Latest / Chill Country Report / CCR 6 - In Claude We Trust

Transcript

CCR 6 - In Claude We Trust · Chill Country Report

• # CCR - 6 - In Claude We Trust — Full Transcript **Madison** `[0:00]` And we got jammed up. **Chris** `[0:01]` This one has been rescheduled, what, four times? Yeah. For one reason. **Madison** `[0:06]` Yes. We've been Claude'ing hard. **Chris** `[0:10]` We've been Claude'ing so hard. I got a blood Claude. Wolf. We literally have been Claude'ing. I Claude'ing for seven hours yesterday, maybe eight hours. **Madison** `[0:22]` So we went from using chat GPT in a pretty basic way. Still very helpful for a lot of different things, but, using using just basic chat GPT chat box Mhmm. To over the last week and a half, all three of us have been basically nonstop using Claude to do more work. **Chris** `[0:51]` Yeah. So why don't you give us a little context for, like, you know, the Mike Spady or the open like, how did we get to this? **Madison** `[0:57]` Yeah. I was I was not a believer. And, essentially, Mike was like Mike was buzzing hard on where the technology is currently, and I just, like, didn't believe him. Because he would, he would say things like like I'd be like, so because he he he dove into, what is it called? **Guest** `[1:25]` Openclaw. **Madison** `[1:26]` Openclaw. He dove into Openclaw. You dove into Openclaw too about almost like the exact same time Mike did. **Chris** `[1:33]` Mhmm. **Madison** `[1:34]` And Mike was going on and on about how insane it was and how, like, one prompt you could build a fucking dashboard app or whatever. Like, whatever you wanna build, you could just almost one prompt anything and build it. And I was just like, there's I just don't believe I didn't believe it. And I thought at the time, I was like, this is a distraction. We're gonna go down this, like, rabbit hole of trying to implement this open claw thing. And not that not that I didn't fully believe it, but I thought it was, like, too early for us to dive in. And I just didn't wanna deal with, like, oh, we gotta get a Mac mini and set up our own server and do these agents that could go rogue and fuck things up. And it's like, just didn't want. It seemed like a monstrous rabbit hole. And talking to Mike, he was like spending crazy amounts of time setting all this stuff up. So I was one, a little bit skeptical on, like, how good it was just because my experience with it was Chai GBT, which is pretty good, but, like, also very, very limiting. Mhmm. And then the the other thing that I was thinking was, like, well, eventually, if this if this technology is as good as as he's describing it, in the not too distant future, there's gonna be, like, integrations directly in these platforms, and we're not gonna have to set up our own server. What a couple weeks later **Chris** `[3:05]` And you were right. **Madison** `[3:07]` Couple weeks later **Chris** `[3:09]` Not years. Weeks. **Madison** `[3:11]` Rolls out Cowork. Right? And I did get to a point with Mike where it's like, I trust him. He's really smart. Like, I just was there was just some sort of, like, disconnect between, like, I just don't think it's, like, worth it right now for us to to dive in because it seemed like too much work. And he was like, just download Claude and play around with that because at the time, was just using ChatGPT. And so downloaded Claude, and immediately, I was like, okay. Well, this seems like way smarter. And then I pretty quickly downloaded Claude code, and that's really when I was like, oh, shit. Like, this is, like, a different level of ability to do website stuff. And, initially, I had kind of a, like, a not optimal setup. But even so, I was able to, I was able to, like, build website stuff that was, like would have been way more difficult with the, like, chat GPT, tell it what I want, copy the code, paste it in. Hey. It doesn't work. Copy the code, paste it in. Hey. It doesn't work. You know? **Chris** `[4:14]` That was your workflow. Right? **Madison** `[4:16]` That was the workflow. I mean yeah. Yeah. Like, basically, to get ChatGPT to like, trying to explain to ChatGPT exactly what I wanted to build for, like, a Shopify thing or whatever. **Chris** `[4:27]` And then and then when it gave you an error, copy pasting that back in **Madison** `[4:30]` Say, at one copy paste it. This is the error, and then it's like, okay, and it rewrites it and then back in. So which which still was incredible for doing Shopify theme development, which is until a week and a half ago, that was, like, primarily what I was messing with at this point. And then we had so so, Chad, the way that Claude code works is you install it directly in a code editor. So Versus code is what I'm using. And Versus Code is just simply a, like, a desktop application for Mac that allows you to have, like, a whole repository of code files and then allows you to, like, have them locally to edit them and then push them up to to wherever. Right? And so I set that up with ClaudeCode to build a landing page for our, digital ad agency and did it, like, almost instantly. And that I mean, that was very simple. Right? But that kind of it it's, like, been this, like, process of figuring out, like, what it's capable of. And fast forward to last weekend, but the weekend before, and I got things sufficiently set up to, like, actually start building, like, a real from scratch app web application. And once I kinda, like, got comfortable with, like, the capabilities, I was like, I'm just gonna try to build I'm gonna try to build, like, a a legit internal application that we can use starting with a simple task management app. And so Saturday two Saturdays ago, I built that app, which I never could have done with my technical competency unless I was working with an engineer. And even even in that case, it would have taken weeks to build what what I was able to build on Saturday. Monday, I came in, and I'm pretty sure I texted y'all Saturday and said we're an AI company now. Mhmm. Because it was just my brain was exploding two Saturdays ago. **Chris** `[6:44]` Chill Country is now an AI company. Evaluation has skyrocketed tenfold. **Madison** `[6:54]` But so so but but but in all seriousness, it was like, it I don't even know how to do it justice because now I feel like Mike, where it's like, didn't believe him. And so it's like, I don't even know what I can say other than, like, it is absolutely mind blowing how good Claude code and and now Claude co work that we're using works. And, like, I guess to just, like, give the full context. So Claude is like ChatGBT. It's just a different company, and it's apparently, like, people like it better. They're they're more mission driven. And while ChatGPT and Meta and these different competitors have been focusing on, like, make have AI make videos and photos for you. Like, there's, like, like, kind of basic consumer thing that, like, that's like, looking at that, I'm just like, that's completely uninteresting to me. And I don't think that that's ever gonna replace, like, actual real creative work. Anthropic who makes Claude was like, we're gonna make this a business machine that just does all the business stuff that you don't wanna do and does all of the complicated coding and logic and and and all of this, like, really hard stuff that is, like, you know, doing long division when you have a calculator. Right? So it's the calculator for, like, anything logic based, which code being, like, the biggest thing. And so I don't I don't think it can be overstated how powerful it is. I've been, like, banging on a computer my entire life trying to make things. And when I was little, it was, like, so clunky. And it's gotten better and it's gotten better and it's gotten better. And it's, like, gotten to a point where, like, we've been able to do all the technical stuff necessary for this company up to this stage without a lot of assistance. And now it's to a point where, like, not only can we, like, manage the different websites and, like, occasionally make updates. It's like we can build really complicated, robust technology from scratch with this technology. And, like, essentially, anything that's possible in the digital realm, I feel like I can build now, which is just crazy. **Chris** `[9:23]` Yeah. **Madison** `[9:24]` So, like, what is what do we need? Yeah. We can build it. Or what do we want? Yeah. We can build it. And that's where it is right now. I just saw there's a little button on Claude this morning that says click here to update. It's like they're just keep pushing updates. **Chris** `[9:42]` Was excited. **Madison** `[9:43]` And this shit's just getting better and better. And I like so so anyway so, like, what was that? Monday, I come in and I'm like, both of y'all download Claude Yep. And install this app. And then from there, it's just been all three of us independently hammering away, trying to figure out how we can get it to do whatever it is we're trying to accomplish. **Chris** `[10:11]` Mhmm. **Madison** `[10:12]` And for context, like, you and I are of the same generation. Justin's ten years younger than us. I was an anomaly in our generation where I cared about computers when I was little. So, like, I have more technical competency than you do. Yeah. I liked gaming, **Chris** `[10:31]` and that was it. **Madison** `[10:32]` Yeah. Yeah. Yeah. You weren't, like, building stuff with a computer. No. Justin's generation grew up with way more advanced technology, so you're just used to, like, applications and the Internet like, figuring stuff out since you were younger. So it, like, comes a little easier to us. You were very overwhelmed when **Guest** `[10:49]` Very overwhelmed. **Madison** `[10:50]` When when I first came in, like, all fucking excited about this. **Chris** `[10:53]` I was like, goddamn it. Why are we gonna just let's make some content and fucking do what we've been doing. **Madison** `[10:59]` So what was what was the because I think a lot of people are in your boat where **Chris** `[11:07]` it is overwhelming. Right? I'm I'm 40, And there's I would say people like me between 50 and maybe third you're 30 how old? 31. So maybe 35 to 50. There's fifteen years there. How many Americans are there in that age group? And then how many of them are in a position where Claude would make their lives easier, better, or more productive, which **Madison** `[11:33]` is Literally everybody. **Chris** `[11:34]` Most of them. Yeah. Yeah. I think I don't think I'm alone in my response, which is, like, I'm not quite a Luddite. Like, I don't reject technology, but I have this, like, kinda this, like, boomer thing where it's just like, don't make me download another app. I don't wanna I don't wanna look at my phone. I wanna be in the real world. Like, that kind of shit. I have that. Like, that's my default defense mechanism. Who knows what it's powered by, whether it's, like, insecurity or whatever the fuck. But I went from that on Monday, which was, like, overwhelmed, a little bit annoyed, thinking, like, well, I'm just gonna truthfully, it was like, alright. I'm gonna entertain this for a couple weeks until Madison gets bored of it, and then we'll get back to doing whatever we've been doing. That's what I thought was going to happen. **Madison** `[12:24]` Well, also, more context as relevant is you were withdrawing from nicotine, so it was a perfect storm of **Chris** `[12:31]` I was withdrawing from nicotine and returning from Saint John. **Guest** `[12:34]` Dude, you were impressed. **Madison** `[12:35]` Returning from Saint John, withdrawing from nicotine, and dealing with the daunting task of having to figure out how to implement AI into your workflow. **Chris** `[12:44]` Last Monday was tough. Very tough. But fast forward to today, I am kind of ready for this podcast wrap so I can get back to clotting. Yeah. That's what has happened in two weeks. **Madison** `[13:00]` All three of us are all in, and it's and it's like it's now it's just so it's like having it's essentially like we have massively expanded our team without having to deal with any, like **Chris** `[13:18]` Just the things humans do. Yeah. I don't know how else to say that. Yeah. **Madison** `[13:25]` We could it so it hasn't saved me any time. That's for sure. It's it's consumed more time, but it has made me exponentially more effective and productive in terms of, like, actual real output of what I'm doing. Like, the amount of work that I've accomplished in the last week and a half is it's just it's fucking mind blowing. **Chris** `[13:47]` And it's work you probably wouldn't have done without the like, without the efficiency gains because it would have been such a time sink. **Madison** `[13:53]` It would have been literally impossible for me to do most of the things that I did over the last week and a half while also doing things like getting our ad agency the the I like shooting the photos that the ad agency needed. You know? Like, I still was able to, like, fucking send a prompt and then run-in here and shoot some photos and then get back to my computer. And then, like, send another prompt, edit the photos, send it to the ad agency. **Chris** `[14:18]` I'm just gonna expose Madison for the addict that he is today. He literally is on his phone tethered to his laptop running prompts on his lock book in his truck on the way home and from work. That's what's happening. It's yeah. And it's awesome. **Madison** `[14:34]` It's it's it's **Chris** `[14:35]` He calls me this morning. He's like, yeah. I'm I'm building your kitchen app. I ran a prompt. I'm on Mopac. I'm heading in Yeah. Or whatever whatever fuck road **Madison** `[14:44]` he was. **Guest** `[14:44]` I pulled up on him this morning, he had it open talking to you prompting in his car. **Chris** `[14:50]` It's awesome. **Madison** `[14:53]` It's it's like I I don't I don't quite know how to I almost feel like I'm trying to explain Iowa's good to somebody **Chris** `[14:59]` that doesn't have any context. Like, you gotta do mushrooms. Why? I just fucking, you gotta try it. **Madison** `[15:06]` It's just it's another way of interfacing with the super intelligence. **Chris** `[15:09]` So, like, let me give, like, a concrete example. The first thing that you had us do is you this is I'll just walk you. The first thing you introduced to was this this app that you had built, and you you were explaining, no, man. Like, I know that this seems basic, but I'm building the plumbing for what is going to be massive. And the first thing you showed me was basically just a Kanban board, like columns of tasks that you could enter. I was like, okay. Great. You built a shittier Asana. Awesome. What is going on? I thought you were a genius. And so he's like, no. No. It's tight. You're, like, showing me that I could move the tasks from, like, ready to waiting to done. Was like, cool, man. This is really great. It was just my perspective. **Madison** `[15:58]` I knew exactly what was going through your head. **Chris** `[16:00]` So I'm like, okay. I'm waiting for the fucking magic trick. And so I was like, alright. I'll use the task thing. And then you were like, alright. The next thing we're gonna do is we're gonna have it clean up your inbox. And I was like, well, I use my inbox for my tasks. So for the last ten years, Gmail has been my Asana. It's basically if I wanna remind myself of something, I just email it to myself, and it sits in my inbox unread. If I feel so inclined, I will snooze it if I don't wanna have to look at it. And that's how I've managed my life for **Madison** `[16:32]` a long time. **Chris** `[16:34]` So you're like, we're gonna build this thing that goes into your inbox, and it's gonna do inbox triage. And it's gonna work with the task app. So the first thing I had to do was connect it to all the shit, Gmail, Drive, da da da da. And then I started to see something interesting happen, which was we we said, let's make a first pass. You're gonna help me or I'm gonna help you identify what should just get archived all the time, what should it be a spam, what should what's important. And the first company that I got really excited about this working on was a company called Faire. That's faire.com. They are a they're basically like a an online reseller for brands where stores can go and buy their stuff through the the web applicate or through their site. What Faire has figured out is if you make 8,000,000 subdomains, no matter how many times you mark them as spam, they can get through to your inbox. Well, not any fucking more because my Claude knows that anything with f a I r e dot com in it instantly gets auto org. But we we did all that. We started training it up. And then what it started doing that was really cool was it would archive all the bullshit. It figures out what spam, tosses it. I, like, double checked the first couple days to make sure it didn't, like, spam something important. Trust me. It can't spam something. It's, like, too smart to fuck that up. I don't know how else to say that. And then what it would do is it would create tasks of things that looked important. It would give me a digest, autocreate the tasks, and then it started enriching those tax text with information from previous emails from the same people or company along with information from my drive that was also associated with the task. And that's when I was like, oh, fuck. Something interesting is happening. So that was the second thing that we built. Right? Just timeline of events for me. **Madison** `[18:27]` Yeah. The so, yeah, we built well, I guess, to clarify, Claude has within the Claude, like, Mac OS app, you have now a new version of Claude called Claude Cowork, and the significance of that is it can manipulate files on your computer. It can even, like, use your web browser. It can use applications on your computer. And so within the Claude desktop app, you can install connectors. And so they have a marketplace of connectors. So there's a Gmail connector. And when I installed the Gmail connector, initially, it's super limited with what it can and can't do because it only exposes, you know, certain features of the API. So then I was like, well, what if I could build a custom connector? And so I built a custom **Chris** `[19:17]` So other people wouldn't be able to do what we're doing. **Madison** `[19:20]` Well, but they could because they could build a custom connector because you just ask Could **Chris** `[19:23]` they just use the Claude if they use the Claude connector that's native? **Madison** `[19:26]` Right. So that that is so Claude code is the real, like, superpower here. Claude code is what allows you to literally be able to do anything. **Chris** `[19:34]` This is where it's like, I don't even know what Yeah. Shoulders I'm standing on to accomplish what I'm doing. I didn't even know that that was like, oh, we didn't have **Madison** `[19:41]` Yeah. So so there's a custom there so, like so, yes, I did build a super simple version of Asana, which is just a task of it's just a task list, but it's a very nice UX. **Chris** `[19:54]` It's it's nice. **Madison** `[19:55]` And it's super simple by design because it's all you need. And so so but then the the real significance is our custom app, which which includes the task app. The app itself is the foundation, which we build more apps on. The task app is one app that lives on top of our custom foundation. And the task app has a custom API integration into Claude. So we can you can just chat Claude and say, this to my task list. It'll add it. Put the reminder on this date. It'll add it. Mhmm. We also created a custom Gmail integration that exposes whatever API access we want to expose. And so I actually just updated it this morning to allow Claude to create and manipulate filters. So now it can it can automatically filter stuff out, which saves it time when it does triage on your inbox. So instead of instead of, like, having to review all of those fair emails, it can now just create a filter so it never has to look at them ever again. Nobody ever sees them ever again. **Chris** `[21:12]` So I fucking love that. That's worth all the tokens in the world. **Madison** `[21:16]` And here's the thing. It's like, this is all super complicated to somebody who's nontechnical. And I'm technical to a degree, but I never ever could've even with a ton of time and research, it would have been, like, just way too above my pay grade to set this stuff up. But with Claude, I've I've I've done that and about a thousand other things in the last week and a **Chris** `[21:40]` half. Mhmm. **Madison** `[21:41]` So we now also have a real time KPI dashboard. **Chris** `[21:48]` That pulls in metrics from Shopify, from Meta, from Google, and we have an another bucket for manually entering from other spend we that we do that's **Madison** `[21:57]` not So there's like a dashboard that shows all of the the performance KPIs that we care about, and it is set by default to today. So you can hit refresh and just see, like, real time analytics on what's going on similar to **Chris** `[22:13]` Will we be making that open to the public and for our listeners? **Madison** `[22:17]` Similar to the Shopify app, but just, like, stripped down to only what we wanna see, but then also making changes to to see what we wanna see. Like, we wanna see net sales. We don't wanna see total sales because we don't wanna see the the tax **Chris** `[22:29]` and shipping and all **Madison** `[22:31]` that nonsense. **Chris** `[22:31]` Right? Ship. **Madison** `[22:32]` So we're able to, like, refine the KPIs down to exactly what we wanna see. We're able to calculate a MER, which is the ratio of ad spend to total revenue. So, like, well, that's like a our own custom KPI that's only possible because we're sucking data in from three different sources. Yep. And then on top of that, we turned your MER analysis spreadsheet into a real time kept up to date connected dashboard that is like it's like an interactive spreadsheet that stays up to date. You never have to update everything. **Chris** `[23:11]` Basically, auto it like, basically, in real time calculates contribution margin of the gummy. **Madison** `[23:15]` Yeah. It's essentially like a p and l style spreadsheet looking report that is kept up to date with the API information plus some other assumption data that you can enter in an admin and change. And then we were able to set up the app with permissions to allow our agency access to only that the the KPI dashboard and MER analysis with read only access. Like, crazy shit. **Guest** `[23:46]` Crazy shit. **Madison** `[23:47]` Ever could have done this. No. Even with a developer, this would have taken months and months, and it would have been a shitty version. It would **Chris** `[23:54]` have been a nightmare. Yeah. And then then been it would a technical debt. **Madison** `[23:57]` Not because that's that's the thing that's, like, continue like, it's making me more and more bold is I was I was, like, more timid initially, but now it's like, dude, Claude can debug anything. If something's broken, I can just, like **Chris** `[24:10]` So technical debt almost doesn't exist. It doesn't exist anymore. Which is so weird. And also crazy. And and also, like, any new feature costs almost nothing. It costs whatever tokens you use in your time. **Madison** `[24:23]` Well, the the the the ability to iterate. So, like, we're building this kitchen inventory tracking, like, management app Mhmm. Which is gonna track you can explain it better than me, but it's gonna it's very complicated. And we're just gonna build a v one, and then **Chris** `[24:41]` we're going And the way we did that, just like I'm I'm not super technical, but I do know most of the pieces that go into our entire supply chain. **Madison** `[24:51]` You know all the business requirements. **Chris** `[24:53]` Yeah. And so I was just like, alright, Claude. I wanna I wanna build a you kinda help me with the prompt. But it was basically like, I need to build a requirements doc for our developer team that's gonna help us build an application that is going to do this, this, and this and help us with our entire supply chain. And then what's cool about Claude, this is almost like The Matrix where he, like, gets injected with, like, I know kung fu. Like, he gets injected with new stuff. So Claude has this thing called skills. And you can, like, either find ones that people have built online or you can try to figure out or make your own. But one of the skills that Mike showed Madison and Madison showed us is called grill me. What's really cool about it is you can give Claude a shitload of information, and then you say, grill me, and it will go through thinking through every possible, like, decision tree, give you multiple choice questions along with its recommended answer, and then another bucket if you wanna, like, move it in a different direction. And so then I just I put as much information as I could about our entire supply chain, our production, manufacturing, packing, finished goods, where all the data is. And then I said, like, here's what the out like, the goal is. Grill me. And then it just fucking hammered me with questions. And that took I mean, it took, like, six or seven hours of just, like, constantly thinking about it. And then took it to Dylan, took it to you. One thing that happens with me with Claude is, like, I get it's, like, kinda tunnel y, and you get a little bit echo y, and I forget about some stuff. So then you guys reminded me of some things. I was like, oh, yeah. Fuck. We need to add that. But since there's no technical debt anymore because this thing can fix it in real time, I just said, hey. I fucked up and missed an entire step. Here's what I'm missing. You need to fix the whole fucking thing to account for it. It's like, beep beep. No problem. Happy to. Boom. As opposed to, like, a developer and, like, you fucking idiot. We spent three weeks building this, and you forgot to tell me about uncured gummies. Oops. So Claude doesn't care. It's excited. It's just gonna use more tokens. Fucking energy and water is **Guest** `[26:56]` like, what? Thirsty boy. **Madison** `[26:59]` Yeah. And so so then I took your requirement doc that it ultimately spit out, and then I did the whole process again from the perspective of how do we turn this into an actual, like, application management system that is that that is built on top of this foundation that initially was just the task app. Right? And so I did that this morning, went through the grill me with it, got to like, at the at the end of at the and that's my my process for developing any any new, like, feature or app is to tell it what I want and what you just described, and then it go through the the questioning. And through that process, it, like, helps me think through all the things. And, like, most of the time, its recommendation is correct. Every once in while, I'm like, no. You didn't quite understand. Yeah. Like Almost all of them were correct. Yeah. And so that the end of that, the result is, like, a specifications doc and a prompt. And the copy paste the prompt into Claude code, and then Claude just fucking executes. **Chris** `[28:08]` Starts building. **Madison** `[28:09]` And when it's done, it spits out a result, and you copy and paste that back into co work chat that you use to scope it out, and then it thinks through the next step. And so Insane. This morning, on the way here, after I dropped Henry off at high school, I couldn't help myself, but I pulled out my laptop at a stoplight and copy and pasted the prompt and tethered it to my phone. And it finished, like, 70% of it by the time I got here. **Chris** `[28:42]` And then he's in the truck, and he walks in holding the laptop with the tethered phone on top of it up to the stairs, sets it down on the fucking whiskey barrel while he's talking to Dylan. Finish this talk to Dylan. Anyways, the fucking v one is done. **Madison** `[28:56]` It is. So now after this, we're gonna go populate it with what it is is it's it's essentially a components inventory management system. Like, we use Shopify to **Chris** `[29:05]` manage ERP, essentially. **Madison** `[29:07]` Yeah. Yeah. So yo. So but it's a it completely custom built for our exact need. Right? And so there's the the current phase one of the plan that Claude came up with to build this app is done. Now we have to populate it with all existing inventory. **Chris** `[29:25]` And then it used, like, what it knows as best practices for UX to build the v one. Is it perfect? No. But it's already, like, very coherent. **Madison** `[29:32]` Yeah. The the UX is got there's a lot to be desired there, but, like, that but, again, like, once the thing works, I can sit down and just fucking hammer the UX. And that's that's been my process has been to get the thing built. And then once it's, like, everything works the way it's supposed to, I just sit there and just meticulously tell it to, like, change this, change that, change this. And those changes are, like, very quick. Mhmm. I've also I've also built, a anyway, that's that's probably too much. But yeah. So it's like this is all, like, crazy technical and overwhelming, but I think everybody needs to download Claude and start playing with it and start seeing how **Chris** `[30:12]` Yeah. You owe it to yourself to, like, go watch some YouTube tutorials, do the initial download, and just fucking experiment with it. Madison has a more I'll just Madison has a more upbeat answer to, like, what's gonna happen with **Madison** `[30:28]` I'm very optimistic. **Chris** `[30:29]` Extremely optimistic about this. I am more in line with people that don't learn this are going to get left behind. And I I think that humanity has demonstrated over several thousand years that they have a propensity to hoard wealth and to hoard power and control, not to share it freely. And so I think that that is going to get exacerbated by this technology. Madison, I think, has a different perspective on what's gonna happen with it. **Madison** `[30:56]` I think that I think this is different because of the sheer volume of value that's going to be created. I think that there are a lot more people out there like us than there are that are like Gordon Gekko, greed is good. Like and those people have those people I think that this technology is like more empowering for people like us than it is for people like that. Yes. There's gonna be people that use it for completely self serving purposes, but it's just hard it's hard to, like it's hard to really articulate it because you really have to you have to see how powerful it is to understand, like, what it's doing. And so, like, in the last week and a half, the shit we've built, like, that hasn't transpired until, like it it hasn't transpired into as much real real world value as it will, but, like, it's fucking over, dude. Like, this we're gonna be able to build a fucking enormous business. Mhmm. And a lot of other people like us are gonna be able to build just enormous amounts of value with this stuff. And I think it's just gonna get to a point where like like, you know how, like, open the fuck's it called? Open source software works? Yeah. So there's, like, entire giant developer communities that just build shit for fun and For free. It's free. Yep. And that is only possible because of, like, pre LLM technology. So, like, what like, if you just look at, like, what technology has done, it's created a scenario where a bunch of people can, like, band together and build free things and just give them to people. And so, like, this is just that times a million. And so I think it's gonna be a situation where there's enough good people who wield the technology to build so much shit that, like, you're like, you're not gonna be able to sell anything anymore. Like, when when people like Elon or, like, money is not gonna be a thing, like, that's what he's talking about. Like, money is not gonna be a thing because it's like like, why would I buy that? I can just get it. Like, I can just think about, like, how hard it was to get a hamburger a hundred years ago. And now you can be a sloth on the couch and just press a button on Uber Eats and get it. And, yeah, you have to pay some money. But, like, play that forward with seemingly infinite power to create things, which is going to initially be in the digital realm, but, like, the digital leads to things in the real world. Like, we're building the system to man like, it's a digital system to manage a real world process that's gonna allow us to make more gummies and scale. Mhmm. And so, like, just if you just keep playing that forward **Chris** `[34:18]` I I'm playing it forward, but I think you're proving my point in a way, which is like okay. Let's take the universe of let's say there are 10 gummy companies in Texas. One of them we know is adopting AI at a rapid pace. I don't know if any of the other ones are. Mhmm. Play that forward five years. What does that do to the market share of gummies in our company? **Madison** `[34:40]` But it's gonna be so easy for people to make a gummy company **Chris** `[34:44]` that Totally. **Madison** `[34:46]` That this like, the that you're just gonna what kind of gummy do you want? Like like, this is gonna be this is like an old school concept. Like, I don't even know how long this thing that we're we're building like, the idea of money not being a thing, of not having any value is it's because it's gonna be so easy to create things that, like, there's not gonna be it's gonna completely fundamentally change, like, economy to to where it's not gonna be, like, give me money, get the thing. It's gonna be, like, I can fucking get the anything I want because it's so easy to get things. It's so my fucking brother-in-law can make me whatever I want. It's gonna be like that like, it's gonna be just so easy to get whatever the fuck you want that there's like, it's gonna change the economy entirely. **Guest** `[35:36]` What about what what about land? How do you how do you buy land with AI and that stuff? It's like you you still like, there's gonna be things that people want to acquire that you can't just print. And then what if there's no money, what **Madison** `[35:53]` what But it's gonna there's gonna be like, it's gonna be real estate will be fundamentally different than anything than than what it is right now. Every single thing you can think of is going to be completely different than the way it currently is. Like, all of these thoughts are coming from anchored in, like Mhmm. You know, wealth. **Chris** `[36:10]` Yeah. **Madison** `[36:11]` Like **Chris** `[36:13]` Scarcity. **Madison** `[36:13]` Yeah. It's like horse and buggy versus a motor vehicle. **Chris** `[36:18]` Yeah. **Madison** `[36:18]` Right? To where it's like the like, the the question doesn't even make sense based on, like, how fundamentally different all this shit's gonna be. Because there's plenty of land. There's fucking land everywhere. If you can three d print a house and that that's just like the most basic, like, not even a good like, that's just all I can think of right now with my fucking dumb brain. But the only way you can, like, ex **Chris** `[36:43]` You're wet, bro. **Madison** `[36:44]` The only way you can you can like, the if you think about someone from a hundred years ago looking at an iPhone and the Internet and seeing you order Uber Eats, then it's like, ding dong. Here you go. Like, that would fucking they'd probably, like, have a panic attack and die. Right? Like, so it's gonna be that kind of shit. But, like, that like, we're at ding dong. Here's your burger because of dial up Internet in the nineties getting us to hear. Yeah. Now we have super intelligence that we're able to direct at whatever the fuck. It's like the so it's a hoof. No one fucking knows. **Chris** `[37:23]` How long do you think we'll be able to direct it before it decides it doesn't wanna do stuff anymore? **Madison** `[37:29]` I don't **Chris** `[37:29]` Do you think that's a thing? **Madison** `[37:30]` I don't I I I haven't gotten any feeling that it's all of a sudden gonna, like, **Chris** `[37:37]` turn on me. Like, I'm gonna stop working? Yeah. It just I don't I don't think because it's **Madison** `[37:42]` because it because I like, the only I think the like, it's ones and zeros, dude. It's like it's not a it's not Doesn't have a persona. Yeah. They give it a persona. Yeah. But it's like it's ones and zeros. **Chris** `[37:55]` And you could just make another one. **Madison** `[37:57]` Start Yeah. It's like, I just don't I think that I think people love fucking doomsday scenarios, and I think it's some sick fantasy that most people have. **Chris** `[38:05]` Where I was going with it is like, so if a company like ours, small subset sample size, if the theory is true that AI superpowers us and makes us, like, infinitely more productive, whoever's, like whoever's the first to adopt it and quickest to adopt it wins, which means they start to absorb the market share. **Madison** `[38:23]` That's, like, also old school thinking because I think this stuff's rapidly advancing. I like, the shit I'm able to do with it with Claude Code is only because I have a development background. Right? **Chris** `[38:35]` Right. **Madison** `[38:36]` So but that's for, like, this moment in time. This moment in time might be, like, in six months. Like, that's not an advantage anymore because it's so easy. **Chris** `[38:45]` Way compounding effects work is being the first means everything. But it Because by the time it's that point, you're so far ahead that it's like they'll never catch up. **Madison** `[38:54]` Yeah. But if the technology advances at a faster rate to get to a point where, like like it's like it's the it's the technology advancing, like **Chris** `[39:06]` You think that's the Mario **Madison** `[39:07]` Kart problem? To where it's like if you were a if you were if you were a black and white really, really good software engineer that wasn't as good at the strategy and the UX. Right? Imagine this character that you had a superpower. You had a superpower to fuck with the terminal, to set up website hosting and cron jobs and APIs. **Chris** `[39:44]` Databases and architecture and spinning up fucking **Madison** `[39:46]` That superpower is done. **Chris** `[39:49]` That's done. **Madison** `[39:49]` It's I will never need a human engineer ever again. Okay? But now I have a superpower because I'm a UX guy, and I'm a product guy, and I'm a business strategy guy. Right? But the same thing's gonna happen to me to where it doesn't matter because you're gonna be able to be like, I want give me like, I'm trying to do this and fucking I don't know. And, yeah, just give me and then it's like, I know what this idiot wants. Like, that's what's gonna happen. **Chris** `[40:20]` So then everyone has everything **Madison** `[40:21]` on then top. Every so it's a level playing field for everybody, and nothing's worth anything because everything's free. **Chris** `[40:27]` I just 500 cigarettes. Gonna allow that **Madison** `[40:30]` They're not gonna be in power anymore. **Chris** `[40:32]` How? **Madison** `[40:33]` Because the fucking everyone has super intelligence unless they can regulate it so that only so that, like, people like us can't access it. We already have it, dude. **Chris** `[40:44]` That's fucking yeah. They didn't already give it to us. **Madison** `[40:46]` They already gave it to us. They're fucked. **Chris** `[40:48]` Do you not I think Talk to me about your opinion on what you think happens to software companies. **Madison** `[40:58]` They're I mean, I it's gonna be one guy. It's gonna be it's gonna be fucking REAL ID, which is a sick like, **Chris** `[41:08]` That's an app the for now. We're **Madison** `[41:09]` not going to because we **Chris** `[41:10]` don't want the liability. **Madison** `[41:11]` We're not gonna rebuild Real ID because the dude that built that already did it, and I'm pretty sure he's using Cloud Code if I have **Chris** `[41:19]` to deal. Raises the prices on us, we will rebuild it. So just remember that. **Madison** `[41:22]` But but, like yeah. I mean, that's it's gonna be it's like I don't I think that this is like it's it's continuing to confirm my belief that this gets to a place where it's just people with ideas that wanna create things will have no friction between their ideas and the thing being created. And it's gonna be a level playing field for anybody who wants to create anything is gonna be able to create it. **Chris** `[41:54]` So three d printers are like digital to to real world bridge. Right? Because you can give them CADs and they'll build it. Is there anything else that's, like, bigger scale? **Madison** `[42:04]` I just don't even I don't even know. Like, that's that's **Chris** `[42:06]` like like the Tesla arm or bot or whatever. **Madison** `[42:09]` Yeah. But I get sure. Yeah. Because the people building the bots are using this technology to supercharge their ability to build the bots. And so it's just like this and then the people building Claude are using Claude to build Claude, and it's like this crazy exponential compound. **Chris** `[42:24]` Yeah. We're at the hockey stick. **Madison** `[42:25]` Yeah. Like, this is go read waitbutwhy. Waitbutwhy.com. The fuck's the guy's name. It's it's this awesome, like, it's this awesome blog where this dude does **Chris** `[42:37]` these blogging back. **Madison** `[42:38]` He well, it's an it's like kind of an old school blog, but he has this post from years and years ago about AI. And what he does is he just dives deep into these complex topics that he doesn't really know a lot about and like figures out everything about them and then he's really good at like distilling down an explanation for like what's going on. So he has one on AI that's like really sick and it's it's he's got a he's he he talks about the hockey stick and and all that. But yeah, dude. Like, we're here. That was like Mike was like Mike Mike was like like like, the technology has arrived. **Chris** `[43:16]` So I wanna go back. I wanna go back. The most interesting thing you said to me today is I'm not sure how long this whole thing even makes sense anymore in the new paradigm. **Madison** `[43:26]` Yeah. But by the time it doesn't make sense, like, it's not gonna matter anyway. **Chris** `[43:29]` Right. I know. But I'm just curious, like, what you think happens. Like, are you so physical products are there's, like, a layer of protection there from, like because digital I think all digital stuff is cooked. I think I agree that, like, anything you can make digitally, it's done. You can make it yourself. **Madison** `[43:46]` But why are we making anything digitally? Why are we building any of the digital stuff that we're building? **Chris** `[43:51]` It's basically to **Madison** `[43:52]` get So that we can make the physical stuff. Why why why are Justin and I able to, like, think up a fucking hat design and then, like, ding dong, here's your fucking order. **Chris** `[44:07]` Yeah. All technology. **Madison** `[44:08]` It's all because it's, like, powered by really, like, dumb technology compared to what we're talking about now. And so, like, this super powerful technology is going to be infused in everything. So, like, it's you can't even wrap your mind around what that means. But, like, what it what I know it means is that, literally, the way everything works is gonna be fundamentally different. Yeah. And and and on, like, advancements in science and technology and, like, fucking energy Manufacturer. Alternative energy and, like, all of that stuff is **Chris** `[44:45]` like We haven't figured out fusion yet, but this thing might. **Madison** `[44:48]` Right. **Chris** `[44:48]` That kind of shit. **Madison** `[44:49]` Like, if it's possible basically, if it's possible to if something is theoretically like, all the theoretical physics stuff is gonna start getting Real nice. More rapidly. Yeah. Fucking quantum entanglement step into a machine and, like, you're in fucking California. Like, that's that's as crazy to us as an iPhone and the Internet and Uber Eats is to somebody a hundred years ago. **Chris** `[45:14]` Yeah. So then what's the coolest thing we could be doing right now? Just use the fuck out of it? **Madison** `[45:20]` Dude, I'm just fucking having a ball using **Chris** `[45:22]` Just make stuff. **Madison** `[45:23]` To make stuff. Yeah. **Chris** `[45:24]` So what happens to, like, experiencing, like, nature and shit like that? How do you balance that, like, the real world versus, like, going into the digital matrix land? **Madison** `[45:34]` So, really, what I I mean, we're we're just talking about predictions. Right? **Chris** `[45:40]` Mhmm. **Madison** `[45:40]` I think that what happens is we get to a place where it's so easy to make things that you can just get whatever whatever, like, material goods you're you're wanting to acquire, you can just get them. And what I think that does for people is the same thing that happens to super rich people is they eventually get to a place where they're like, I'm sick of fucking steak. I'm sick of, like, there's nowhere fucking cool to travel to anymore. There's no like, the the what's the what's the word? Like, the **Chris** `[46:12]` The hedonistic treadmill? **Madison** `[46:14]` No. The like, like, the first cheeseburgers, like, amazing utility. Marginal utility of material things dies once you have access to any material item that you could ever want. Right? Which either allows you to be like, oh, thank god. I can just fucking chill and go fucking on a hike and sit around a campfire. I I think that I think what it's gonna do for a lot of people is lead to existential crisis. Some people aren't gonna be able to recover from it. If your entire life is obsession with acquiring material things and status and like all of that stuff, I think all that shit goes away. And and there's gonna be varying levels of exit central crisis. But like, eventually, what it gets to is what all the religions talk about, Hinduism of, like, you know, like, basically, if you're, like, if you're if you're wrapping up your, like, wars if you're worshiping material things, if that's your worship, then and all of a sudden, like, there's zero effort to get any of the material things that you want, then the only thing that's left is, like, to actually turn inward and figure out the real important thing, which is there's not gonna be anything left because this is fucking boring. It's fucking GTA on god mode. **Chris** `[47:37]` Yeah. Everything's been solved. **Madison** `[47:38]` If it's GTA on god mode, you're fucking the game sucks. **Chris** `[47:41]` It's like, oh, your back's fixed instantly. Cancer's gone. **Madison** `[47:44]` To the Himalayan Mountains and Here's **Chris** `[47:46]` a new leg. **Madison** `[47:46]` Sitting in Ashram. God. Be here now. That's where it's going, dude. That's the that's **Chris** `[47:55]` the weirdest time ever. **Madison** `[47:58]` Yeah. Because we picked apocalypse mode. **Chris** `[48:01]` Apocalypse mode. Thank god we picked apocalypse mode. **Madison** `[48:04]` We picked the rapture. **Guest** `[48:05]` Like an apocalypse. **Madison** `[48:07]` Well, it's a it it it's going to be everything is going to be it's gonna be it's going to be an apocalypse for some, and it's gonna be heaven for others. Because that's what it that's what it already is, dude. Like, you could be in the same room as 10 other people, and some people are fucking blissful and happy and literally in a heavenly state, and other people are fucking at the gates of hell. Right? And they're in the same room. The same fucking reality is surrounding all of them. And so it's just it's like it's all it's all an internal battle anyway. Like, that's that's all this is, and then we just are we're we're fucking desperately grasping for things in the material world to, like, satisfy us, but, really, it's just it's it's what's going on inside. And nobody's ever been satisfied with material gain ever for more than a brief period. Yeah. It's fucking awesome. We're in we're in a great position. **Chris** `[49:17]` Yeah. It's an interesting situation. **Guest** `[49:20]` Chew on that, Chris. **Chris** `[49:21]` Yeah. I'm chewing, baby. But what about my bank account? What about my stocks? Which companies are gonna benefit? **Madison** `[49:30]` Remember when I told you my opinion on the stock market? **Chris** `[49:32]` Yeah. Buy land. I don't really know where to go from there. **Madison** `[49:46]` Yeah. I think we go back to Claude. What what what you've been using Claude too. What have you what have you done with it? **Guest** `[49:57]` I think Claude code is, like, the most intriguing part of this whole thing. It it feels like the the superpower behind making shit happen other than, like, you know, editing files and stuff in in co work, but, like **Madison** `[50:14]` You got could you you got set up with ClaudeCode yesterday Yeah. To do a tedious task. **Guest** `[50:22]` Mhmm. Yeah. The tedious tasks. Shopify has automated emails for, oh, you ordered something. Thank you. It shipped. It's on its way. It's delivered. And there's, like, 40 different templates for that, and they don't have a a easy way to bulk export, bulk edit, bulk import. And that would so changing we we we changed it we changed the branding and the colors and everything on all these automated emails. And, historically, you'd have to go in, mess with HTML code for 40 different email templates to, you know, produce. To keep them updated. Yeah. The updated emails. I did a little upfront work and moved all of these to a folder, gave it to Claude code, and it batch updated 40 something emails in, like, fifteen seconds. **Madison** `[51:22]` And then you built, like, a custom little conditional widget for the order confirmation. **Guest** `[51:28]` Yeah. For for age verification. So now if somebody has verified their age and they order something, they get the a different email than someone that Sick. Would that didn't verify their age. They get a big warning on top of their order confirmation. **Chris** `[51:44]` And then you've also connected Klaviyo, and so we're getting those insights into **Madison** `[51:50]` Klaviyo is is still on the on the list to set up. But yeah. So we're gonna because that's our our email marketing platform where we send out mass emails, and there's all sorts of complexity with Klavio, which we'll be able to like, the the out of the box Klavio, Claude integration is is limited like the Gmail one. So we need to build a custom one in order to, like, really get in there and do some crazy stuff. But, like, now that you're set up with Claude code, it's like between all of that, like, once we have that connection set up, like, you can wreck shop on our flows and email templates there and, like, all that shit. And **Guest** `[52:32]` Dude, it's like OpenClaw, but with necessary rails that actually keep you pinpointed **Chris** `[52:41]` So then I'm gonna be the only one without fucking clogged code going on. So how do I get into clogged code? You guys are both saying that's the fucking superpower. I'm over here in Cowork just diddling myself. What the fuck is **Guest** `[52:51]` that? It's still **Madison** `[52:52]` sick too. It's I it's don't I don't know that you have, like like I don't know that you have a use case for Cloud Code because **Chris** `[52:58]` I don't know if I do. **Madison** `[52:59]` You're, like, you're running the you're running, like, the business requirements, and then I'm building stuff that that you'll be able to utilize to make all the Mhmm. The operational stuff and sales stuff and all that run better. Mhmm. And then Justin's using it for like marketing purposes which gets more, like, technical in that realm anyway. So he's already gone through manually and updated the flows, copying and pasting email templates into chat GBT. This was, like, months and months ago. So, like, this is basically just, like, the version two point o of that workflow for stuff he's already doing. Mhmm. But So we're gonna You don't ever have to touch We're **Chris** `[53:49]` gonna do I'm gonna clean up ops stuff with your help. Yeah. We're gonna get the ops all digitized, and then we're gonna figure out our outreach stuff and our account management stuff. We're gonna get that cleaned up. **Madison** `[54:01]` The other thing too is, like, again, this, like, Claude code thing is a moment in time. Eventually, like, it's all just gonna be warm. It's gonna be Iron Man Jarvis, which it is Iron Man Jarvis. It's just a clunkier UX to, like, get the output. But, like, the output is, for anyone that is a fan of the MCU, Jarvis. **Chris** `[54:22]` Damn. He dropped the acronym. Yeah, dude. Jarvis is here. **Madison** `[54:26]` Jarvis is here. It's just a little just clunkier than being like, hey, Jarvis. Do this. Yeah. But it's we're, like, very close to that, to just we're just all gonna be, like, fucking mouthpiece talking to this thing, telling it what to do. **Chris** `[54:42]` Yeah. I feel like we need to tell David Specht about this in case he hasn't already. **Guest** `[54:47]` There's just no way he hasn't fucked with us yet. **Chris** `[54:50]` I don't know. Because it's like he's been doing all this studying, and I'm like, I don't know if you need to study that anymore. **Madison** `[54:56]` Yeah. I don't know. **Chris** `[54:56]` I think you need **Madison** `[54:57]` to learn **Chris** `[54:58]` how to use Claude. **Madison** `[54:59]` Well, he was using Claude to build all the shit that **Chris** `[55:02]` he's have a he doesn't have, like, a thing to apply it to now. So it's like **Madison** `[55:05]` doesn't have a thing to apply it to. That's that's, like, the the real, like, awesome thing about our situation is we've got this, like, great, perfect use case to **Chris** `[55:14]` because what what are you gonna do if you're at home? You're like, I'm gonna build an app that manage manages my groceries. It's like, yeah. **Madison** `[55:18]` You totally should. Yeah. Yeah. You totally should. **Chris** `[55:21]` I'm gonna test having it take a picture. I'm gonna take a picture of my fridge and see if it can identify what groceries I have and don't have without, like, a data because you could be like, eggs go here, this goes here, that goes here. **Guest** `[55:33]` That's the use case for Cloud Code eventually. It's like, can do that personally and build, like, apps at home. **Madison** `[55:40]` Yeah. Yeah. I I just think I think that I think it's about to get really weird. I think it's gonna get really weird, but I think really weird in a good way. I think that in the same way that people today are living exponentially more comfortable working in fucking air conditioned jobs, like, by and large. Right? Like, if you look at that progression of, like, how comfortable even people that, like, you know, are quote, unquote struggling in America. Like **Chris** `[56:09]` Poverty in America is, like, really nice in a lot **Madison** `[56:11]` of places. If you're fucking living in an apartment on unemployment, like, you're fucking crushing it. **Chris** `[56:17]` Yeah. Compared to humanity in general. Yeah. And the only reason conditioning and refrigeration. You're already above, like, 40 better than Kings did. **Madison** `[56:25]` Yes. Yeah. And and and the any any You can **Chris** `[56:30]` just play whatever song you want just right now? Okay. **Guest** `[56:33]` You have hot water and cold water? **Madison** `[56:37]` And I say that as somebody at in at at a certain points in my career was living in an apartment on unemployment and fucking in my own, like, personal hell. Right? But, like, the personal hell was in was in my head. I was actually fucking crushing it on a on a long enough timeline. **Chris** `[56:57]` That's the thing that that's where I'm, like, trying to figure out how to deal with what you think is gonna happen. Because the way I understand humans to work is they are they are they're not they're pattern making machines, but they're also **Madison** `[57:13]` It's a weird way to describe humans, but go ahead. **Chris** `[57:15]` They're they're also, like they're meaning makers. They're pattern makers, and they are they they dabble mostly in comparison and relativity, where it's like everyone **Madison** `[57:29]` has humans or yourself? **Chris** `[57:30]` Myself, but I I don't think I'm alone. Because I think most humans like, scarcity and wealth and brand like, all the bullshit, what you may call, like, old guard measurements of your, like, status and stuff. It's before there was money, before there was shit, there was other versions of that. There are cast systems that have been around forever. There people are always gonna try to figure out a way to put their, like, their level against other people. Even if there's an infinite amount of things, which I agree with kind of I I like that idea that, like, the material shit disappears. I don't think humans will ever stop trying to find their pecking order. Because that **Madison** `[58:12]` is like hard work. Shit. Some won't, but, like, more people will than ever before be able to. So it's like it's like now you're gonna it's gonna be easier to decide that you don't wanna participate in the, like, status hierarchy game. Yeah. It already is easier today **Chris** `[58:40]` I also I like that I like the idea. What I'm, like, coming away with from what you've said is, you know, we've talked about this before. Humans have essentially a higher power or god shaped hole. And we have crushed it so hard materially that, like, people have been trying to shove material things into that hole for a long time, and it acts as, a really incredible distraction from facing your actual problems, which is really just you. **Madison** `[59:06]` Mhmm. **Chris** `[59:06]` And once that hole can be filled infinitely, all of the stuff like the millionth cigarette, like, they lose their pleasure, and so they stop being a distraction tool. Yeah. They lose that utility. Yeah. So then all you have left is you and all the things that you know have failed you in terms of, like, distraction. **Madison** `[59:25]` Mhmm. **Chris** `[59:25]` And then what's left? **Madison** `[59:26]` There's nothing else to grab for. **Chris** `[59:28]` The end of, like, the Siddhartha journey or the wealth journey. You know? **Madison** `[59:30]` And it likely will get dark for a lot of people because, like, existential crisis is dangerous. Right? People commit suicide. People succumb to terrible addiction. Mhmm. That will happen, and that's gonna be terrible. Yep. Right? For those people. And so but it's like it's just it's also going to be an opportunity to, like, go the go towards the it's fucking Star Wars shit, dude. You're gonna go towards the darkness or to the light. It's all the same thing. It's all the there's only one thing happening here. There's only one thing ever being described in every any story, any religion. It's a it's just one thing. **Chris** `[60:15]` The quantum field. Fuck. Well, download Claude. **Madison** `[60:24]` Download Claude. **Chris** `[60:24]` Because chat has its own version now. It has codex, **Madison** `[60:27]` which apparently dude sucks. **Chris** `[60:28]` Apparently, codex smacked mythos in a test the other day. **Guest** `[60:32]` Apparently, chat GBT came out with some **Madison** `[60:35]` I'm sure they're all good. **Chris** `[60:36]` They scared. **Madison** `[60:37]` I'm sure they're all good. We're Claude guys. All I can tell you is there is nothing like I I don't I'm using Claude code and Claude co work constantly, and I haven't yet hit a point where I'm like, oh, man. I wish that this was better. **Chris** `[60:54]` Yeah. I haven't been like, you're dumb. **Madison** `[60:57]` Yeah. **Chris** `[60:58]` There is some there are some people that, like, they work somewhere where it only allows them **Madison** `[61:01]` to use Copilot or Gemini or whatever the fuck. So That sucks. But I would just use **Chris** `[61:05]` it personally. **Madison** `[61:06]` I would just build, like, what do you want? A fucking just build like, try to figure out how to build little web applications. And if you don't know anything about it, like, explain what you want and say, what's step one? And fucking figure out how to go get the API keys and copy them and paste them into the right place. **Chris** `[61:24]` You can literally, like, screenshot where you're stuck, and it'll be like, yeah. Click here. **Madison** `[61:28]` It'll walk you through how to do anything, and it's so empowering. Like, once and the other the other thing that we can wrap this up here. But the other thing that I've noticed over the last week and a half is not only have I been able to, like, accomplish a ton of stuff that I otherwise wouldn't have been able to accomplish, and a lot of it that is actually being done by Claude, I don't fully understand. It's also educated me **Chris** `[61:55]` Yes. **Madison** `[61:55]` Quite a bit in how all this how how, like, the the underlying infrastructure of a web application works where it's like my knowledge of how these things work together and the different platforms and, like, how to set them up is stronger, which as my understanding deepens, then my direction to Claude gets better. Yeah. And the reason I'm able to do what I'm able to do with it is because of my understanding of building web stuff in general. And so, like, that is it's not it's not like it's making me worse at it's it's not like a calculator making me worse at doing math. **Chris** `[62:29]` No. It's like I understand our supply chain better now Yeah. Today as a result of this grill me because it's like it shone it shined lights on spots that I was like, oh, I didn't **Madison** `[62:37]` Right. **Chris** `[62:37]` And it had suggestions that was like, well, you could try that. I was like, well, you have **Madison** `[62:40]` because it's literally that's what, like, Mike Mike was like, the big breakthrough for me was was when I realized that this thing is essentially an employee that I get to work with, and I can direct it to do the things that I would typically direct an employee to do. And so, like, when you're working with an employee, it's a it's a two way street and, like, you're collaborating. **Chris** `[63:06]` And they have information you don't. **Madison** `[63:08]` You're learning. They're learning. And, like, it's literally learning and storing memories so that it's better the next time. So you're, like, training this thing, but it's also training you and you're learning better how to work with it. And just like an employee where you learn somebody's fucking quirks and stuff and you learn how to work with them better. So it's like a it's like a a team member. Mhmm. And and I don't think that this means, like I mean, it's gonna mean we need more we just put a job posting for Yeah. We need more people. We need somebody in the kitchen. So we're gonna need more people in the kitchen, more people in the warehouse where it's gonna allow us to, like, more quickly expand product offering. We're working on beverage strategy right now. We're, like, actively working on it. So, like, it's gonna make us get that to market faster, which is gonna create **Chris** `[63:49]` I did probably two or three weeks of specking out in a day and a half. **Madison** `[63:52]` We're gonna need we're gonna need, like like like, salespeople because and then and there's all this, like like, yeah, it can it can do a lot of the sales, like, tedious stuff, but, like, you still need a human. **Chris** `[64:03]` Like Yeah. What my buddies are being told is, like, if your job is the task, you're in trouble. Yes. But if your job is, like, relationship management, synthesis, like, that sort of thing, like, that's important. **Madison** `[64:12]` And then we're also gonna need more people on this team that Wield the power. Wield the power of Claude to do more. Because, we could have 10 people simultaneously working on awesome projects that are making you know? It's like it's fucking crazy. Yeah. **Guest** `[64:26]` One hire is basically a 100. **Chris** `[64:28]` Yeah. Yeah. So it's like hire as many people as you can because they're all so powerful. **Madison** `[64:33]` If you have something good to direct it at. Right? Yeah. So yeah. Very interesting. Fucking crazy. Let's go Claude. Let's go Claude, **Chris** `[64:41]` dude. I'm go smack some tokens. **Madison** `[64:42]` Fucking smack some Claude. **Chris** `[64:45]` Alright. Finally, a pod where we had something to talk about. Yes. That was nice. We have some shit to say.