Respite from the Dopamine Machine
#40

Respite from the Dopamine Machine

Justin Jackson:

Welcome to the panel where founders talk about building a better business and a better life. I'm Justin Jackson, cofounder of transistor.fm.

Brian Casel:

And I'm Brian Casel, creator of Builder Methods.

Justin Jackson:

And on the call today, very special guest, longtime listener often in the live chat. I've appeared on his podcast. He's building stuff in Atlanta, Georgia. Ryan Hefner. Welcome to the show, man.

Ryan Hefner:

Yo. What's up?

Brian Casel:

Yeah.

Ryan Hefner:

Happy to be here. Obviously, a longtime listener. And during the panel livestreams, there is a certain, I don't know. Like, want to just kinda dive in and get into that conversation because all the stuff that you guys are diving into is like the things that are racing through my head constantly. So it's fun to be on here and just kinda riff with y'all.

Brian Casel:

Yeah. It's great to great to have you on, Ryan. Great to catch up with you. I it's awesome that you're always in the live chat. And, you know, for those who don't know, I I I don't know if you're or if you're still doing it, but you've had one of the one of the long running, like, build in public podcasts going on and just kinda talking about your real projects over the years.

Brian Casel:

It's been pretty cool to follow.

Ryan Hefner:

I I kinda fell off on it a little bit. I'd say around last August for whatever reason, I think with the client work that I was doing and then just kind of like I mean, truthfully, of almost kinda like burning out a little bit to a certain degree, you know, like trying to get side stuff going and then balancing the client stuff and, like, dealing with that kind of stuff. It all kinda got got a bit much, so it was, hard to find, like, good topics to to talk about. But actually, right now, you know, I think with AI and I'm I'm also part of the whole, you know, winter break AI converted. Basically, I feel like

Brian Casel:

Oh, so you you were a bit slower than than most last year to come around on

Ryan Hefner:

it? You know, I'd say I wouldn't have fully adopted it. I had Copilot in my IDEs, like, whether that was, you know, Versus Code or Cursor. I had dabbled with mostly just kind of like using it as conversation, but not having it do full coding projects at the time. Maybe throwing it like a a bug here and there.

Ryan Hefner:

But over the Christmas break, I base I have a backlog of projects that I wanna do and domains and all this other stuff. And so I just started spinning up GitHub repos. And while I was playing with my kids, I would just start throwing ideas at it. And then I would just let it rip. And the this was the first time where I felt like I was actually getting meaningful output from the models that was legitimately, like, reducing a lot of, like, the boilerplate cruft stuff that you have to do when you're spinning up new projects.

Ryan Hefner:

And also fleshing out, you know, whether it's article templates or landing pages or whatever, but just giving you at least something to kinda like hop off on as opposed to just try and come up with, like, the blank canvas issue. Totally. And and and then I was just hooked. And so now, yeah, like, clawed max to the turned up and I get

Justin Jackson:

Oh, yeah.

Ryan Hefner:

90% news by like Tuesday and my my my subscription re like, re ups on Thursday or whatever. So then I gotta throttle it to get me through the rest of the week. But

Brian Casel:

Yeah. I've been hitting that max

Ryan Hefner:

actually been introducing This week. But Codex too. I think Codex is an amazing actually, I I started using them when I was actually running out of credits on Claude. And actually, I think for certain languages and certain apps, it actually performs way better. Knowing it on honestly, yeah, that's me.

Ryan Hefner:

Sorry. I didn't want to

Brian Casel:

Nice.

Ryan Hefner:

Get too too far into it already.

Justin Jackson:

Yeah.

Brian Casel:

Perfect. So so, you know, today is supposedly my my week to host. So I guess I should take the hosting duties today. So, yeah, like, today, I I did prepare a little topic that I think is a little bit of a change of pace. We'll see.

Brian Casel:

And the idea is how we think about our daily drivers across all things, tools, services, just like general things that we choose to use as our daily driver. And I think this we have a long list here of things that that happens in our work lives and are in our in our, like, personal lives, tools, services, and and things. And like, I think what's interesting about this question is, like, why have we chose to to stuck why why have things really stuck for us long term, even if there are, like, popular alternatives that didn't stick? It's just kind of fun to to talk about those preferences. But what's more interesting is like this idea of product market fit.

Brian Casel:

So let's we'll we'll get into that stuff. But real quick, why don't we just get like a quick update? I mean, Justin, what's been happening with you?

Justin Jackson:

Dude, so much is happening. We're just I'm I've actually been having these weeks where I'm feel like I'm addicted to drugs. And AI has definitely contributed to this, but it's like Monday, Tuesday, Wednesday, I felt like I was addicted to drugs. And then Thursday, I had kind of like the come off. Like, I was just like, nothing much was happening.

Justin Jackson:

And I was kind of like like like, come on. I need more of that same energy. So we TBPN, which is one of the big tech podcasts.

Brian Casel:

Oh, yeah. Yep.

Justin Jackson:

And they were just in the news. They were just acquired by OpenAI. So here's a podcast that doesn't have a ton of viewers or and listeners. They say 70,000 per episode or something.

Brian Casel:

I don't listen to them a lot, but I I I've known that they're pretty big. Right?

Justin Jackson:

It's big, but not really big. But they were acquired for hundreds of millions of dollars by OpenAI. So let's say 200,000,000. They were on track to make $30,000,000 in revenue this year. So we did a project for them because they interviewed Eddie Q for the fiftieth anniversary, and they wanted it on Apple HLS video.

Justin Jackson:

We still have to do that manually. So we had to coordinate all this. And so we had this big day, Wednesday. That was April Fools. Right?

Justin Jackson:

Yeah. April Fools Day, we did this big Eddie Q video launch, which you can watch on Apple Podcasts in video. Go check it out, TBPN. And then the very next day, they announced this acquisition by OpenAI. And was just like, what is going on?

Justin Jackson:

It's probably like Joe Rogan, Spotify acquired it for like 200,000,000 or 250,000,000. But this could be bigger. And Joe Rogan is a much bigger show. So it just goes to show like this show has the right kind of reach.

Brian Casel:

It's it's such an interesting acquisition. Like why? Like, I understand why Spotify would acquire a Joe Rogan or Ringer or like whatever else, because they're trying to move into podcasting. Like, is is OpenAI trying to move into podcasting? I don't I never really know what like, that's the thing with OpenAI.

Brian Casel:

I never really understand their strategy on anything.

Justin Jackson:

Oh, yeah. I mean, who knows if it'll work out? But this is why I'm bullish on podcasting is if you look at all of the CEOs and executives, they're all going on podcasts. Podcasts have become this channel. And I think it's because it's one of the few remaining genuine authentic human channels that real humans are responding

Brian Casel:

A 100%. I get all that. And I and I also see why even other companies acquire media brands. Because it's like, okay, if if we can't be the YouTube channel ourselves, let's just acquire one to to get that like marketing juice, right, or the or the exposure juice. But that that's for most companies and smaller, medium sized companies.

Brian Casel:

OpenAI is they're they're chat GPT. It's like they don't need exposure. You know? Everyone knows that.

Ryan Hefner:

The commercial default. Right? I mean, if anything, it's probably to control the narrative around AI to a certain degree. You know? If you have the show that everyone's listening to and you can kind of position the narrative around that to align with your whatever, you know, whatever other, like, initiatives you have behind the scenes and stuff, I mean, that's where it's gonna be.

Brian Casel:

Mhmm. That that makes the value. Of course, like well, they're supposed to be, like, independent and this isn't gonna change anything, but, of course now it it would happen. You you would think it would have some sort of point of view. So anyway, that's that's a whole other side interesting conversation we can have.

Brian Casel:

Yeah. But that that is super exciting. And and also that TPBN has been hosted on Transistor, and you guys have been part of that.

Justin Jackson:

Yeah. Yeah. It was an exciting week. And, of course, we're building stuff this whole time. And, yeah, it's just a lot feels like lots is happening.

Brian Casel:

I I did wanna kinda talk about, like, my my only update for this week before we get into the topic is, you know, I've been thinking really hard lately about my content strategy going forward. Obviously, Builder Methods essentially is a content company. And it's it's just a constant it's like a constant, like, evolution for me as a founder to really embrace this thing. Like, what I'm building here is a content media company, essentially. I'm building a lot of stuff and I'm teaching and I'm growing a community that's sort of the product behind it.

Brian Casel:

But on the front end, it's a YouTube channel. And so I did make the decision that I'm gonna really try to invest in this and try to like double down on the content production. I have been working with a part time editor to help me churn stuff out, but I am gonna be looking for so I I kind of posted a a job opening. This is like the first, like, real, like, official job opening to work with me in builder methods for, I think someone to cover some video editing stuff, but really I'm looking for somebody who's a little bit more like a producer, a creative collaborator. So really strong video chops, motion graphics, like, you know, videography.

Brian Casel:

Like, you should be really strong with with video editing. Like, that'll be part of it. But more is, like, I have some very specific, like, new, slightly different direction ideas for content on the YouTube channel going forward. And I think it's gonna require a heavier lift when it comes to editing and production and and storytelling and and taking a lot of, like, raw footage and crafting a a narrative and, like, documentary type type stuff. I just really wanted to kinda take it all to the next level.

Brian Casel:

I wanna try to differentiate even more from the rest of the pack in in the YouTubes in this build with AI space. And I I have some ideas around that. There's some I have some kind of bigger directional stuff for builder methods in general that I haven't talked at all publicly about that I'm really, really excited about, but it's gonna take a long time to build into that. And I think this is the first, like, kind of big step. So anyway, I'm I'm looking for somebody who if you happen to know anyone who is interested, there's there's a a new jobs page on on Builder Methods.

Justin Jackson:

I just saw that.

Brian Casel:

Yeah. I built that into the app, and video editor, producer, creator, collaborator type type person is what I'm looking for.

Ryan Hefner:

So

Justin Jackson:

Yeah. I think this is key, man. I think this is like this is really oh, man. I was having a big conversation with my teenage son who's all stressed out about AI and it it affect how is this gonna affect his jobs and everything? And I'm like, but dude, every day I go to work and I need help from humans to do stuff.

Justin Jackson:

And you just it's hard to communicate to folks that there's going to be opportunities like this. And it's also hard to communicate like, what that would look like. Like, I don't want somebody that's just gonna just take video from me and edit it in your case. Like, I want someone who's gonna show up and be like, help me craft this. Help me make it into something incredible.

Ryan Hefner:

Yeah. Bring the ideas. Bring the creativity. Like, help me like, challenge me and also elevate the entire thing. You know?

Ryan Hefner:

It's that's I mean, AI helps automate and, you know, like, rapidly get to a conclusion, but it doesn't actually write the narrative. You know? You still have to basically kinda guide it and direct it and figure it out.

Brian Casel:

100%. I wrote this in my newsletter last week that, like, how I think about scaling businesses now in 2026 is a little bit different from how I've done it in the past. In the past, it was always just like build a thing, build a process, hire someone to carry out that process. But now it's it's more like, okay, if if something is highly process oriented, I'm I'm definitely gonna try to automate it with an AI agent first. Because most most things that are, like, purely process oriented can be agent automated.

Brian Casel:

It that that in itself takes a lot of work, but but it can actually be done. And I am doing a lot of content development stuff and and process oriented stuff with like Claude agents and stuff. But where I do think when it comes to scaling and hiring, I'm gonna look to hire a person, a human, when it's like an owner type situation. Like, where where a role where it's a role where I just really wanna want an owner to be in it. That, you know, that most of the time is not gonna actually mean, like, equity ownership in the business or anything, but I mean, like, own your your role.

Brian Casel:

Like, Justin, I've heard you talk about this a little bit, like, where the value of someone like this in in the business is more like them just bringing unexpected ideas to the table Mhmm. And using their own creativity and their own taste to take things in a direction that might might surprise you, might might not have been part of your vision for a process, but, like, helps helps further the ultimate goal of, like, let's let's just make this whole thing better. You know?

Justin Jackson:

Yeah. Yeah. Exactly.

Brian Casel:

Ryan, what what are you working on this week?

Ryan Hefner:

You know, part of AI wave of of winter, the last three months, I've been so focused on actually just building, like, way too many projects and projects and products and everything. And so, actually, next week, the kids are on spring break, and they're going off to Savannah with their grandmother. And I'm actually using that week to kinda just consolidate and actually start figuring out, like, the launch timeline. And so, Justin, actually, I've been meaning to reach out to you, especially regarding, like, marketing for developers. One of the sites slash tools that I'm working on is called Launch Weeks.

Justin Jackson:

Okay. Cool.

Ryan Hefner:

And it's basically like a launch planning tool. And so I'm I'm using it as I'm dog fooding all my stuff through it. I'm basically planning out launch sequences, you know, warming social warming kind of scheduling and stuff like that and tracking

Justin Jackson:

Oh, sweet.

Ryan Hefner:

That kind of stuff.

Justin Jackson:

Do you have a domain for that yet?

Ryan Hefner:

Oh, it's I'm gonna be putting it up later or, like, next week, basically. But it's gonna be at launchweeks.com.

Justin Jackson:

Launchweeks.com.

Brian Casel:

Woah. That's a that's a great domain.

Justin Jackson:

Yeah. That's really great. Seriously, the the thing I've been talking about the most in terms of marketing for developers is people underrate the time that happens before you launch.

Ryan Hefner:

Yeah.

Justin Jackson:

It is the most potent time for marketing when you're building up anticipation for something. This is why

Brian Casel:

And it's the hardest time to

Justin Jackson:

move to market too.

Ryan Hefner:

Some well, some

Justin Jackson:

people say it's hard, but it's it's drives my team a little bit crazy. But like right now, we don't have video podcasting. Right? It's not built. We're building it.

Justin Jackson:

But I am actively going on podcasts, talking about how it's coming, how we're gonna have it. I'm talking about it in a lot of cases like we have it already.

Brian Casel:

Yes.

Justin Jackson:

We're gonna update the homepage to say, video podcasting coming soon. People click on it. They can sign up for the waiting list. The whole idea is to build anticipation. It's just like anything.

Justin Jackson:

Like, when the your favorite movie is coming out, and you're just waiting and you're seeing the trailers and the teasers and you're just waiting for it to come. That's the most potent time. Or you know a band is coming to town and or they're coming out with a new album and you're you're like reading mean, you don't do this anymore, but you read Rolling Stone to see what they're talking about it. Like, you just and Yeah.

Ryan Hefner:

You see their posters start showing up in town or whatever.

Justin Jackson:

And That's right.

Ryan Hefner:

That's Yeah. Just builds the excitement. Yeah.

Justin Jackson:

So I think that'll be a big that's a big idea launch to to have help people with a launch plan is yeah. I think there could be something there for sure.

Brian Casel:

Yeah. Like, the way that I've always thought about launching, like like, the prelaunch stuff is to treat it like a big mostly like the marketing site. That's a whole project in itself. So I just try to use that as my pre launch marketing, meaning like I don't do pre I don't do like a coming soon page. Right?

Brian Casel:

I don't like that maybe I did that a long time ago, but now it's like if I have a new product that I'm working on building, even just from from the start of it, when it's just an idea, one of the first things I end up doing is like designing a whole marketing site for it as if it's a real thing in the market for a few reasons. Number one is I don't want people to evaluate this as something that's coming soon. I want them to evaluate it as something that exists today and give me a real feedback. Like, would you are you ready to buy this today? Even though it doesn't exist, I wanna see if you might try to attempt to buy it today.

Brian Casel:

That but that that's number one. But the other thing is that is a whole project. So I don't like everyone like builds and launches and then they're like, alright, right after launch, we're gonna get on that marketing site and and we're gonna work. But that is like the last thing that you wanna do the day after you launch a product. Like, I'm I don't wanna embark on a whole website design project at that point.

Brian Casel:

Like, I have a thousand other things that I need to be doing with the very first customers. So so the idea of having the marketing site already done on day one is Mhmm. Huge. You know?

Justin Jackson:

Yep. Yeah. That's cool, Ryan.

Ryan Hefner:

And you've been good about using the marketing side as also basically like like recon where you'll have stuff behind an email, and then you'll present a form to get more information

Brian Casel:

Always.

Ryan Hefner:

Before there actually is, like, a full sign up. Yeah. And so, yeah, I

Brian Casel:

think email first, then goes to a page that which goes to a survey about, like, why are you interested in this? Mhmm. And and then that I use to probably invite them to a call or learn something or analyze those as I as I go through the building. You know?

Ryan Hefner:

Yeah. And I think now, more than ever, and to your point, Justin, like, for marketing, for developers, the launch is what's gonna really distinguish you from, like, just anyone throwing up a random app and trying to get people onto it versus building the anticipation, kinda building that momentum around it. And to your point, the launch week, I think it actually is gonna be like these launch full weeks where you're kinda doing it to where it's like, you know, Tuesday is the Product Hunt launch. Maybe Wednesday or Thursday is the hacker news thing. You also then have some blog posts, maybe a live stream that you're introducing to this, like, during that week to kinda have keep that momentum going.

Ryan Hefner:

And there's a two or three week lead up period to that where you're warming everything, and then there's also a one or two week trail off after that. So it's not really like a launch week. It's like a launch month.

Justin Jackson:

Yeah. That's that is so key. And the other thing is you're building trust. So in the age of AI where there are autonomous apps being built by Polsia, what you want to do to stand out is say, you can trust us. And how how how do we know to trust you?

Justin Jackson:

Well, in the in the pre launch, you're building up trust. You're you're saying, here's my reputation. Here's we're real human beings behind this. I I just reviewed Freak's new app. Freak is a big big in the Laravel community.

Justin Jackson:

Their their dot app. Sorry. Their dash their dot app. And they just did a few things with like putting the avatars of the human beings on the right side, and then down at the bottom having a nice human testimonial. And then just having here's a photo of the team sitting around a desk in a real office and saying, hey, we're the team behind these products.

Justin Jackson:

We've been doing this forever. I think building trust in your lead up is gonna be so key and people a lot of people aren't doing it. A lot of people are launching these anonymous apps Yeah. Like, you can't see the humans. I'm like, dude, you gotta show the team because I don't wanna give my credit card to some fly by night vibe coded app that a thirteen year old could be running on a you know, like, I so, yeah, I think there's lots in the launch there.

Brian Casel:

I I, you know, I I also, like, used to always, like, launch these marketing sites without my name or or face on it. And and part of it was like, wanna see if this product is gonna do well without without me talking about it. Not that I have such a huge audience or anything, but but I did learn that it is really helpful. So, like, I I have started putting like a little, like, letter from the founder type of thing. Like, there's one on Clarity Flow.

Brian Casel:

There's one on Build Your Methods. I do that on basically everything now because there are quite a few people who, like, especially on Clarity Flow, nobody knows me or follows this podcast or anything anymore. So, yeah, it's just really helpful to have a face on there.

Justin Jackson:

Yeah. Plus now, Brian, you're a celebrity, so you gotta put your

Ryan Hefner:

face on

Brian Casel:

about that.

Justin Jackson:

Yeah. Dude, are you're like one of the few people in my little town that I'm like, oh, yeah. I talk about AI with my buddy. He's, like, into it on YouTube. And they're like, wait, Brian Casel?

Justin Jackson:

I'm like, you know who Brian Casel is? They're

Brian Casel:

like, god. What? Funny. With YouTube, it's so weird. Like, I hope that people in our industry come across my videos, which is nice nice to hear about.

Brian Casel:

And I also hope that other people in my life, like, never see my videos. Like, I don't I don't want, like, friends and family in the real world who are not in tech. I just don't want them stumbling into my videos. It's just such a weird Like I was on a snowboarding trip with two longtime buddies two weeks ago, and like they're not in tech. And like they had no idea that I've been doing this YouTube.

Brian Casel:

I've been talking about it because just sharing like my business updates with them a little bit. But like the idea that I'm a quote unquote influencer on YouTube is like so foreign to most people, you know? It's foreign to me.

Ryan Hefner:

Like it doesn't

Brian Casel:

really feel like that. But anyway, All right. Let's talk about daily drivers. This is a term that's actually been on my mind lately. So specifically, I've been in the market for a new pair of shoes, like everyday walking around shoes.

Brian Casel:

They actually just just came. I haven't opened them yet. So I'm, you know, kind of looking around. And and I'm I'm actually not tied to any, like, particular brand of of sneakers. Like, I don't just buy the same ones every year for for decades.

Brian Casel:

Like, I I just sort of jump around for whatever reason. I'm also in the market right now for a new daily driver tool for building with AI, like a coding tool. Okay. I've been on Cursor for years now and like sort of Versus Code based for several years. Cursor three looks kind of interesting, but I'm so tracked into Claude code at this point that, you know, like, I I'm basically just using Claude code in the terminal in Cursor, and and it just feels like it's not like, that tool is not purpose built for that sort of thing.

Brian Casel:

So we'll get into it. I've been testing out like four or five different tools to see like which candidates could become my new daily driver. Because that is the app in my work life. Like that type of app is the thing that I literally spend the most hours in every day. So that's the daily driver.

Brian Casel:

So when I'm talking about daily drivers here, and I wanna kick it around to you guys, it's we're talking about not products or tools that you try once or twice and you kinda liked it. I'm talking about the things that have stuck around for you for years. Like, the things that have just embedded themselves in your everyday routine of life or work. There's obviously a bunch of categories where where, like, you always need something. You always you always need like an email client.

Brian Casel:

You always need a bank provider. You always need a this or that. Like, everyone has like settled on a thing. And, like, also, as we get into the more, like, personal life stuff, like, where where do we get our news? Like, what what like, how do you read news every day?

Brian Casel:

Or what type of vacations do you go on? And, what what brands or services do you use when you do vacations or whatever TV or like, so let's start to kick it around. Maybe we could jump back and forth between work and life. We I have this Google doc with a bunch of bullet points of, like, categories. I don't know.

Brian Casel:

Where do you guys wanna start on this? Like, what do think?

Justin Jackson:

Well, I talking about shoes, shoes are a big deal for me. I walk to my office every day. Okay. I wear out soles of my shoes regularly.

Brian Casel:

I also on your feet these days right now?

Justin Jackson:

Well, right now, I'm wearing Ultras. But I actually wear these probably, I don't know, 30 of the time. And I just bought a pair of Red Wing boots that I wear most of the time. So if your Redwing boots are like they've this is an American classic. It's been made in Detroit.

Justin Jackson:

I'm sure you've seen these before. But I've got something similar to to I've got something similar to the Iron Ranger or the classic Mach, but it's it was really fun. And these are, like, $280. Like, they're a bit pricier. Oh, this is the one I have.

Justin Jackson:

This one right here.

Brian Casel:

Oh, that's nice.

Justin Jackson:

The Heritage. So, yeah, $300. I decided to treat myself. I've always wanted a pair. And there is just something about buying into a big story, like buying into something that has a reputation.

Justin Jackson:

And I the first day I wore them out, people were like, oh, are red wings. Those are sweet. You know? So there's like this social proof.

Brian Casel:

Yeah.

Justin Jackson:

And and also something that maybe I could, like, resole and like, we actually do have what do you call somebody who works on shoes? We have one of those guys here in town. So

Ryan Hefner:

Yeah. I forgot. Cobbler. Cobbler. Cobbler.

Ryan Hefner:

Yeah. Cobbler? Yeah. Yeah.

Justin Jackson:

Yeah. So shoes are actually one of those things that are huge for me. Like, because I buy them all the time. I always buy almost always buy a new pair of shoes before like, a month before I go on a big trip or a big conference. Like, I'd like to have a new pair of shoes, but I like them to be worked in a little bit.

Brian Casel:

Yeah. I break

Ryan Hefner:

them in.

Brian Casel:

And that always makes me a little nervous if they're gonna come, like, the a couple days before a trip. I can't I can't commit to a pair of shoes and be uncomfortable on

Justin Jackson:

a trip. No. You can't do that. Yeah. You got you gotta work them in.

Brian Casel:

Yeah. What are you what are you kicking over there, Ryan?

Ryan Hefner:

Yeah. It kind of varies depending. I don't you know, when I'm around the house, I actually just like wear some Birkenstocks kinda, you know,

Brian Casel:

I'm all about

Ryan Hefner:

the Birkenstocks. Tupper ones and kinda shuffle around. Oh, the closed toe. Okay. Yeah.

Ryan Hefner:

I do closed toe. No one needs to see my feet.

Brian Casel:

Right. I'm, you know, I'm I'm like the the hippie listening to fish with the old with the old Birkenstock sandals all summer long. They're just I I don't even really like how they look. They're just so damn comfortable that

Justin Jackson:

Mhmm.

Brian Casel:

They're just awesome.

Ryan Hefner:

Yeah. Easy to slide into.

Justin Jackson:

What's the what's the point of being in your forties if you can't be comfortable?

Brian Casel:

Well, speaking of of, like, getting old, the the pair of shoes that I just ordered that are sitting on the table over there, I haven't opened them yet. New company for me, Xero Shoes. Do you guys know them?

Justin Jackson:

Oh, yeah.

Brian Casel:

So Xero, like, xeroshoes.com. And my friend Brian Marble recommended them to me. I hadn't heard of them before. Right there, you could you could there's a little marketing there, like social proof. Someone mentioned it.

Brian Casel:

Like, I I wouldn't have even known of this brand unless somebody I personally knew told me about them. You know, for me like, for the past year, my everyday has been like these Adidas slip in, like like slip on shoes. Like, this is like the old guy thing. Like, I just don't wanna, like, bend down all the time when I'm when I'm just, like, throwing them on and wanna go, like, walk or do do an errand or something. So these Xero shoes are, like, thinner, kinda cool, all slip on.

Brian Casel:

I don't know. We're gonna see how they work. We'll we'll see where we go. But

Justin Jackson:

I think shoes is a great example of often we buy so these Ultras that I bought, which were also a a zero drop shoe. My friend, Andy, bought them first. We are so influenced by our friends with products. So even in this age of, like, search engine optimization and LLM optimization and all this other stuff, definitely influencers who buy products and share them can influence us. But our friends are still, I think, the number one.

Justin Jackson:

Like, people in our circle that are like, oh, yeah. I bought these. You gotta get these. These are the best. It it almost makes it okay to get them.

Justin Jackson:

You know? It's like, is it okay for me to buy this stuff?

Brian Casel:

Alright. So I'll throw it to you guys first. So the question is, I mean, Justin, I know you you are building stuff, but maybe not quite as much as, like, John or the rest of your team is doing it every day. But, like, this is probably a multipart question. So there's, like, the the main, like, IDE, but then there's also, like, the the choice between Cloud Code or Codex or Gemini.

Brian Casel:

Which which one or ones of these, for for you personally, and maybe what you're also seeing on the transistor team, are, like, the go tos, like where you're spending most of the most of your hours doing that kind of stuff.

Justin Jackson:

I actually don't even know what everyone else uses. I use iTerm for my terminal and then Claude code in there. I've used iTerm forever, I think. I use Visual Studio Code when I do open things up. Every once in a while, I go into Cursor, but not that often.

Justin Jackson:

That's basically my setup.

Brian Casel:

So I have a question about about those. So would you say whenever you are gonna build something with Claude code, you're using iTerm and you're not using Versus Code with a terminal?

Justin Jackson:

That's right. ITerm. Yeah. Yeah. It's just a preference.

Justin Jackson:

It's like I can't having the terminal inside of Visual Studio Code, I just never got used to it. And so I just like having my iTerm setup and all on one monitor too.

Brian Casel:

And do you do you, like, rarely or never touch the actual code files, folders like you

Justin Jackson:

Yeah. So, I mean, these days, it's not as common. But every once in a while, I'll, like, still open up a page. Especially, like, if I'm working out like, even, like, if I'm doing some, like, I wanna do some custom HTML in a email newsletter and I just know what works. I just wanna build that myself.

Justin Jackson:

I'll I'll use Visual Studio Code for that. Cool.

Brian Casel:

Do do you know, like, what the rest of the team is using? Like or at least, like, is it is it Claude code or is it Codec?

Justin Jackson:

Definitely Claude code is the almost all of us are on Claude. And then I don't know what I think John uses iTerm as well. And I can't remember what they use for their IDE. I think Visual Studio Code, but I actually don't know. I should ask.

Brian Casel:

Ryan, what what are you doing over there?

Ryan Hefner:

Cursor for my code, but and then just using the cursor models, you know, depending on which what stuff I'm doing. I actually do the majority of my coding through the Claude mobile apps, actually. So I'll spin up sessions, like cloud sessions, and I do it all through either the phone or the iPad. I like that because it basically means that I can sit there and, like, be working on something no matter where I'm at. I'm not bound to my computer, and there's no, like, get pushes or anything that get lost on my local.

Brian Casel:

I run tasks with Cloud Code on mobile too, but only during my off hours. Like, if I'm if it's like at night or if I'm traveling or something. But, like, are you saying that you use it, like, as part of your workday?

Ryan Hefner:

For everything. Yeah. The entire day. Like, so I have my I like, an iPad Pro sitting next to me, and I just tether over to it. And then I'll fire off code, like, commands in there.

Brian Casel:

That's easier than just running Cloud Code on your computer that you're sitting at?

Justin Jackson:

I don't

Ryan Hefner:

know if it's easier, but I it's one consistent workflow regardless of where you're at.

Brian Casel:

Yeah. Like, you you can, like, get up and go with it.

Ryan Hefner:

I can get up and go anywhere. I can you know, I'm, like, driving the kids to school and I'm text prompting stuff into, like, create this plan and then I'm taking that plan and throwing it over into or wherever.

Brian Casel:

Not dangerous at all.

Ryan Hefner:

I can just be on the go.

Brian Casel:

One of the I mean, one of my favorite new features in Claude is this remote control feature, which I use a lot now. So you could start a a session locally on your desktop computer, enable remote control, and, like, literally pick up the same session on mobile. And then you're still communicating into your desktop computer remotely, which is pretty cool.

Ryan Hefner:

Yeah. So I wrote an app to let me do some of that stuff. So it's basically it's called like TermVisor, and it creates a local daemon that I can remote into and launch TMux windows in. So I'm not even bound to necessarily Cloud Code. I could be using Codex or Cloud or whatever.

Ryan Hefner:

But there's a lot of times where, you know, depending on the project that I'm working on, if I need for whatever reason in the cloud environment, sometimes the dependency loading or regenerating a migration or something like that, it just has issues with it. So to be able to then remote into a terminal and be able to run some commands regardless of whether it's actually within like a Claude code instance or not is super helpful. So That I've been building a lot of tools around Claude to help me work more on the go and be less tethered to a computer. And, like, for instance, last night before I went to bed, I basically started a I had to kick off, like, a new project with, like, a 20 spec 20 individual, like, spec, like, markdown files that it, like, generated that I pushed over into a repo and kind of, like, scaffolded it up. But then it just, like

Brian Casel:

Love it.

Ryan Hefner:

Chunked on it all night. Yeah. That's amazing.

Justin Jackson:

That's cool.

Brian Casel:

Yeah. Like, I I was But all through app nights where the mobile app is incredible. I I was I I I had trouble sleeping. I woke up super early, like an hour earlier than I usually do. I just sat in bed, picked up the phone, and fired off.

Brian Casel:

Like, I basically spec'd, built, and shipped for pretty significant new features in in an app that I'm working on all before I left the bed. Like like, no joke. It just it's just unbelievable right now.

Justin Jackson:

So this is a is this a sorry. I missed it. Is this a slash remote session that you guys are talking about or just using the Claude app?

Brian Casel:

I use it a couple different ways. Like, what I just described where where I was I initiated a new task from mobile. I'm using Cloud Code on the Cloud mobile app, which starts a new cloud session.

Justin Jackson:

Yes.

Brian Casel:

Yeah. So that's that's like typically if I want to, like, start up something new when I'm already on mobile. But then there's this, like, remote control option where it's like if I had started it at the desktop, but I wanna continue working on it, I can just kind of pick it up and and go. And then I actually just did a whole video, a whole YouTube video about, like, mobile options with with Cloud Code. There's few a few more, like, what you were just describing, Ryan, with, like, are tools I I guess you spun up a a custom one.

Brian Casel:

There there's tools. There's one called, like, Termius, which you can, like

Ryan Hefner:

Yeah. Termius is the the big one.

Brian Casel:

Remote into your into your

Ryan Hefner:

It's mostly

Brian Casel:

Yeah. But that, like, that's like if you anyway, like, we're getting really in the weeds here. But if you need to, like, get in and and and access something on your desktop when you're on mobile.

Justin Jackson:

Yeah. In the chat in the chat, Nathan Gathright is wondering if you guys have used t three code

Brian Casel:

yet. I don't I don't know what that one.

Justin Jackson:

Okay. T three dot codes is what he's talking about. This actually

Ryan Hefner:

Oh, that's that Theo thing. Right?

Brian Casel:

Yeah. I don't know what that is, but it actually looks similar to some of the tools that I've been trying for this, like, for my new for like a replacement for cursor for me. So I've been on cursor for years, and before that, Versus Code. And I still like to have even though I almost never write code myself, I still like to have the file tree and the files here with me in the same application where I'm working in Cloud Code. I I like the merging of having my terminal with Cloud Code and probably multiple agents in different tabs and having being able to, like, run servers and stuff and being able to see my files and and edit them all in one application.

Brian Casel:

I don't like to have to, like, jump between an IDE and a separate terminal. So that's why I've always loved, like, the Versus Code cursor thing. But, like, more and more, I'm I'm just and I and I for a while, I was using Cursors AI chat as well to to use other models when I wanna, like, switch off of Cloud Code for a little bit. But I've just been doing that less and less and relying more and more on Cloud Code for almost everything, which I don't know if that's good or bad. I might jump back and forth again in the future.

Brian Casel:

But the what I'm what I'm landing on now so so yesterday, I spent the day using Conductor. I don't know if you guys have heard about this. It's been around for, a year or two. Really, really nice app. It is especially known for being really good at running multiple agents each on a different work work tree.

Justin Jackson:

Conductor.build.

Ryan Hefner:

Yeah. Work trees. Right? Yeah.

Brian Casel:

Yeah. Conductor.build. So it's very Work Tree based, and you can, like, you can basically run multiple agents, you know, on multiple projects or within the same project using Work Trees. You can do that normally just using Git or whatever. But like, it just gives you a really nice interface for doing that.

Brian Casel:

And just overall, just a really tasteful UI for building agentically. Really nice workflow for like, you know, issuing, like like PRs and and bunch of other stuff kind of baked in there that I really like. There's a couple things that I don't love about it. Like, it's not it it it wraps Claude code in its own chat interface. It's not the actual it so they're using, the Claude code SDK.

Brian Casel:

So they're they're not it's it's not the actual Claude code CLI interface.

Ryan Hefner:

Which so the harnessing and everything that Claude code brings is lost, basically.

Brian Casel:

Yeah. Yeah. Under the hood, it's still Claude code, I guess, but it's so that's that's not great. It's nice what they've built, but it's missing Claude code features. Mhmm.

Brian Casel:

So the other one so that was what I tried yesterday. And then today, for for the day, I've been using this other new one called Superset. Okay. I think it's I think it's called Superset. Sh, which is a very, very similar concept.

Brian Casel:

If you look at it, it it I'm almost certain, like, they're basically modeling it after Conductor, but they sort of fix two missing features for me in Conductor. And one one is that, like, you could just use pure Claude code and pure terminals instead of like this this wrapped interface like I was just talking about. The other is in addition to WorkTrees, you can have just your main tree, your your main branch without it being a work tree. Whereas conductor, it's like you're in work trees all the time, which can be kind annoying work tree. Yeah.

Brian Casel:

Yeah. Interesting. So other than that, they're almost identical. A few differences here and there, but, like, I'm I'm actually really liking Superset today, and I this one might be the one that I stick with, but I don't know. I just like them both a lot.

Brian Casel:

So we'll see. Yeah. It is a different way

Ryan Hefner:

of working from solo?

Brian Casel:

Yes. Yeah. There's solo as well.

Ryan Hefner:

I haven't had a chance to

Brian Casel:

to try that. I'm waiting to see if he if he takes the 5,000,000 investment, and then I'll then maybe I'll commit to it. But, yeah, that that's been looking really, really solid too.

Ryan Hefner:

I've been definitely obviously clawed Opus four six extended, like, This basically have everything throttled up to the max, but I am surprised at how I would I would definitely push you into doing a little bit more Kodak stuff because I'm really surprised at how well it architects things. And it took iOS app that, you know, that I thought Claude was doing a relatively good job on. You know, it's like you're trying to use your Spidey sense to kinda work through these things and, like, is it really doing it? It's like it's kind of creating it. But it evaluated the code and completely rewrote it, and it's way more stable, way more performant, like memory usage is down.

Ryan Hefner:

It's kinda wild. Like, I'm trying to figure out how to blend those a little bit more, like, into a a more unified workflow where it's like I can ping pong between Claude, have Codex evaluate it, then, like, throw the Codex output to Claude. And there doesn't seem like there's that still seems to be, like, a lot of friction right now.

Brian Casel:

Yeah. I mean, I I do like that both Superset and Conductor you can use them with Claude code. You can use them with codecs. You can use them with with any other thing. So so that is nice.

Brian Casel:

I I do keep hearing really strong reviews of just Codex as a coding model more and more. I did try their their new Codex app a little bit, which is it's nice. It's a similar layout to this. It's similar layout to Google's anti gravity. But, you know, the thing with, like, anti gravity and codex is that you're basically I from what I understand, you're sort of, like, locked into their models.

Brian Casel:

I'm totally correct about that. I don't know. But I like having these, like, other independent apps that where we can jump between them.

Justin Jackson:

I mean, this is the other thing now is that the combining of models, you know, I've used it with Aaron Francis's counselors thing, but there's other things like it where you can get them to compete. You give them a task. Yeah. And then you get them to go out and fight it out. That's interesting.

Justin Jackson:

And then you get to see, you know, which models come up with something different than what, you know, the other I I just have it connected to Gemini, Codex and Claude. I I was just gonna say a lot of us have access to these models. Because, like, if you have a Google Workspace account, you have some level of access to Gemini already out of the box. And then I have a ChatGPT account for my family. So you have some level of access.

Justin Jackson:

So you can try these things out without actually fully committing to max plan on everything.

Brian Casel:

Yeah. I have the max plan on Claude and the $20 a month plan on Google and ChatGPT. So like so for the few tasks that like, I I do go to Gemini when I when I need an image generated. You know? If I if I ever need any sort of image, I'm gonna use Gemini for that.

Brian Casel:

I I do still use ChatGPT for life stuff, but but I have moved all of my work stuff, like, out of ChatGPT, and that's all in in the Claude ecosystem at this point. I, you know, I know that a lot of people do this type of stuff where it's like you're you're playing the the coding agents off of each other and you're getting their opinions and having one review the other or either that or you're jumping around a lot between just building a lot of stuff. And like sometimes you're using Claude models, sometimes you're using OpenAI models, and then you can really get a feel for like what's what they're good at, what they're what they're weaker or stronger at. I guess there's some value to that. But in in practice for me, this gets into one of my other questions, which is like which AI provider are we most loyal to, if you will.

Brian Casel:

But, like, in my building process, like, I've just become so comfortable with how Claude Opus collaborates with me and builds with me that it's like to to me, it's like I I maybe it's I just don't wanna spend extra cycles of having it build something and then giving it to another agent and having it review and tear it apart, and maybe I need to rebuild parts of it. I'd I'd rather have it build something, and I can review it and finish finish out the design, or then I can decide that, like, something was fundamentally misarchitected, and that's mostly on me. Like, I gave it the wrong direction. Yeah. Like, okay.

Brian Casel:

You might find a a a few optimizations in the code base that could be coded a certain different way. But like, to me, it's much more about design. And I mean, like, not even the look and feel, but the design of the product. How does it work from a user experience standpoint? Like, and all of that happens during the planning phase either.

Brian Casel:

So all of these models, all the frontier models are super good at getting you there. It's a it's more of a question of comfort, I I guess. You know? Mhmm.

Ryan Hefner:

I was just wondering, when you're working on a plan, are you are you creating one unified markdown file, or are you actually broken the breaking that out into, like, feature specific plans? Because I've actually started to actually fan it out a little bit more. So that way each of the individual features can be a little bit more robust and, like, detailed than what there is as opposed to, like, having it try to consolidate it because I feel like it it creates gaps when it tries to do that. And then then I try to fill it in, and then it tries to, like and sometimes it synthesizes it to a point to where you're trying to refine it, but then it loses a detail somewhere else. And actually so I'm trying to fan it out to, like, 20 or 30 spec files.

Brian Casel:

Yeah. So what I do is I usually do a big preproduct planning phase, and and that sort of generates like a big PRD, but that's not gonna be the plan or spec that it builds off of. That's just to guide the road map of the project. And then I'm gonna break that out into milestones. Usually not 30.

Brian Casel:

It depends on the tool, but or it depends on what I'm building. But like the one that I'm working on right now is this thing called SparkDrop, which ended up being broken out into, like, eight milestones. And for this one, it was the first time that I actually used a Ralf loop to to build most of these milestones. So I took so it there's, like, multiple phases of planning before it act ultimately builds. So it goes for for me, it goes like ideation and con and conception into a a big, like, product scope plan Kinda a thing.

Brian Casel:

Sort of thing product vision sort of thing. And then out of that, I'll turn that into milestones. And then each milestone is then broken out into, like, user stories that get fed into this is if I'm doing a Ralph loop, I I would, like, put all these user stories into a Ralph loop, and that can that can sometimes build, like, one milestone. Occasionally, I had it build, like, three or four milestones at a time, and it did pretty good with that. When that's all finished, I'm I'm still gonna hack on additional features or refinements or fixes, and that's mostly just like prompting.

Brian Casel:

Maybe occasionally, if it's more complex, I'll do if I'm doing like a big refactor or something, I'll I'll do go into plan mode for that and and create a and and then have it build off of that.

Ryan Hefner:

No. I'm I'm kinda curious, though. Are you doing that in the context of code, or are doing it in chats?

Brian Casel:

The pre the early planning stuff is mostly happening in chats, like in Claude, just claude.ai. But but then once then once I know what I'm building and I know what the vision is and I know what the roadmap is, that's when I move into code and that's when I turn it but I'm not actually coding it. I'm still gonna turn it into, like, specs. I think there's a difference between putting your product manager hat on or product designer hat on. That's like the vision stuff and the user experience.

Brian Casel:

Like what am I trying to build for users? That's first. And then it's like technical implementation. So how do I turn that product vision document into a technical implementation plan? That happens in Claude code.

Brian Casel:

That might have multiple levels it's broken into multiple milestones, and then the milestones go into Claude Codes' plan mode. And the only reason I do that is so that Claude can really like follow its own plan so that it that's how it really sticks to it and and completes it.

Ryan Hefner:

Yeah. So it doesn't create the gaps and stuff. Yeah.

Brian Casel:

But there's there's, like, a lot of different styles of building for different use cases. Like, I lately, I've I've been really into building these, like, internal tools. Right? And if it's an internal tool, I'm actually more likely to try to build it as using, a Ralph loop because it's usually smaller, more focused, and and I can get, like, maybe 80% of the product built that way, and then I go and then I work on the final 20%. Anyway, that that's a whole tangent we can

Ryan Hefner:

get in.

Justin Jackson:

I I do I do think it's interesting. This is in the age of AI, this is something to consider, which is everything is moving so fast. Everyone's coming up with new products, new models. You know? Now you know, now OpenAI is really going after anthropics market with developers and businesses.

Justin Jackson:

Humans just do get stuck in certain workflows and certain apps. Like, why like you said, like, why do I use iTerm instead of just using the terminal inside of Visual Studio Code? And I'm like, I don't know. I just prefer it that way, but it's silly. But that's the way it is.

Justin Jackson:

You know? Like, you we we develop these preferences. And, I mean, this is a big problem in podcasting. A lot of people who kind of grew up in the iPod era still use Apple Podcasts. But then millennials that came into podcasting through Spotify prefer Spotify.

Justin Jackson:

So we always end up preferring these things, and it's remarkably sticky.

Brian Casel:

Yeah. Oh, yeah. And I think a big part of that, matter what product we're talking about, is switching cost. Right? Like like, I feel it this week when when we're talking about these coding tools, like, is absolutely slowing me down to work in Conductor or Superset compared to me working in Cursor because I've been using Cursor for years.

Justin Jackson:

Yeah.

Brian Casel:

But but these these are totally new. So so I have to learn new muscle memory. I have to learn where things are. It's it's all different.

Ryan Hefner:

Mhmm.

Brian Casel:

So that means I'm I'm spending an extra two hours in my workday just reading their docs or or reprogramming their their keyboard shortcuts to my preferences. So so I am slower, but it's like I'm this week, I was like, I'm gonna I'm gonna commit to slowing down for a week and reevaluating this tool that I that I only reevaluate every, like, three years, really. And and, yeah, it's like, is the thing. It's it's like everything is like like I'm just like some a lot of times, it's like, I know that there's a better tool out there. I I know it.

Brian Casel:

I I know that that exists. Yeah. I just don't wanna spend the time switching to it.

Justin Jackson:

Well, this is why maybe Nathan was saying here in chat that Codex just released a plugin for Cloud Code, which is pretty smart because if, you know, if you're used to Cloud Code and OpenAI wants you to, like, check it out, well, let's just go where the market is. Let's go where the users are. You know?

Brian Casel:

That's really interesting.

Ryan Hefner:

So do you think you're gonna have a daily driver that comes out of this?

Brian Casel:

Oh, me?

Ryan Hefner:

Yeah. I mean, do you do you feel like a daily driver is evolving?

Brian Casel:

Yes. I like, I spent all day yesterday in Conductor and most of today in Superset. I'm feeling pretty good about Superset right now because it's sort of I'm I'm getting used to it, and I'm gonna give it a few more days. But I I could also see Conductor coming back and, like, evolving a little bit more, and they they just raised a bunch of money too. So that that'll be interesting.

Brian Casel:

So I don't know. But, yeah, I I do I would say that I am almost certain that it's gonna be that I that I am off of cursor now. Like, I I still I I I have still been firing it up just to, like, like, go grab some things that I know are just in there, But I think I'm I'm actually moving on from cursor at this point. Mhmm. Let's let's just jump around to a totally different category for a second.

Justin Jackson:

Okay.

Brian Casel:

Switch it up. Here's something alright. So here's one of those things. Let's talk about news and news sources and news reading apps. Okay.

Brian Casel:

Because this is one of those things that I feel like I need a new daily driver, and I've been lazy or haven't found a good option. I I mean, I get news from from lots of different sources and methods. But my in terms of, like, reading the world news headlines every day, I'm doing that in Apple News. Like, using the Apple News app, which kinda sucks, to be honest. Like, it's not great at that.

Brian Casel:

There's nothing smart about it. They're they're just I I do what I do like about it is that it is a very mixed and unbiased mix of headlines from Yeah. Major major pretty you know, mostly trustworthy news outlets. I also like that I have a a healthy mix of just, like, world news, like, top stories in the world, plus plus like top stories in sports. I I read a lot of sports updates, especially my teams.

Brian Casel:

Mhmm. I like that that's mixed in there. I don't know. Like, it's just a it's it's it's a quick scan. I mean, I'm actually usually reading it at night.

Brian Casel:

I I don't read news in the morning. I read it at night. I mean, like, like, there's like other podcasts and and some YouTube stuff where I get some news. But that's like, if I wanna know, like, what's happening in the world, I don't watch, like, TV news or anything like that. It's I'm I'm gonna read the headlines, you know, which tend to come from, like, what do they have?

Brian Casel:

Like, Wall Street Journal in there or some like CNBC, some like some a little bit of New York Times, a little bit of everything. You know? I like the mix.

Justin Jackson:

The nice thing about it too is that it it it it if you're on Plus, it gives you access to most of those publications without having to pay again, which I found helpful. Often, I'll I'll, like, find a news article I wanna read. I'm like, oh, you can either put the archive.ph in front of it, or you can go to Apple News and just search for the headline and find it. And then

Brian Casel:

But it's not everything. I I am on Plus, but it's not everything. Like, you it's not a full Wall Street Journal

Justin Jackson:

No.

Brian Casel:

It's not a description.

Ryan Hefner:

No. Yeah. There's still some key walls there. Yeah.

Justin Jackson:

I mean, I'm I'm still so I have a reading area in my iPhone, and it's just like I've got news there. That's probably the one I use the most. There's some that read my Blue Sky feed and bring up links, that's called Sill. I like I like that one. There's another one called Feedseer that I use, and then Reddit.

Justin Jackson:

Then And sometimes I go to the New York Times directly. There's a new one by Flipboard called surf.social, which is also looks interesting. I've been checking that that out. I loved back in the day wasn't Flipboard. It was oh, someone's gonna remember it.

Justin Jackson:

But there was this app that would just surface all the interesting links from your Twitter feed. And I loved that app. And I've been kind of finding these little replacements, and this is another one of them.

Brian Casel:

I was using maybe about a year or two ago, and it was around for about a year, and then they shut it down. The creator of Instagram

Justin Jackson:

Yeah. Yeah. What was that called?

Brian Casel:

Yeah. Like, now we even forgot the name of it. Like, he he shut it down. Like but it was great. It was it was really well done.

Brian Casel:

It was a news reader Artifact. Agri or it was called Artifact.

Ryan Hefner:

Yeah. Artifact.

Brian Casel:

Yeah. And it was really smart at at following, like, what are the articles that I actually click on and I read? And then it gives me more of that kind of stuff from from all different sources.

Justin Jackson:

Yeah.

Brian Casel:

It's such a bummer when they when they shut that down. And that and that's when I switched to Apple News, which does a really crappy job of of the same idea. You know?

Justin Jackson:

Yeah. What about you, Ryan? What do you use typically?

Ryan Hefner:

It's a blend between Apple News and then obviously podcast and newsletters. But the problem is those things are all over the place.

Justin Jackson:

Oh, yeah. Tangle. I'm a big I'm a big Tangle news guy. I I don't know if you guys are aware of Tangle newsletter, but that one's awesome for, like, unbiased kind of two sided news in The US. I I'm a big fan of them.

Brian Casel:

Nice. There's another news site that I tried, but it just didn't stick for me for a reason. Was it Semaphore?

Ryan Hefner:

Oh, yeah. Semaphore was cool.

Brian Casel:

I don't I don't know if this is it. There's one that's like they have, like, ratings of, like, how biased each article is. There's, like, data ratings underneath, like, each Yeah. Article. I forgot what that is.

Justin Jackson:

Yeah. They're they sponsor Channel five News. What's what is that Channel five News sponsor? Sorry. I interrupted you, Ryan.

Justin Jackson:

What are some other newsletters you're reading right now?

Ryan Hefner:

Oh, I mean, most of them are kinda like industry ones, but sorry. I got distracted by some before. I was seeing if they had the the rating breakdown. Yeah.

Brian Casel:

That's I don't think that's the one.

Justin Jackson:

I think it's ground.news. Oh, ground.news.

Brian Casel:

That's it. Yeah. Yeah. Like, I don't know. Like, I I I tried reading that for for a couple days and, like, I don't know.

Brian Casel:

It just didn't really stick for me, but I love the concept. Mhmm. Yeah.

Justin Jackson:

It's hard to build a habit. I mean, if we're all honest, I mean, maybe you guys I'm on Reddit the most by far. That's like interesting. I I

Brian Casel:

don't go in Reddit as much as I I would like, but I by far, I'm I'm reading about sports. Like, when I'm reading news, I I catch the main headlines, but I'm so less interested in that stuff than I used to be. And now so much more interested in sports updates, especially my teams and the leagues that I like to follow and tech news updates I like to follow. You know? I was listening to Jerry Seinfeld.

Brian Casel:

He called in to WFAN, which is the New York sports radio here, and he was talking about how sports is one of those it's like a fish tank where it's isolated from the rest of the world and politics and all this crazy craziness that's going on. Mhmm. But it's this whole, like, insulated world where it's like everyone is super, super invested and soup and and have all these, like, heated arguments debates and competitions, and they're and they're really, like, emotionally invested in all this stuff that doesn't matter at all in the grand scheme of things. You know? And and it's like and it has nothing to do with politics and red and blue and and any of that nonsense.

Brian Casel:

And it's so true, but it it has that to me has been such a a breath of fresh air and, like and just, like, following all the little storylines and, like, the like, right now, like, one of the things that I'm really fascinated with is in like, I followed the NBA a lot and this idea of tanking. I don't know if you guys are familiar with this story at all. No. Okay. So if you haven't been following NBA basketball, there's a there's a huge there's a huge problem in the NBA, and that is eight to 10 of the teams intentionally lose for almost half the season

Justin Jackson:

Oh, interesting.

Brian Casel:

To try to get draft picks. Oh. And it used to be like they it used to be that, like, they might start to tank in the last three games, five games of the season if they if it if they might make a movement. Now you're starting to see teams do it from the all star break on, like halfway through the season and onward. Teams are intentionally losing.

Brian Casel:

Like, they're like, they're benching their best players. They're just you know, because, like, because they're like, if if we're if we're the Utah Jazz, our only option for getting better we're we're not New York or LA, so we can't afford the big names. Our only option is to just try to tank the season and get a good draft. Early.

Justin Jackson:

Yeah. Yep.

Brian Casel:

And it's a massive problem. Like like so there so there's all these different proposals for, like, different rule changes and how to how to penalize tanking and all and all these different incentive structures. It's it's actually a really fascinating and difficult problem, but it's and and there's, like, business involved in it. It's it's so anyway, I I've been, like, super invested in that whole ongoing saga.

Justin Jackson:

I mean, people always make fun of folks that are into wrestling, but it's basically the same thing. It's just like a way to turn your brain off and be in a different are are you into any sports, Ryan?

Ryan Hefner:

Storyline. You know, I I have two young younger kids, five and eight, about to turn six and nine. So it's like my world is pretty much just them and and trying to get some work done during the day. So I I don't really follow too many sports. But I do like watching I got into NBA playoffs last year.

Ryan Hefner:

I I was, like

Brian Casel:

Mhmm.

Ryan Hefner:

Really rooting for the Knicks and hoping they were gonna push through. And, you know, occasionally, I'll watch whole game here and there. But no. Well, I mean, I was in New York for about, like, fifteen years. I mean, actually, if my if it if there was a team, it was probably it'd probably be, like, more like the Pistons or the Cavs because I'm from Ohio.

Ryan Hefner:

Mhmm. But That's

Justin Jackson:

kind of the other fun thing about sports is that it it is I I mean, talk about jobs to be done and product market fit. It's like, help me to turn my brain off of the real world so that I can just have a break.

Brian Casel:

Yes.

Justin Jackson:

And for me, like, watching I'm a big Edmonton Oilers fan with NHL hockey. But why? I don't live there anymore. Mhmm. You know, I haven't lived there in over, I don't know, fifteen, twenty years.

Justin Jackson:

But it's the hometown team I grew up with. And and even like the name, like, don't I'm not super excited about the oil industry. But that's like the whole identity of the team is Yes. The the the area. Like, they they they at the beginning of the game, a big oil tower comes down and then they skate through it.

Ryan Hefner:

That's great. Oh, wow.

Brian Casel:

I love stuff like that.

Justin Jackson:

But there's something about and just something about caring passionately about something that doesn't really matter is kind of this relief. And I think this is gonna be a bigger and bigger thing in the age of AI. I'm seeing way more people talk about burnout. Like, even what I was saying, like, you're on drugs and you're just like, I'm gonna burn out. I'm burnt out by 11AM.

Justin Jackson:

You know?

Ryan Hefner:

The dopamine machine.

Justin Jackson:

Yeah. People are looking for a some respite from the dopamine machine and respite from the dopamine machine. That's my new hardcore band.

Ryan Hefner:

Yeah. I love it.

Brian Casel:

Should we do like one more quick one or wrap it up? Like

Justin Jackson:

Yeah. Let's do one more quick one.

Brian Casel:

How do you guys think about like like, vacation and travel? You wanna do something like that?

Justin Jackson:

Oh, yeah. I would that's a good one.

Brian Casel:

Because this encapsulates encapsulates a lot of things. Like, there's, you know, there's like airlines. There's like, do you like Airbnb versus hotels or or VRBO? What types of vacations do you tend to prefer or go to? What are you most loyal to?

Brian Casel:

Like, again, we're thinking about, like, daily drivers, probably the wrong word, but, like, what are your go tos? If you're thinking, I wanna take a summer vacation with my family. What are we thinking about?

Justin Jackson:

You go ahead, Ryan. Do you got anything?

Ryan Hefner:

Yeah. I mean, I'm I guess I've kind of evolved into being like a Delta flyer over the years, and also it helps being in Atlanta, a pretty major Delta hub with pretty much direct flights to anywhere you wanna go. Mostly Delta. I'd say I kinda air towards VRBO, but at the same time bounce between VRBO and Airbnb just depending. I feel like some of the fees and stuff are maybe I don't know.

Ryan Hefner:

It seems so weird. Like, the fees are kinda like over the top on Airbnb right now. So actually, in a weird way, I've been reverting back to maybe hotels or like different types of things like that to have if they have accommodations that kinda like work well for our family and if we have guests or something that are joining us on the vacation. Yeah. That's that's kind of some of the stuff I've been using.

Ryan Hefner:

And some of the way you know, the things I've been like teeter tottering on as far as like which one to lean into more.

Justin Jackson:

I I'm seeing more and more people teeter totter. Like that that feeling of Airbnb often sucks. And when I'm traveling by myself or with my my wife, I want a hotel. Like, hotels are way better than Airbnbs in so many ways. And if I'm traveling with my family though, you typically need more space.

Justin Jackson:

You you need so it's like we gotta go with Airbnb or VRBO. But it feels like there's like still a there's something about what's going on on Airbnb that people are teetering on. You know?

Brian Casel:

We're I would say we're pretty we're still pretty loyal to it. Although, like, we go between Airbnb or or hotels and almost never use VRBO. Part I think for me, of it, like, if if we're so we we go for an Airbnb when we're gonna be there for multiple days, typically with kids. And sometimes we go on Airbnb like, there there have been some trips where, like, Airbnb drives our vacation decision. Like, where can we find the coolest Airbnb to stay at?

Brian Casel:

That's where we're gonna fly to.

Justin Jackson:

Mhmm. Yeah.

Brian Casel:

Like, if we just wanna be sort of, like, creative about, like, we don't really know where we're gonna go this this this next vacation. Let's just see if we can go, like, stay in, like, a teepee on a farm or a cave or some some some crazy place on on Airbnb.

Ryan Hefner:

Yeah. That's cool for the experiences, you know, if you're trying to just, like, hunt for an excursion.

Brian Casel:

The other thing that I found is that, like, I tend to trust the reviews on Airbnb. And I have like, we have stated at VRBOs where it was like, it's some company that markets their their property through VRBO, and then the company tries to screw me out of some payment. And, like because that's the thing on on VRBO. It's like they connect to you, but then you're on your own. Whereas, like, if if they if they screw with you, like and and they're through Airbnb, like, they'll get Airbnb will take care of it.

Brian Casel:

You know?

Justin Jackson:

Don't don't you guys think that, to me, travel I've always struggled with, like, booking vacations, taking vacations, planning vacations. And I know, like, when I have a good trip, I'm like, okay, here's what made the trip good. You know? And it depends on what stage of life your family's at. But like one summer, we made this last minute decision and went to this lo this family camp that was like not even an hour away.

Justin Jackson:

And it was just like perfect. They had this big beautiful pool. They had motorboats that would take you water skiing. They had mountain bikes you could rent. They had a basketball court.

Justin Jackson:

They had a big dining hall. It was just like everything about it. A zip line. And so there's all these things to do. The kids were at the perfect age.

Justin Jackson:

They were probably like, I don't know, 12 and under. And it was just like, perfect, you know? But I find it very difficult. And this is somewhere I was hoping AI would solve my problems. It's just like, I want you to in a even in a recurring way, like, don't want to have to drive this all the time.

Justin Jackson:

But in the same way that I set I set like flight reminders on Google Flights. I don't know if you guys do this. Just like, just give me price emails all the time so I get different destinations that sends me emails. I want something that's just like emailing me all the time going, hey, you haven't booked anything for spring break. Do do you wanna do that?

Justin Jackson:

You haven't booked any like, summer's coming up and It's coming up. Yeah. Like and also knows me. Like, it feels like booking a good trip is like one of the hardest things to do.

Brian Casel:

Yeah. I have been using ChatGPT more in the vacation planning stuff.

Justin Jackson:

Okay.

Brian Casel:

So it was like, we just booked a quick trip for spring break down to Florida. We got a place on the beach on the on the West Coast Of Florida. But for that trip, we were, like, up in the air. We're like, basically, both both Amy and I started on ChatGPT, and we started saying, like, we know that we wanna be on a beach. It's gotta be just for a couple of days, flyable from from Connecticut.

Brian Casel:

We want a direct flight somewhere. And we started looking at, like, potential, like like, maybe, like, Grand Cayman or, like, one of these Caribbean islands. We looked at, like, different places in Florida. We've been to Florida lots of times. Like, what's gonna be nice?

Brian Casel:

What's like, can we get something nice on the beach? Does it work with our budget? So, we ended up settling for this spot on the beach. But like, yeah, like we use I I like to use ChatGPT to first just start high level. Like, give me a bunch of destination ideas with with this criteria.

Brian Casel:

Did

Justin Jackson:

I why ChatGPT and not Claude or Gemini or anything else?

Brian Casel:

One of those, like, comfort or in in my mind, I now have it where it's like, ChatTPT is for life. Claude is for work. Yeah. I don't know why. Yeah.

Brian Casel:

That's just what I do. Like, I I use ChatTPT for life shit. And but so whether it's, like, vacations or, like, I got this weird shoulder pain, what do I do about it? Or or, like, I a replacement for Google random Google searches, or I I need a fun joke or or story to play with my kids. That's ChatGPT.

Brian Casel:

But similar similar searches, like, what's a what's a good tool for this thing in my work? I'm gonna go to Cloud for that. What's a what's a project that I'm working on in my business or business strategy? I'm using Cloud for that. You know?

Justin Jackson:

Are are you doing something similar, Ryan?

Ryan Hefner:

You know, I think this is actually a pretty underserved one thing that I think is actually kinda underserved is right now AI feels like a very solo experience. And I think eventually there's gonna be this collaborative AI thing so that but where you and your wife can be sitting there, GMing on stuff, and you can see the responses that she's getting, and you could be upvoting that or something, and then that's feeding into kind of creating this thing.

Brian Casel:

Totally. That and that's like sort of a new development. Like, just just now on this vacation planning stuff, and we're also planning our summer vacation now too. And she's much further behind me on like tech adoption kind of kind of stuff, obviously. Right?

Brian Casel:

But the but like like, I just share like, I made that ChatGPT conversation shareable, and I I shared it to her, and she started looking at it and and doing her own searches. But, yeah, like, definitely that that, like, multiplayer AI kind of experience is

Ryan Hefner:

Yeah. I think even amongst teams, I think that's gonna be huge. And then, Justin, to your point about how you like interfacing with ChatGPT via Slack to be able to, like, have, like, working rooms where you can kinda be jamming and either people have their own agents that they're working against and you're everyone's seeing the output and then, like, snagging stuff. I think I think there's some stuff there too that it's gonna be interesting to see what evolves.

Justin Jackson:

That's a really good point. I haven't thought about that, but, like, the, yeah, the co driving model is such a especially with something like travel planning. Like, it just makes so much sense to say, okay, we we're both in here. But, you know, like, you know, I might say, well, I wanna do kind of this kind of stuff. And my wife might be like, well, that's fine.

Justin Jackson:

But I I I wanna make sure that we have, you know, these kinds

Ryan Hefner:

of things. Like, that's job

Brian Casel:

to be done. Right? Like like, we always have these we really center, like, our trips around, like, we just want one. We gotta just we gotta make sure we have this one thing in the trip and everything else sort of clicks into place. Like, this next one is, like, we wanna be on the beach.

Brian Casel:

Right? Summer, we're we're gonna probably go to Glacier National Park and then drive up into Canada and go to go to

Justin Jackson:

Banff. Nice.

Brian Casel:

Like that so, like, we have we have these, like, models of vacations that we continuously come back to. So summer trips, we've we've done a lot of national parks, and and we just, like, center the destination choice around national parks. Usually, like, fly somewhere and then and then drive and do some Airbnbs around those areas, do a lot of hiking and and seeing stuff with the kids. Sometimes it's like a quick beach vacation. Sometimes it's like more of like a city or exploration, usually just me and me and Amy.

Justin Jackson:

This is this is the other thing I think is missing though from all of these because I've tried some of these AI planning trip planning tools. And I think everybody's still stuck too much in this mindset of like, here's a single chat, single prompt, like, you wanna go to Banff. I'll recommend you like, what days are you going? Okay. I'll recommend you an itinerary every day.

Justin Jackson:

It's like, no. No. No. What I want you to do is I want you to first develop some sort of codex knowledge about me. Like, I want you to understand what I like, what my wife likes, what my kids like, and at this stage of life.

Justin Jackson:

Right? So, like, now all of our kids are older. They don't want to travel with us. And so, like, we we've thought about doing trips where we will fly somewhere and then our kids will fly and meet us there. Okay.

Justin Jackson:

Well, that's interesting. Like, that's a new development.

Brian Casel:

What think that the the main apps do that now. Like like, ChatGPT knows about me, knows knows I have kids of this age and like

Justin Jackson:

Yeah. Yeah. That that part is interesting. I I think the other part that's still missing is this proactive like like, you can set recurring tasks and you can prompt it, but those still feel very much like manual things I have to implement. I want some in certain cases, not all the time, but in certain cases, I want it to be proactively reaching out to me.

Justin Jackson:

I want it to be prompting smart alerts. Yeah. Yeah. Seriously. You

Brian Casel:

haven't done this sort of thing in a while. Isn't it time to

Justin Jackson:

Yeah.

Brian Casel:

Plan the next thing?

Justin Jackson:

Yeah. Exactly. And even just like, hey. Like, I don't have your kid's school calendar yet. Can you add that to, you know, my memory?

Justin Jackson:

I I still think there's too much, like, self driving with these LLM tools. And they need to be there needs to be a mode where I can turn it on and say, okay, in this particular area, I want you to prompt me way more. I want you to be like, hey. I don't have this information. Hey.

Justin Jackson:

I don't have this information. Hey. If you give me access to your calendar, I can help with a lot of your planning. Hey.

Brian Casel:

I mean, that's the vision. Right? Like and that's that's the thing with agents. And and I don't I don't know how comfortable I am or we are as a as a society with, like, just giving over everything, our personal accounts and and stuff like that to have it. Like but that's sort of what like, I I feel like the models and the technology are capable of that sort of thing today.

Brian Casel:

It's it's more a question of adoption and willingness to first of all, like, are we willing to just give over our personal email and and calendar to a ChatGPT or to an OpenClaw or to a Claude? Some people, yes. Some people, no. But then the the but then I would say, like, the the tool companies, the product those companies are yeah. They they've been trying to push these, like, agentic tools, but they haven't quite hit the mainstream because I think the mainstream hasn't quite had the demand for them in the big way.

Brian Casel:

You know? Like that's the thing for me with agents is that like ChatGPT has been trying to do this for years now. It's like, get me to have ChatGPT choose restaurants and choose hotels and choose flights for me and book them for me. But I've actually tried that a couple times just to see if it could do it. And first of all, it can't.

Brian Casel:

It just finds stuff that's not actually available or that I'm just not interested in. But second of all, like, for me, at at least at this point, I like the balance of, like, having a little bit of control, but also offloading more of the research legwork. Like like for these vacations, I like to have ChatGPTJ, like, alright, high level, Give me a lay of the land. Give me some good destinations and maybe a few sites that we that we might wanna see. But then I'm gonna wanna pick my own hotel and look at the actual photos and you know?

Justin Jackson:

Yeah. I mean, this is what feels like missing, though, is for every tool, what I really want is a proactive, smart notification or prompt. I don't I want to still book my flight. Like, I'm not gonna I don't need it to do that part. And like Ryan was saying, you know, we often develop a allegiance to a certain airline.

Justin Jackson:

We have points. We have things we want to do there. So and picking flights is like such Yeah. In the weeds Don't screw me. Like, I gotta get in there.

Justin Jackson:

It's like, I don't want your agent going, oh, you're you'll like this seat. It's like, no. No.

Brian Casel:

No. You can handle a 4AM flight. Sure. You

Ryan Hefner:

know? But Ten hour layover.

Justin Jackson:

Yeah. Ten hour fifteen hour layover. It's like, oh, thanks, Claude. And my wife's just looking at me like,

Ryan Hefner:

I told you we

Justin Jackson:

shouldn't use AI for this.

Brian Casel:

That's the vacation disaster right there. But smart prompts. Through an airport.

Justin Jackson:

Smart prompts is this is what again, we we this is a recurring theme. What what do I want a really good employee to do? I want the them to bring to my attention something that I need to be reminded of, something that I need to be aware of, something that I haven't thought of, whatever. That's a good employee. That's what I want these agents to be doing.

Justin Jackson:

I want them to be prompting me and going, listen, dude, I know you're busy. I know you get super in the weeds, and you're not thinking about the rest of life. Here's an idea. Maybe you should book a trip for this time of year. It's good for all these reasons.

Justin Jackson:

It's the shoulder season in Vernon, you know, there's not much happening in your hometown. This is the time to go, dude. Like, here's some opportunities and list it out

Brian Casel:

for like a you want like an AI life coach or life

Ryan Hefner:

yeah. Mhmm. Then no big deal. I

Justin Jackson:

it feels like that's especially for something like Claude or ChattyBT that already knows so much about me. It feels like that I do use them. I mean, this

Brian Casel:

is more on the is more on the work side, but I really use it a lot as a as a I don't know if you call it a coach or a therapist or just a thought partner or just a sounding board.

Justin Jackson:

Mhmm.

Brian Casel:

But at this point, it's almost every medium sized or big or big decision that I make, I think about it, and then I voice note it into Claude, and we have a back and forth about it to just to just clarify. And and it's like, I'm aware of the issues with it's just gonna agree and support most of what I say. And sometimes I'll actively just be like, just push back on this, like poke holes in my logic. These are my unknown these are the questions on my mind. But, like, it's still even knowing that, it's still more helpful in in clarifying than just thinking about it and not having that that, like, service available to me.

Brian Casel:

You know?

Ryan Hefner:

Well, sometimes even the process of just saying it out loud helps solidify it in your head. It's kinda like when you go to someone with a question, and then all of a sudden you answer to the question, like, as you're ask yeah. Like, asking the question. Just like to get it out and then have that kind of sounding board, the whole rubber ducky thing. Actually, on that topic, do you use any sort of, like, voice transcription stuff?

Ryan Hefner:

Do you use, like, a WhisperFlow or anything like that when you're when you're interfacing? Yeah.

Brian Casel:

In the computer, use WhisperFlow for almost everything.

Justin Jackson:

I've just started using WhisperFlow. It's taken a little while to get it get the habit to remind myself to use it, but I have been using it more often.

Brian Casel:

Yeah. I use it a lot. I I love it.

Ryan Hefner:

Yeah. Yeah. I was just curious if that was like a daily I

Brian Casel:

just use the iPhone voice memos app a lot, and and it has built in transcription. So what I end up doing a lot and and and Claude's thing actually kind of sucks for this, at least right now on on the if you try to voice note into Claude, you can do it, but, like, not for more than, like, a minute or two. So a lot of times I'll go I do a daily walk almost every day, and I'll speak for ten, twenty minutes on a topic. And let's say it's like one of these like sort of business strategy, like journal stuff. I can I can do that whole walk, and twenty seconds later, it's all transcribed?

Brian Casel:

I can copy it, copy the transcription, paste it into Claude, and then we go from there.

Justin Jackson:

Yeah. I I don't trust the like, if I'm on a walk and I need to record a lot, I'm just using voice memos, and then I'll upload that for transcription. Because a lot of these tools, they just I was using AI Pen for a while. Is that right? That was like an indie app that had been around for a while for this kind of stuff.

Justin Jackson:

And it's I'll also use it every once in a while, but the every once in a while, it'll, like, mess up, you know, and not even WhisperFlow sometimes garble up. If it's too long.

Brian Casel:

Yeah. And I I haven't taken the time to, like, really program WhisperFlow. I know you can, like, put your whole, like, vocabulary and all these, like, unusual brand names or or phrases that you might say. Like, there's still a lot of stuff that I have to correct after you know, because it just it adds a space where there shouldn't be a space or whatever it might be. But but yeah.

Brian Casel:

Especially for prompting AI. Like in Cloud Code, I use WhisperFlow all the time. Yeah.

Justin Jackson:

The the only other thing I as we're winding down here, the only one thing I wanted to just say quick was that idea that Ryan had about about this dual mode. Because you were Brian, you were just saying it tends to agree with you. Right? ChatGPT and Claude tends to agree with you. This is why Ryan's idea of having like this dual mode like, if my wife brought her ChatGPT and I bring my ChatGPT, and now they have to like It's

Brian Casel:

marriage counselor.

Justin Jackson:

That's that would be very interesting. And even like

Ryan Hefner:

And you're seeing it side by side work.

Justin Jackson:

Yeah. But but then they have to negotiate between themselves. In some ways, like Aaron's counselors thing does. Right? So in that case, it's all still my models.

Justin Jackson:

But I think for the AI companies, this I don't know what you would call this co pilot mode. If you have a if if all of a sudden, the LLMs have to negotiate between two people's memories

Brian Casel:

Have have your agents call my agents.

Justin Jackson:

Yeah. And then but then they can agree the same way they would normally agree. That I think is fascinating. I think that would expose I think it would help AI companies actually train their models better. But it would also just expose how just how agreeable your particular LLM has been.

Ryan Hefner:

I mean, it's kinda interesting. You actually see some of these people's responses that they screenshot out of Claude, and they must talk to their Claude in a totally different way than I do because it's a completely different personality. And I think that that actually plays into it. Like, I'm pretty polite. I, like, say, like, please and thank you and, you know, like, that kind of stuff even though I know I'm blowing the tokens.

Ryan Hefner:

But I see some other ones where the responses are like, wow. You must say some, like, I don't know, just like kinda like some hardcore stuff to your to your claud if you're getting those responses. You know?

Brian Casel:

Yeah. It's like I don't know if some of it is, like, made up or or what kind of crazy stuff that that they're just trying to get a crazy reaction out of it. But like, yeah, I I never see I I have seen like that's that's part of why I migrated from ChatGPT to Claude for any kind of work. Because I used to use ChatGPT for the business strategy stuff, and I just found it way over the top, way too, you know, sycophancy. Like, like, it's just just agreeable, like, to a fault where obviously to a fault, but, like, to it's not helpful.

Brian Casel:

Like Mhmm. You know? Yeah. And I have I've noticed like, there's still some of that with Claude, but like, it also just has this different vibe to it where it's just trying to be constructive, and does naturally like, let me push back on on these things or maybe help help you reframe these a little bit differently. Mhmm.

Brian Casel:

Or rather than just being like, yes. Yes. Yes. Yes. Yes.

Brian Casel:

You know?

Justin Jackson:

I mean, there's also some there's some stuff that they're just bad at that like, anything novel, I just find like, I asked it to give I because Claude has lots of information about me and my business, and I asked it to give me a new novel marketing idea for Transistor that I'd never thought up before. And we went a couple rounds, and I'm like, these are the dumbest ideas ever. They're just dumb. If you bring it a novel idea, it can help you execute on it. But, like, the ideas it came up with, I was like, these are so dumb.

Justin Jackson:

And I'm like, no. I want an idea that will actually move the needle, that will make us more money. Like, give me an idea like that. And then it comes back with just something so dumb. Oh, okay.

Justin Jackson:

We're But if it reminded me to book a vacation, I'm in.

Brian Casel:

I don't know. And if that chat was sponsored by Airbnb.

Justin Jackson:

Yeah. Yeah.

Ryan Hefner:

By a cruise. Yeah. Alright.

Brian Casel:

Alright. Should we wrap it?

Justin Jackson:

I think we should wrap. Yeah. That that was great. So nice to have Ryan with us. First time we've had, I think, someone from the chat.

Justin Jackson:

So this might be something we'll do in the future. If you're in the chat and you wanna be or you're a listener. If you're listening right now and you wanna be on a future episode, let us know. We might we might have you on last minute, ten minutes before we go live.

Brian Casel:

That's right. Good

Ryan Hefner:

Yeah. Was seven minutes. I was like, sure.

Justin Jackson:

And I'll put all of the links to what we talked about in the show notes today. But yeah. Thanks for being with us, everybody.

Ryan Hefner:

Alright. Later, folks. Yeah. Thanks. Have a great one.

Ryan Hefner:

Bye.