do wanna hear about Lars Loughran Loughran.

Justin:

Let's do a quick thing. Yeah.

Brian:

Well, this is just like a,

Justin:

I don't know, like

Brian:

a mindset, like a challenge thing that I that I've been feeling a lot lately. And part of it was what led to wanting to do this agents thing with OpenClaw. But, like, I'm a solo founder, solo creator. The business is going really well. You know?

Brian:

So I used to feel like last year, it was more of the urgency around, like, I gotta keep hustling and pushing to to make this business happen so that I can get revenue going, and then I feel a little bit more secure. Like, that's I've moved on from that. I feel I feel pretty secure with this business now, and and it and it's pretty smooth sailing and definitely growing. But now I still find myself working all the time, and I still have this urgency. Mhmm.

Brian:

And now right now, the the challenge, I think, is this like, there if if we split my whole work life into two halves Mhmm. One half is the public facing you can call it marketing. Like, all all of my work on YouTube and email, like, weekly email newsletter and this podcast

Justin:

Yep.

Brian:

Or and my and my new Builder Methods podcast. And, like, all of that is, like, public facing stuff.

Justin:

Yep.

Brian:

And especially with the wave of competition that is rolling into my space

Justin:

Yeah.

Brian:

Here in January, February

Justin:

Yep.

Brian:

Like, I constantly feel a pressure to keep keep that pedal to the to the metal. Like, keep pushing on growth and exposure and audience and lead flow and funnel. And I've got a funnel that's working really well. Every time like right now, like since I had the big snow trip, I couldn't record and publish that week. And then the week after, I still didn't have enough time to record a new one.

Brian:

So there's gonna be a two week gap in a new video on my YouTube channel. That's way too long for me in my space right now. At the same time, the other half of what I do is delivering a Cloud Code course, delivering private content for my members, interacting with my members every day, and then actually building stuff using my own tools, Design OS, improving Design OS, improving Agent OS, Yep. That is so incredibly time consuming. And then not to mention, I need days per week just to tinker and research and learn.

Brian:

Like, I'm gonna do a big YouTube video on OpenClaw at some point based on all this learning, But I need the time to do to do that learning.

Justin:

Right?

Brian:

And so it it it's really challenging right now.

Justin:

I think I mean, I think you should do a video just on this topic because that pressure you just described I mean, your pressure is unique in a sense because of the business you're in. But I think everybody's just feeling this pressure. Like, this is the land grab moment. This is the the trains taking off. And if I'm not on the train, I'm gonna miss out.

Justin:

Everybody's just thinking, I gotta build all the time. It's like we're all addicted to drugs right now. And Mhmm. That it's exhilarating. It's fun.

Justin:

And a lot of people on Twitter are talking about the fun part of it. But we're starting to also get some of the reflection like Caleb Porzio just had on his podcast, Notes on Work just had. He's like, is this better? Am I still having fun?

Brian:

I heard that. I actually have some separate thoughts about that. But but the yeah. Like like like, one of one of the things that has this tension that that I that I felt for a while now is that, like, yeah, there's a lot of guys out there who are tweeting nonstop every day about Mhmm. Like, claiming to to be the the AI guy for for this or that.

Brian:

Right? And but I think that my whole angle on the on the YouTube channel is that I'm a little bit more thoughtful. I take a lot more time to create. And I only create a video when I actually have something to say about something. I'm not just gonna copy any popular topic that's out there.

Brian:

Yeah. But at the same time, like it's that Jerry Seinfeld thing. Right? It's like, yeah, you I don't know if it's Seinfeld or some other person said this, but it's like, yeah, you gotta like create your best like creative ideas, but you have to have those creative ideas at 8AM every single day. Yeah.

Brian:

You know, like you still have to have a schedule to to make sure that this thing is actually still producing content, essentially. You know?

Justin:

Yeah. So so you're feeling the pressure of just how do how do you stay on top of it? Things are changing so fast.

Brian:

It's it's really even more like I what I actually want to do is deliver twice, five times, 10 x more value to my pay to the paying members. Like, if I could spend all my time or even just most of my time inside Builder Methods Pro creating and giving them my full attention and creating tools and creating private videos that show really interesting projects. Yeah. Because what's fun about those types of videos is that I don't have to polish them for YouTube. Mhmm.

Brian:

And they don't have to be so scripted. They can be just like, watch me build this really cool thing. Yeah. You know? Or teach a lesson on Cloud Code.

Brian:

Like like, that's actually more I I wanna be delivering even more value to my members. And, yeah, like, there's a lot of stuff, but but but, like, this it's it's still this, like, I I feel an obligation to the funnel. Like, I I can't just, I can't, like, lay off the funnel too. You know? So it's it's just really, really difficult.

Justin:

I mean, I think this

Brian:

And then, like, not to mention the sponsorship stuff. Like, I I get I I could probably triple my revenue next month if I just open the doors to sponsors. Yeah. You know?

Justin:

Yeah. And is there a way of doing that that won't kill you, that won't overwhelm you, that won't

Brian:

I was talking to a sponsorship coach, Justin Moore Oh, know Justin. About about this. And yeah. Like, he he he was really helping me out with some strategy on how this could work and and what kind of value is or like untapped value there is for for my channel and what I'm doing. And Mhmm.

Justin:

And I

Brian:

and I plan to do it. I I should be doing it now, but it's just a it's I I don't have the bandwidth to insert the sales process and and and the and the operation for doing sponsors just yet. You know?

Justin:

I mean, this is the challenge in any market, but a market that's a rocket ship that feels like it's taking off. That's my favorite Peltie quote is he he said he launched Balsamiq, the wireframe app, and there was so much demand in the market. He felt he said it was like holding on to a rocket ship by my fingernails. Like, that's how it felt. And it sounds like you're feeling the same thing.

Justin:

It's like, you know, the rocket ships there. That's often the problem is the rocket ships not there. Right? You're just like trying to trying to make something happen. And it's like, this thing's not taking off.

Justin:

But then the reverse of that problem is like, this thing takes off, and you're just holding on, hoping that you can keep holding on. Right?

Brian:

Yeah. Yeah. Well, I'm curious if Lars Lofgren has any wisdom that could that could help with that problem.

Justin:

I mean some of Some of this is going to some of this will probably make your life worse. Probably. Yeah. So let me

Brian:

actually He's a good I I can't wait to hear that episode, but he is such a good speaker and communicator. You know?

Justin:

Yeah. Yeah. So let me let me see if I can play a little clip from that because I think this is one of the big takeaways I had. Let's see if this works here. Okay.

Justin:

I won't make this live.

Justin:

Google isn't just looking at links in order to figure out entities.

Justin:

Mhmm.

Justin:

They're looking at the entire web. Yeah. So they're looking at Reddit. They're looking at YouTube. They're looking at social.

Justin:

They're looking at everything else. If you go build a website and you're just relying on SEO and links, like you don't have that footprint everywhere else. Right? You don't, you don't have reviews. You're not on Trustpilot.

Justin:

You're not on G2. You don't have a Google business profile. You're not on Reddit. No one's talking about you because you have a brand new business. Like you haven't built up your brand, but more importantly, haven't built up that, like, entity across the web.

Justin:

And if Google's looking at your shit and says like, wait, who is this entity? They won't even look at the content. Yeah. They'll just say, hey, we don't know who they are. We don't know this thing.

Justin:

So we're not going to rank them.

Justin:

The this is this was one of the key insights. And I and I'll I'll synthesize it again. Today, the challenge with LLMs, and this is for personal brands like yours, this is for companies, is now LLMs and Google aren't just looking at, oh, Brian Castle. Let's go over to briancastle.com and see what's there. No.

Justin:

They're not just looking at transistor.fm. No. They're not just looking at the backlinks to transistor.fm, which was in the past, all you needed was a bunch of authority links linking to your site. Today, they are looking at your overall reputation as an entity. Meaning, if all you have is a website, they're going to look at you suspiciously.

Justin:

But if you have a website, plus 100 reviews on Trustpilot, plus a bunch of real people discussing your brand, product, etcetera, on Reddit, plus a bunch of people doing the same on YouTube, plus a bunch of people doing the same on LinkedIn, plus it just keeps going and going going. It's the footprint that matters. And this also has to do with personal brands. So one of my favorite things to do these days is to go into one of the LLMs. I actually like Grock for this because it's connected to Twitter.

Justin:

And I'll say, what is Brian Castle known for? What is Justin Jackson known for? What is Ian Landsman known for? And you'll see how sticky some of this stuff is. So if all the if if the primary thing that LLMs and Gemini have on Brian Castle is that he's the process guy.

Justin:

He's the productized service guy. That's what they're going to list out. So if all your reviews, if people talking about you on Reddit, people talking about on you on Twitter, etcetera, that's what they're going to talk about. If your brand if your product has done a pivot, like Transistor did from being like the podcast hosting for brands and professionals to just podcast hosting for prosumers and creators, that needs to be everywhere. And your brand needs to be everywhere.

Justin:

And he's and Lars is just like, this is this is foundationally changing everything. It's the reason he's way more active on LinkedIn now as his personal brand. Because he needs people interacting with him and more information that now when the LLMs scrape him or their their sense of who he is as an entity, it won't say Lars is the CMO at Kissmetrics. He wants he wants them to know something different. And the way you do that now is you have to manage a very wide footprint across multiple platforms.

Brian:

I get the concept. The question I think everyone is probably thinking is like, how? What is the how do we do that realistically? Even if you're a company with a with a lot of resources, that's still super hard. Right?

Justin:

I mean, some of this is gonna depend on personality. So the thing that Lars and I discussed is he's like, basically, this system's perfect for Justin, because I've always taken this kind of gardening approach to marketing where I'm just walking through the garden. It's like, oh, there's my Reddit part of the garden. I'm just gonna go and help that to grow. And I'll go over here.

Justin:

Here's Trustpilot.

Brian:

I noticed that about about you. Like, you're obviously very out there, especially on like podcasts and YouTube, but you do jump around platforms a lot more than people might realize. Like, you you show up in a Slack channel, and then you show up on Blue Sky and then in Reddit, and it's like, dude.

Justin:

I'm just I'm just walking around the garden. And I got lucky in this case that this approach just works awesome. And then stuff I used to garden in the past, like Quora, is now a ranking factor again. It was a ranking factor like ten years ago. Now it's not as much of a ranking factor, but I invested so much of it in it.

Justin:

Well, not ten years ago.

Brian:

But When you say walking through the garden, is that like literally throughout the day, throughout the week Mhmm. You do is there any process to this, or is it just random? Like, I'm gonna log into my Reddit and see my notifications. I'm gonna log into my Twitter and see structure. My

Justin:

This is gonna maybe I mean, you could probably do it in a structured way. I'm I look at Reddit every day. I look at LinkedIn every day. One of the reasons I'm trying to do more podcasts, and this is Lars' approach too. The reason by the way, the reason that all of these AI CEOs are still doing podcasts is exactly this reason.

Justin:

They know that the more stuff like that they do like this, the more it feeds in to the models. They're doing this on purpose. And so it is for me, it is just like, I have a rhythm to my day and my weeks, which is just yeah, I'm gonna go around and just keep and then every even to the point where it's like, every quarter, every half, every six months, I'm gonna go back. I don't spend very much time on Indie Hackers. I'm gonna go back to Indie Hackers, just update some of our stuff.

Brian:

Yeah.

Justin:

I'm gonna go back to Product Hunt, our initial Product Hunt launch. I'm just gonna update a bunch of our stuff. In our in our next newsletter, I'm gonna ask I'm gonna put a PS and say, hey, we're a small company. We really need some reviews on Trustpilot. This is just all the approach of just moving things forward in different areas throughout my day and my week.

Brian:

I'll tell you how my vision for how I want to approach this. I'm totally failing at this right now. But what I really want to happen in 2026, sooner rather than later, is I think that this is like a general strategy that maybe other people have been using and and can use, which is like focus on your primary one or two things where not only is it natural and easy and and placed to your strengths as a creator, like for me, it's like video and podcasting. Right? Mhmm.

Brian:

Like YouTube and podcasting. Like the but also like, it's also where you like to hang out. Like, I like to watch YouTube. Mhmm. You know?

Brian:

Twitter, I I like to scroll it and I tweet from time to time, but I I'm not a high volume tweeter. LinkedIn, I never touch. Facebook, I never touch. So I'm naturally terrible on those platforms, like from a strategy standpoint. I I do like, it's the age old, like, repurposing concept.

Brian:

Right? And I I've been doing a little bit of that, like, where we cut my my editor will cut shorts out of my long forms to to but I'm only publishing them as YouTube shorts. I should be publishing them on other places. Mhmm. I should be so my my thought now is again, I I I do I really do wanna use AI agents for this.

Brian:

This might be, like, the number one use case for me for for use for for OpenClaw is, like, I'm still gonna create content the way that I I've always naturally created it, YouTube videos and podcasts. Yeah. And I don't wanna change that. I like doing that. That's that's always where my best ideas are gonna come out.

Justin:

Mhmm.

Brian:

And I want those to come from me, the human, Brian. And so, like I was saying before, even if it ends up being me or even if I create social accounts for my agents, which I plan to do. You know? Like, I still want them to be able to like, what you talked about, like, like, you you jump around and log in and tend to the garden on Reddit and tend to the garden on Quora and on Blue Sky and on YouTube. I don't do a lot of that.

Brian:

I just stick to my one or two channels that I just like to consume anyway. But if the agent can prompt me to say like, Hey, you haven't posted to Indie Hackers in six months, But here's something that might fit for you, or here's something that you should reply to. Yep. You know? Like, that's something that, like, a a good team member would be able to watch and prompt me to go do.

Brian:

I mean, and

Justin:

I would use that too. I think that would be helpful for me as well. I think an easy way for anyone to start is to just like I just said on Google, what is Brian Castle known for? And

Brian:

Product I service

Justin:

was The interesting thing is, at least on my screen, the the number one kind of feature snippet is from tuts+.com in battle. So there What? There's a

Brian:

I mean, that's like fifteen years ago.

Justin:

This is why you need to start googling yourself, asking Gemini chatting. What was it?

Brian:

Like a like an article that I that I guest wrote in '20 Yeah. '11?

Justin:

Yeah. Here's a here's a nice fresh face of

Brian:

How to use WordPress or something

Justin:

like that? Writing and designing a killer headline and how to scale and grow your online business by systematizing. So that's one thing they pulled. Then your website, then your LinkedIn profile. Brian Castle Builder Method.

Justin:

So at least you've updated

Brian:

your you I've the company that I work for.

Justin:

I'm noticing a lot of people I won't mention names, but there's a lot of people who have not updated their LinkedIn profile. Even just to update your job and your company will help a lot with this.

Brian:

Yeah. Like, you're you're right. Like, because even just people that you meet in the world, like, they're gonna Google you and and and also, like, professionally, like, people are gonna just check you out on LinkedIn. Like, what what company is that person from? Like Yep.

Brian:

Yeah.

Justin:

And I I also think it's it'll bring up some other things like your Mixergy interview from 2017 is on here.

Brian:

It I think And that's like eight years ago.

Justin:

That's a good prompt, though, to say, oh, you know what? I should just reach out to Andrew and just see if he wants to do a follow-up episode. Yeah. Because then we have an updated version of this thing that's already getting pulled. So yeah.

Justin:

And even like some of these past interviews, like there's one from 2022 where it says you're the founder of Zip Message. If you know that person, you can just write a quick email or get your agent to write a quick email that says

Brian:

It's so tiring to go out and get and, like, change all those names and links and, like

Justin:

Just but even just to write the email to at least say, hey, I just wanted to know if you could update your your the interview with something. And 10% of the people will do it.

Brian:

But And I've I've received those emails from you know, because we had random links in our Clarity Flow blog. And like, I'm too lazy to go update it. Yeah. I'm gonna I'm gonna ignore that.

Justin:

I mean, this is part of the work right now. And so, you know, there's part of gardening that sucks. Pulling weeds is not fun. So another ranking factor that's just showed up in the last couple months is now Facebook discussions and groups are ranking factors in Google again. And I don't like Facebook.

Justin:

But that's just part of the garden. I got to go on there if I want to win in this space. And the way people are finding transistor is by googling podcast hosting and seeing what and they're like, oh, Facebook discussion. I got to be there. That's just the job.

Justin:

I think you can

Brian:

But I also think that, like, fundamentally, like like like, yes to all these things that we're saying that you gotta go technically go to these websites and be active or respond to stuff. But that's not enough. Mhmm. To me, the the the way more important than any of that is just be interesting.

Justin:

Yes.

Brian:

You've got to be doing something And people hate to hear that kind of thing. But the thing is, your product has to be interesting now. It can't just be another product that does the same thing as all your competitors. You have to know the reason why it is interesting to the customers who love your product the most. Right?

Brian:

Yes, that sounds like salesy and marketing and you're self promoting, but at the core of that, you're understanding why why your best customers are paying you the money that they are paying you is because they really resonate with some idea and you you peel that back 10 layers deeper, then you get into, like, content that is not sales y, but it's act it actually means something to people.

Justin:

Mhmm. You

Brian:

know? Like, that's that's where that's what people actually connect with. I mean, you know, like, it that you you have you have to just really, care on that level. But once you do, then then you can start to find the opportunities. Like, where where does it make sense for me to, like, actually be helpful?

Justin:

Yes.

Brian:

You know?

Justin:

This is another we're we're probably wanna wind down here. But the other key part so interesting from that Lars Lofgren interview. I think this will be a an earthquake for some people, and some people have already realized this intuitively. He basically said, if you're going to do organic now, really one of the only approaches that works is founder led marketing. He said that was that's always been a superpower.

Justin:

But now it is the interesting founders who are being themselves on Twitter, LinkedIn, YouTube, whatever, who are generating kind of the the the best organic footprint to get found on Google, but also via LLMs. And his the the big line that that I shared, I think it's at the beginning of the episode, is he's like, you either have to be the influencer or you have to hire the influencer. He's like, that's those are the two options for organic these days. And he said, if I had a SaaS company with a marketing team, basically, every person I hire on the marketing team would have to agree to posting personally on some social network. So not he said branded channels are dead.

Justin:

Like, your branded YouTube channel, like your company YouTube channel. That's not and we've seen this. Like, go to

Brian:

go to There the are some branded stuff that comes up where it's still a person.

Justin:

That's right.

Brian:

Like, clearly, they hired somebody and and, like, their video is answering a question that I happen to be googling

Justin:

Like, Lyricast is basically Jeffrey Way. But if you go to the Tailwind channel versus Adam Wathen's YouTube channel Yeah. There's a big difference in terms of engagement as well.

Brian:

To that point, though, the definitely, I could see how, like, the founder led thing is the thing that works best. But I don't know that I totally like, yes, it's technically possible for a company to, like, hire a YouTube influencer type person to be the your sole job is to be the person who creates YouTube I know, like, companies hire, like, developer education educators. Like, I guess that's a thing that sort of works. But Anthropic, I think, is killing it on this front, where they have their whole real team. Is what I'm talking about, Twitter accounts.

Brian:

And they have super interesting Twitter accounts. And you can even tell that, like, members of their team are not long time Twitter people.

Justin:

Mhmm.

Brian:

Like Boris Cherny, the creator of Cloud Code. He just joined Twitter, I think, three months ago.

Justin:

But

Brian:

he's not the only one. There's more than 10 Anthropic employees on Twitter. And yeah, they're promoting Cloud product updates, but they're talking about what's interesting about them. And a lot of them are like the person who is responsible for this new feature is the one who writes the tweet about what was interesting about it. Yep.

Brian:

And I mean, I started creating a list Twitter to follow just those people. The Cursor team is also pretty good at this. I think they have a a fewer number of people on their team doing it, but this is like real content talking about what's actually interesting about their product. Yeah. The real problems that they solve.

Brian:

It's like real so you don't have to be the CEO. You you can have your team. Yeah. It goes back to this thing.

Justin:

Does that. Like, the the the person that everyone sees representing Ahrefs, I can't remember his name, but that's not the CEO. That's that's their CMO, I think.

Brian:

I didn't even know that until recently. Yeah. Right? Like, I I like, for many years, I thought he was the CEO. You know?

Brian:

Yes. Yeah. I I But there there are

Justin:

Another company that's been good at this forever, Basecamp. You got Jason Free. Yeah. You got DHH. You got Ryan Singer.

Justin:

You got Jameis Buck. There's all these old names of people that were representing more.

Brian:

Some of the some of the guys, like recent like, with their recent stuff, I feel like they're Jason and David are seem to be a little bit more open to having their team be more visible.

Justin:

Mhmm.

Brian:

Like, sharing like, the designer behind Fizzy or whatever shared a thing and like

Justin:

It's be and and and by the way, the reason it works is the thing that you and I have been talking about basically since the beginning of this podcast, which is in especially in an age of AI and synthesized content, human contact, human connection, being interesting as a human online is going to win. And so if you have the whole Anthropic team posting on Twitter all these interesting things about their work, about what they're discovering, little hacks. Oh, here's how we work with Claude. And here's a nice little list. Of course, that's gonna do well and way better than if someone sees like the Claude code account.

Justin:

It's like, okay. Some corporate shill is just shilling out

Brian:

Mhmm.

Justin:

You know, whatever. No. People want real human beings. Real people. They wanna hear from Richard Branson.

Justin:

They don't wanna hear from Virgin Mobile Mhmm. Twitter account. Right? So Yeah. This is And

Brian:

they they clearly have different channels. Like like their their Twitter game is really aimed at the developers using Clog code.

Justin:

Mhmm.

Brian:

But then they're gonna do the Super Bowl commercial that where they're which is fantastic, by the way. Like, the hits that they're they're doing on OpenAI for doing ads.

Justin:

Yeah.

Brian:

So that's obviously for the masses, for the consumers. You know?

Justin:

I Yeah. I think this is gonna be a very interesting time. And, it's disruptive because asking your whole team to post on their personal channels is a big ask for a lot of teams.

Brian:

You mentioned Caleb Porzio.

Justin:

Oh, yeah. Yeah. This episode. Yeah.

Brian:

I want to speak a little bit to to what he was saying, but this is more general because I hear this sentiment come up

Justin:

a lot.

Brian:

Right? Which is, okay, all these people keep talking about AI every day, but where are the products? Who's actually building products with like real products with AI, real SaaS tools with real customers? Like where are all the products? Right?

Brian:

Like I hear that all the time. Number one, there are real products that are 99, a 100% coded with AI today. Mhmm. I mean, I started a new show called Builder Stories, interviewed Arvid Kal. 99% of his code is is AI.

Brian:

He he's he's doing serious MRR with his pod scan. Mhmm. Talked to Brennan Dunn. Same thing. Write message.

Brian:

Like, yeah, it started years ago, hand coded. It's Brennan Dunn solo, him and agents doing all the work on this is a real product with tens of thousands of of MRR. Yeah. I I I have weekly episodes coming out showing real builders, how they are really I talked to Colleen Schnetler last week. That one just dropped.

Brian:

Real builders doing real product. It is really happening. I I just did a private workshop with a team who who's doing serious MRR for for a niche industry.

Justin:

Mhmm.

Brian:

This is a team of eight engineers who are all using Cloud Code. It like, it it might not be the loudest thing you hear on on Twitter, but, like, it's real now. Like Mhmm. You know? And and there isn't gonna be this line in the sand and saying, like, that product is AI and that one is not.

Brian:

All the products, all the engineers are using Cloud Code now. It's just a fact of life. That's the first thing I just wanted to say. But I was hearing I'm such a fan of, first of all, how not only Caleb's work and and and the products that he does, but, like, the way that the way that he does this solo podcast, and he's so open and and honest, and and it it puts so much personality into it. I really respect that.

Brian:

He's really good at it. And I've been hearing his episodes of I especially tune in when he's talk when he talks about his journey with AI lately.

Justin:

Mhmm.

Brian:

I think Adam Wathen's journey also plays into this.

Justin:

Mhmm.

Brian:

You know, Caleb on his on his recent one moment was talking about is is AI really actually speeding me up? Yeah. It makes a lot of things easier, but but what I've actually built and shipped this even faster, I could sort of see how that would be the experience that someone like Caleb would have

Justin:

Mhmm.

Brian:

With it. And I'm not saying that he's using it at any, like, rudimentary level. He's probably using it super at a super deep level. But the difference between what he is doing and what most other product teams are doing is that Caleb's work his Caleb's product is code. Livewire like, his product is Livewire, is code, or or Alpine or, you know, like, Flux.

Brian:

Like, he is creating code products, which means he has to be super hands on with exactly how his products are crafted at the code level. Yeah. But product people who are creating job to be done interfaces that real customers not that his customers are not real, but, like, business you know, businesses have use cases to use SaaS tools that that you know, like, your customers on transistor don't care about how the code is crafted underneath the inter interface.

Justin:

Yep.

Brian:

You know? And that's obviously not to say that like it's bad code or people building with Claude code don't care about the architecture. They obviously do. And it's gotten much better now. But there's just no question that teams can ship and build real products for real customers like a thousand times faster now.

Brian:

It's just the nature of the product. Like if you're using AI to to refactor a service class that that that that, like, powers one component of a code related product

Justin:

Yeah.

Brian:

That you're only touching a tiny thing. You're you're not you're not actually shipping an end to end solution for a job to be done product. Like, that those it it just really comes down to the actual use case around around the product. Like, so yeah. Like, if if you're really just using it to as like a at at the code level

Justin:

Mhmm.

Brian:

Like, I could see how that would that would slow things down. You know?

Justin:

Yeah. I think I mean, the thing I appreciate about what Caleb was saying, which I think we do need we we need a counterbalance to all the hype and excitement and put this into my veins and, like, let's go, is is some thoughtful reflection of, like, how do I feel about this? And and and, I mean, the challenge and there's still lots of unanswered questions. Like, what we don't know like, for sure, there's gonna be, you know, six months, twelve months, thirty six months from now. There will be a a whole bunch of these apps that have been built with Claude code, not even opening an IDE.

Justin:

Nobody really knows what the code looks like. That will start to have problems that all of a sudden Cloud Code can't fix.

Brian:

And I would push back on that, though. Yeah. Because I do think that these engineers do know how the code is architected. They're just not writing the lines of code. There there's a new there's a new type of engineer.

Brian:

There's the there's the vibe coder who has no idea Mhmm. What their what their code base does. There's the hand coder who's resistant to AI, and they are they are behind the times now. It's 2026. You are not doing it the professional way if you're not using AI.

Justin:

Yeah.

Brian:

But there but the the more the now it's becoming mainstream where you are a professional full stack engineer, designer, developer, product person. Mhmm. You know how your product is architected. You know how database modeling works and interfaces and user experience. You know you know how all those dots connect.

Brian:

You you you are putting that knowledge into specs and into prompts and into and and using Claude code to bring those ideas to life, like, so much faster. Yeah. And like, that's the thing. Like, the they they can write code. They can read every line of code.

Brian:

And a lot of times, they don't have to read every line of code to to trust that they've crafted their Cloud Code systems in a way with the right training, the right prompts, the right systems, the right specs. Yeah. They they they they're now at a level that that they trust that it's actually writing and and architecting things the way that they expect with without even needing to to of course, they have, like, code review and test suites and and and QA processes. But, like, yeah, there's of course, there's a lot of code that I don't review that I still ship. You know?

Brian:

But I know how it's built. You know?

Justin:

Yeah. I I I can understand, though, people's feeling of, like, where's this going? Because there is this let's see. Even put the code issue aside. There's still there's so many like, even me, I'm just like, like, our

Brian:

It is scary. Especially the the open cloud.

Justin:

You know, I've I've always said, listen, people, you can vibe code an app all you want, but the true product people will still stand out. Maybe that won't be true in six months. Maybe there won't even be we won't even need apps because everything will be an API that is just talking to your everyone's personal Cloudbot. And we will have no websites, no apps, no SaaS, no anything. Like, that's all theoretically possible.

Justin:

Right? Like, why why have podcast hosting when you can just get Open Claw to just like, hey, build me this thing, and I'm gonna record the first episode now?

Brian:

I think it's like yes to all, really. I think that I think I think it's like on the one hand, I I think it's overhyped that SaaS is gonna go away so quickly. I don't think it's gonna go away so quickly, especially for well established companies and especially for companies that serve businesses that and and businesses it's almost like the larger the company, the more behind the times they are with Yeah. With adopting internal AI AI solutions. Right?

Brian:

But at the same time, I I do I know we're rehashing stuff that we've talked about before, but I do think that there's gonna be more there already is more of a any new idea that comes out, everyone has has the option of, like, go use that new product or just vibe code a similar thing that's perfect.

Justin:

Mhmm.

Brian:

Like and and I think that pattern is gonna play out again and you're already definitely seeing it in open source stuff. Like things like AgentOS, there is like a thousand other AgentOS. Yeah. You know? And it's literally like every developer's, like, personal opinion on how spectrum and development should essentially work.

Brian:

It's like and, you know, and that's that's part of it is like, that's great. Like like, don't like, I was talking to Bruno Bornstein on on the Builder Stories podcast yesterday. Like, he he just created an open source thing that that creates a Kanban view in in the CLI. It's like a CLI Kanban coder thing. Yeah.

Brian:

And and it's just like like because that's how his brain works. You know? And and it's like

Justin:

But and so here here's the this is

Brian:

So, like, like, fit like, Fizzy or even, like, even the video podcasting thing, it's like, yeah. I I look at that. I'm like, I could buy that

Justin:

Mhmm.

Brian:

Or I could vibe, coach.

Justin:

Or I could build it. Yeah. And and here's my thinking. In the short term, I think I I have a fair amount of confidence that a, distribution is gonna matter more than anything. So the reason Agent OS is more important than all the other ones is Agent OS has more distribution.

Justin:

More people know about it. That is gonna be an advantage. But the other thing that kind of worries me, and I'm sure other people are feeling this intuitively. And I think it's one reason everyone's so panicky and like feeling all this pressure right now, is I'll say the quiet part out loud. Part of me is feeling like I need to make as much money as I can right now.

Brian:

Before it all comes crumbling.

Justin:

Because it feels like this is the this is the window. And my worry Yeah. Is that in twelve months, twenty four months, thirty six months, all of a sudden, the only people making money will be DigitalOcean, AWS, and OpenAI, Anthropic, etcetera. There will be no other businesses except for, you know, these basically hosting and infrastructure companies and LLM companies. And I I that just saying that is just like I part of me is joking, but there's this other part of me that's like, feeling this pressure.

Justin:

Like, I need to make as much money as I can right now because I might need to, like, provide for my kids and their future for a long time.

Brian:

You know, I was hanging out with a with a SaaS founder at Big Snow last week, and he runs a very successful SaaS. It's been it's been well established for a very long time, has a great customer base, and it is still growing Yeah. In 2026. And he was saying how and and they sell to, like, small business owners, a lot of, like, nontechnical customers. And he was saying that he's he's he was sort of expressing the same type of thing that you were just saying.

Brian:

Like like, kind of kind of Like, I just don't want anything to change because it's been so good for our business for the last ten years. Like, why does this AI thing have to shake everything up? I'm kind of scared for the future. A lot of uncertainty. And I was saying to him, and I believe this, and I'm like, you know, of all the people that I know, I feel like he's one of the most safe.

Justin:

And

Brian:

I think that about you too. Podcasting space has its own unique competitive challenges, I would say. But the fact that you have an established SaaS with with a well established customer base Mhmm. Means means that, like

Justin:

And a footprint that currently LLMs Yeah. Are highlighting.

Brian:

Yes. Yeah. And and exactly. Same thing with him too. And it's like

Justin:

I just wonder I I think the anxiety

Brian:

It's it's much more to me to me, more of the anxiety is like, this is why I'm not starting a SaaS business in 2026. Like, if you were trying to start one Yeah. Uh-uh. Like, it's like, I know that there's a lot of people who are and and you can be successful. There's lot of different ways to be successful, but I'm not I'm not playing that game anymore.

Brian:

It's just too hard to do that, in my opinion, for And to me, it's like I could see why it would be so difficult because of AI to try to break through as a new player. But if you are established you have thousands of customers, like, your customers are not are not all gonna vibe code your app tomorrow. There's no that's that's just not gonna happen.

Justin:

The stress is is like, first, they came for the sites that ran on documentation like Tailwind. Then they came for the established course brands like Lyricast. You know, like, now we're seeing it was like before it's like, no, this is you're gonna be fine. Like, Jeffrey Way. Like, you're you'll be fine.

Justin:

You know, Lyricast is an established brand. It's loved. Everybody who starts in Laravel starts with Lyricast. And he's just posted this video on his YouTube. I'm done.

Justin:

It's like, he it it's he's he had to fire 40% of his staff. He's he's like, this is having a real repercussion.

Brian:

I gotta check out that that video. I haven't seen that. I I did hear about the layoffs, though. But the you know, I also think that there's a spectrum of product type, and it goes back to this is an age old, like, product marketing thing, which is like the more essential your tool is to to the the way that revenue comes in or the way that you like Like podcasting, any sort of hosting based business is, to me, has always been fantastic because it's like your podcast is hosted on it. The pain of canceling is very disruptive.

Brian:

Even if you stop podcasting, you still want your podcast to be online. Yep. And you can say the same thing about any other tool that is like really part of the way that you sell your product or the way that you deliver your services to customers. The ones that are much more that I would say are much more at risk now are the nice to haves.

Justin:

Mhmm.

Brian:

The the to do list apps, the note takers, the the little project management stuff, the, you know, the the little super niche tool that does this one utility task over there. Like, yeah, I'm still subscribed to a lot of them. But, like, at some point, there could be a little paper cut that I'm like, I'll just I'll just vibe code a better version. You know?

Justin:

I I do think there's something about this pressure that a lot of people are feeling, which is, you know, for Adam and Tailwind, part of him was like, listen, he was making a lot of money for a long time. And that window was open for a while. And then he saw that window start to close. And I think a lot of people can look at that and go, fuck. Are we in another like, is this the window?

Justin:

Is it open now? And and then

Brian:

But also, like, what Adam is doing with UI is it UI dot SH?

Justin:

Yeah. Yeah. I mean, he's

Brian:

What a domain. I mean, but this is the kind of thing, I'm not saying everyone needs to go out there and start a new brand or start a new product, but he was obviously very public about the business challenges he had. He's taking them head on. Him and Steve are learning how to design with AI and launching a whole new product and brand line out of it. Yeah.

Brian:

And, like, like, just sort of like, I hate this term, but, like, lean in and start to be creative. And, like, yeah. Especially for him where his product is code related clearly it has had an impact on his business. It's like, okay, time to get creative and actually figure something out.

Justin:

Yeah. Time to pivot pivot into something related. I think I think in the short term, that will work. I just I want to recognize for a lot of people out there, I think there is a lot of people that are feeling like, I gotta work super hard right now. I gotta code my brains out right now.

Justin:

I gotta figure this out right now because eventually, the only thing left to do is gonna be plumbers, You know? And that's part of what Caleb was talking about. It's like, man, it's like, eventually, the only thing left is gonna be plumbing.

Brian:

Start start an HVAC

Justin:

company. Yeah. Alright. We should probably leave it there. Thanks.

Brian:

Yeah.

Justin:

I do. Everyone for joining us here in the chat. We had all sorts of great people. Keith and Dave Jones and Ryan Hefner. Great Beau is here as well.

Justin:

Great to see you all.

Brian:

Good times. Good to catch up.

Justin:

And I think I'm here next week. Are you here next week?

Brian:

Yeah. I'm here all of February. And then I I'm supposed to have a trip out to Park City in March. Okay. But there is no snow in Park

Justin:

There no snow.

Brian:

We're trying to figure out if we should go.

Justin:

So we'll be back February 12. Good to see you again. Yeah. See you next week. Later, folks.

Justin:

Bye.