[00:00:05] John: Hello. You are listening to standing in the fire where Kris and I, the owners of fireside.fm talk about what it's like to be in the heat of SaaS growth. And today, this is like our quickest recording in between... The the duration between recordings is the shortest that it's been in quite a while. Mhmm. Like...

[00:00:26] Kris: We missed last week because you were out having some fun.

[00:00:29] John: I definitely have some of these segments. I've got some awesome material. Had some life changing experiences in Austin.

[00:00:39] Kris: That's awesome.

[00:00:40] John: Yeah.

[00:00:41] Kris: And we even talked about some of those, so I'll be excited to...

[00:00:43] John: Yeah. I know. This will literally... This won't even be fake. Like, it won't be me, like, saying it again. Also, I'm noticing the the What? Tire upgrade. Wonderful shorts there. Kris is now he's in his ever, you know, increasing quest to be more like me. He's now wearing the same kinds of shorts. They're so comfortable though.

[00:01:03] Kris: They're just easy. Yeah. They're so easy. They're so easy.

[00:01:06] John: Nice enough.

[00:01:07] Kris: Yeah.

[00:01:07] John: And they're really comfortable And it's it's just funny. Again, Kris and I haven't seen each other in a while, so now we're gonna start nerd out for a second. But, yeah. So these takes are gonna be totally new to each other and real. It'll be authentic, which is what people want in this AI era. So cool. Alright. Let's go into the fire. What has been easy, great, or hard and messy in the last week?

[00:01:30] Kris: It seems like those are different questions. Yeah. Having something easy or hard... What was the... Or great, you said?

[00:01:37] John: Yeah. It's it's mostly just, like, into the because like sometimes, you know, things go easy. Sometimes, they go hard. So, it can be either one or both, you know?

[00:01:44] Kris: I I think, you know, so Susan, Chad, and I did a panel last Friday. Okay. For some students. So you have about 40 students here.

[00:01:56] John: Notre Dame students or?

[00:01:57] Kris: Holy

[00:01:57] John: Cross.

[00:01:57] Kris: Holy Cross. Yeah. So they're in an entrepreneurship innovation class. And one of the... We've talked about lot of stuff. It was just interesting because like we haven't done that to... Together and... But we're all like visionary people. Yeah. And so no one was there like grounding us in like the questions when people would ask us certain questions. Yeah. But also just like the ups and downs. So, thinking to the what's good and what's been bad. Like, some days have been terrible and other days, it's like, oh man, like, we made so much progress. Yeah. On either getting more sponsors, you know, getting some more partnerships, getting more offices closed. And I feel like in the last week, we've had a lot of the positive. Yeah. Oh, good. We hosted a tailgate with Irish Angels last weekend, which was awesome. It was, hopefully, we're gonna get some more partnerships with them and just some other stuff that I think has been going really well for maybe two weeks ago. It felt like everything was, you know, crashing down.

[00:03:01] John: I I mean, that's I that's just doing stuff like this. Yeah. I hate the entrepreneurship word. I absolutely hate it. Yeah. But I'm like, that is that's what it is. Some sometimes it's the worst. Sometimes it's the best that but the I feel like the great part of it is that if you're the owner or the whatever the the CEO, the leader, like, you get you get to change it.

[00:03:24] Kris: Mhmm.

[00:03:24] John: So if it's really bad, it's your fault. And if it's really good, it's also your fault. So... Yeah. That's cool.

[00:03:30] Kris: What I've also sometimes found, and we've probably talked about it like a year ago is having then those celebration moments. Like, I feel like we're missing some of them.

[00:03:40] John: So

[00:03:41] Kris: I'm trying to create ways to do those because we need to in order to keep going versus just like passing them by and getting to the next But it's hard when you also then have to be the person leading that

[00:03:58] John: thing

[00:03:58] Kris: as well.

[00:03:58] John: Yep. But you did do... There was like a week

[00:04:02] Kris: ago,

[00:04:03] John: didn't Rides to You have like a Yep. I missed it because I was getting ready for Austin.

[00:04:06] Kris: Yeah. So, yeah. So we... One of our startup companies raised, like closed one of their rounds. And so we did a happy hour with them at the cocktail bar, which is again, perfect use case of why the cocktail bar exists. So so, yeah, absolutely. That was fun. And I... It made me think of more ways that we should in our space have almost like permanent celebration markers for the member companies. I don't know what that should look like. The idea that I had was, well, I didn't have it. I saw it from someone else. So I think it was the the Buffalo Bills who have a new football stadium. Oh, cool. And there's people season ticket holders. When you commit, I think it's them. I'm not sure if it was different, but I think it was the them who they create these little engraved key chains you put on a wall so they can see all the other people who have... Are seasoned ticket holders.

[00:04:58] John: And

[00:04:59] Kris: it's probably there for... I don't know if it's there forever or not. Yeah. But I think something like that where we could engrave a thing and put it on a wall so that everyone can see it, I think is a cool thing... Idea. I just don't know how to yet, like, execute that here.

[00:05:14] John: Yeah. I think the biggest problem is collections. It's just how do you actually know about the things that happen and that they're significant and get them? It's super easy to give them a treatment, you know, on a wall or on a plaque or on a

[00:05:28] Kris: tag or whatever. Well, and I even wondered if we just... If there's if there's something public, do we give them something? Because I would say on a on a similar note, we had a member that got a really big client, and so they're upgrading to an office from coworking membership. That's awesome. On month, like the on, you know, October 1, and hopefully this comes out not too, like, not too far away from that, but like, that's the thing that we should we should signify somehow. Now, whether it's on a wall or giving her a little thing that they should put on the desk, like, I think I I think that can create more community. I'm just not sure yet of what the execution of it is.

[00:06:03] John: Yeah. It could be like football helmets and touchdowns, then they get Oh, yeah. Stickers. Yep.

[00:06:08] Kris: Yep.

[00:06:08] John: Like, you could have that, like, there's some kind of... Not a sticker, but, like, some kind of thing like that to mark on, like, the... You know, on their their name tag or... I don't know. Something like that can be cool too. Mean, I like the idea of a shared space where someone can go and see all the things for all the people. But I also like the idea of, like, walking around and seeing it all as well.

[00:06:25] Kris: I I think it could be both. Because I'm also thinking, like, I don't want you to be able to just get it once Yeah. Because so and so how do you make it so that a company could put a thing on there multiple times, still tributes to them? Yeah. But yeah, I don't know. But anyway, it's, it's cool stuff for the, the member companies that are working here. So that's been, you know, motivating and exciting.

[00:06:45] John: Yeah, that's cool. I mean, Into the Fire for me is

[00:06:48] Kris: like,

[00:06:48] John: it's, it's kind of weird because I was like gone in Austin last week. So like, I mean, I think like high level summary of like what I saw. So it's, it was called rails world, which is the most difficult thing to say in a row because you always just say, Rails

[00:07:02] Kris: World. Oh, I was

[00:07:03] John: gonna say,

[00:07:04] Kris: Whales World.

[00:07:05] John: Yeah.

[00:07:05] Kris: Whales World.

[00:07:06] John: It it comes out funky every time. So now I just have to slow down to, like, half speed to get it correct. But the thing that was interesting is, like, this was the first conference that I've been to in the AI era. Mhmm. And just a a range of emotions and vibes because of that. So that was really fascinating. It was the biggest Rails conference ever over a thousand people Wow. Which is cool. Yeah. And DJ, creator did like a keynote where he basically was like, y'all need to stop writing and looking at code and Rails doesn't matter, which was like totally accurate. It didn't really bother me, but some people got pretty up in arms about it. So that was kind of interesting. But I think like the first day was like this euphoric high, like of getting into Austin, seeing all my friends again that are, you know, my Internet friends Yep. And hanging out with them. We did Terry Black's barbecue, which was incredible.

[00:08:01] Kris: It

[00:08:01] John: was, like... It's... My best description is McDonald's for barbecue, but, like, amazing. Right. Because you just... It's instant. It's Yep. It's instant barbecue. You just walk up and receive barbecue and sit down. And I was like

[00:08:14] Kris: Did you... It was there a long line?

[00:08:15] John: No.

[00:08:15] Kris: What? Oh, time I was there, I was standing outside and it took like twenty

[00:08:20] John: why is because Jesper was like, hey, I get in like fifteen minutes after you and I'm gonna get barbecue. Do you wanna go? And I was like, yes. So... But by the time we got to the hotel and I got an upgrade, so I had to wait fifteen minutes on my room. I got for free.

[00:08:33] Kris: I had to wait fifteen minutes for my upgrade. Yeah.

[00:08:36] John: Had had to wait with the peasants. But, like... So I I got the upgrade, so I had to wait. And then by the time we got that and stuff, and we talked to people in the lobby and then actually headed out, it was, like, 04:00. So we we kinda intentionally didn't eat all day in preparation so we could just go straight to barbecue. And then it ended up being like, that was perfect because then I didn't need like supper. I was good for the rest of the day. So we timed it very well. We just walked straight into the front of the line and I was like... And they were just like, what meat do you want? And I was like, all of the meat. You know? It was good. But then, you know, so, like, that happens. We we took a robo taxi

[00:09:14] Kris: there,

[00:09:15] John: which was like... That was a wild experience to have no one in the driver's seat.

[00:09:18] Kris: Yeah. I I canceled the one that I was there, and they tried to give me one because was... Yeah. I just No.

[00:09:23] John: I I was in a rush too. I was in a rush. Yeah. If you're in a rush, that's difficult. It does take longer. Yeah. A little bit. Little bit. Just depends on how aggressive your cab driver is or Uber driver. Yeah. But that that was kinda cool. Like, that was super cool. We were nerding out. There's four of us in it, and we're all just taking pictures. We're putting YouTube on and all that kind of stuff. So that was really cool. And then... But, you know, you still like the steering wheel's moving. It's a normal car. And then we went to Terry Blacks and when we're at Terry Blacks, we're like, okay, we're taking Cybercaps home. Like, we're going all in, you know? So we each get one because it's only two seater instead of four. And so it just pulls up and, like, they have a light on the front that, like, changes color if there's two of them so you know which one's yours. The app, like, has an arrow at the top and you just, like, move it around and it points you right to your cyber cab. It's it's really cool experience. And so we did that. We rode that back. And that was like, I legit... When I was in the cyber cab, there was a moment where I all... I'm not... I almost shed a tear because I was just like, this... I didn't... I don't think I... In that moment, I was aware of how different things are gonna be in like five or ten years. I like saw the future and I'm like, my son's 11 in five years. Is he even gonna need a driver's license? I don't... Yeah. Probably not. Like, and it is... It's crazy to think that, like, basically a remote control car driver is gonna be like a, like a position that people will have. It will be a career,

[00:10:55] Kris: You know? Like, I Although, I would say I would equate it a little bit maybe to, like, airplane Yeah. Pilots? No. Not the pilots. The people in the towers.

[00:11:11] John: Oh, flight control.

[00:11:12] Kris: Air traffic controllers. Yeah. Yeah. So very stressful job. Yeah. Wonder if it would end up being similar to that where one person's doing a bunch of different things and coordinating it. Yeah. I don't know.

[00:11:23] John: I think it seems very reasonable. I mean, they are of concurrent, like, locked when they're, like... I think when they talk to you and stuff, it's gonna be... They can't talk to two people at once.

[00:11:32] Kris: Okay.

[00:11:32] John: So there would be some, like, moment where they're just... And then after that, they kinda, like, get you unstuck and move on to the next on... Unsticking, you know? But like that was... I don't know. I was just like, I was all of sudden, like, very

[00:11:43] Kris: aware

[00:11:44] John: Mhmm. Of, you know, the first boat length Cadillac that I drove, you know, when I was 16, the old ugly eighties, you know, like to being in this tiny little car that's just driving myself and my friend around and we're just talking and messing around and like completely unaware of anything else going on outside. And I was just like this is wild. Yeah so that was like I

[00:12:13] Kris: think

[00:12:14] John: for me you know what's been easy great hard messy like it's just going to the conference and then you know all that stuff was great on Tuesday and Wednesday the keynote was like fine. I'm like whatever. And then after the keynote, everyone that I talked to that day was just like consulting has went to zero, especially in small to medium sized businesses because they can just get a cloud subscription and that's effectively a consultant to them. It's it's they can't tell the difference between that and a high quality consultant. So like that work has dried up several people. It's dried up so much that several people I know have taken jobs.

[00:12:51] Kris: Wow.

[00:12:52] John: But there's also

[00:12:53] Kris: not

[00:12:53] John: very many jobs. You have to know people, which makes conversations like that more important.

[00:12:57] Kris: Yeah.

[00:12:58] John: It was just like... And then everyone that I talked to, their app has either gone down in revenue a little bit or significantly. The only people that are basically up are, like, developer tools. Like, everyone I talked to who, like like, Honey Badger, you know, up a little bit. I'm not gonna spoil their news or whatever. But, like, you know, anyone that's kind of tracking the like get software into production is doing okay and everybody else like it's it was real hit or miss. Whether they were being gutted. Right. Down a little bit or doing well. Which is interesting to see, I guess.

[00:13:35] Kris: Yeah.

[00:13:37] John: And again, could just be like totally who you talk to.

[00:13:39] Kris: Right. Right.

[00:13:40] John: But that was kinda, I think, the hard part. The easy part was like cyber cab, all this kind of stuff. The hard part was, like, people that are, like... That I like that are legitimate, good people, good businesses before. Businesses that, like, I even thought before, like like Kris Oliver from

[00:13:56] Kris: GoRails.

[00:13:57] John: Mhmm. You know, they're... They had... He had, like, videos. So screencasts, things like that. And people just pay monthly and they just keep releasing videos and like it it's, and he's talked about it publicly, but it's just getting gutted. And I mean, a couple of years ago, was like, that would be an awesome business to own because then you have

[00:14:13] Kris: distribution

[00:14:13] John: and then you just upsell developer tools. And so it's literally just like, this is almost like cost of doing business. And then these are the backend products you make money on. And now it's like, no, that's not, it doesn't work that way anymore with AI. No one, no one needs the video. So I don't know. There's just a lot of weird things

[00:14:29] Kris: like that.

[00:14:30] John: So that was kind of like, I think the hard part, but then the next day I talked to people I liked.

[00:14:36] Kris: Nice. It picked back up. Yeah.

[00:14:37] John: All right. So you, you beforehand, you were like, oh, I'm thinking of some hot takes. So that's where we are. We need sparks fly. What what's... Do you have any hot takes?

[00:14:45] Kris: Well, I I I need some. I I probably do because I I feel like the last week and a half, I've been very what do I want to say? Emotional is the wrong word with a different Not emotional. But very, like... Articulate is probably not the right word, but excited, but also in a sometimes negative way where it's, like, not annoyed, but, like, frustrated. And... Yeah. I I I don't know. I I I... Let me think for a second. Do you do you have any that come to mind?

[00:15:24] John: I mean, my hot take is... I don't even know if this is a hot take, but even if you hate

[00:15:29] Kris: Elon,

[00:15:29] John: you should go fly to Austin right now, get some Terry Blacks, and ride a cyber cap. It's... I just... Literally the whole conference, that's all I could think about. Like, Jesper and I just were like cyber cab to the conference, cyber cab. Home from the conference, to the conference, home, to the conference, home. Like...

[00:15:47] Kris: Was the pricing similar or was it...

[00:15:49] John: It was honestly, it was pretty cheap. It was like we were going... We weren't going far, but it was taking, you know, so less than a mile. Okay. Typically between the conference and hotel. It would take anywhere from twenty to thirty five minutes

[00:16:02] Kris: because the traffic was

[00:16:04] John: really bad. So the way there was like fifteen, twenty minutes away back was really bad because it was always four or five. And I would say it was like $10.11 bucks.

[00:16:12] Kris: Oh, yeah.

[00:16:14] John: Super cheap. Mean, you're like

[00:16:15] Kris: because you're also, guess they're not paying a person to see the car. Did...

[00:16:19] John: Hilariously, though, they do have a tip. And so you can tip if you want.

[00:16:23] Kris: Who are tipping?

[00:16:24] John: Exactly. If you hit tip, they say, just kidding. We don't want your money. Like, that was a nice little, like, surprise and delight. Yeah. Because Jesper, like, I I I said several times to people, I was like, if you put Jesper in one of those cybercabs and you put my son in the cybercab and you tell my son he can press any button he wants, Jesper would touch more buttons. Like, he was just touching every single button that existed and it was very hilarious to see him like, well, what do you think happens if we hit the door button while we're going down the road? Yeah. And I'm like, alright, let's do it, you know? And it stopped. It was like, close the door, please. You know, so we closed the door. Like, we weren't going very fast. Yeah. We were just kinda creeping like five miles an hour. Yeah. In traffic. So, it wasn't a big deal. I imagine they, over a certain speed, they just won't let you open it. Yeah. That's what I would expect. Like the... Yeah. The the windows only came down a little bit. So I guess Interesting. I guess my hot take would be like, definitely go do that. But also, like, now I feel like I'm... With AI and that, it's like I'm... I feel like I'm living in the future.

[00:17:24] Kris: Mhmm.

[00:17:24] John: And I just think about so many people who are like, never gonna let anyone drive me around. And like, you know, like, I don't trust robots and stuff. And I'm like, the robots are so much more accurate and better drivers than you are. Yeah. Like, and to the point that, you know, I was just talking to Paul in the hallway and he was like, yeah, one time I took a Waymo and I was in a hurry and I was like, this was a mistake. Because the Waymo Correct. And actually the Waymo is what I was talking about that was taking longer in Austin when I was there last. Was like, it was taking for the the time it was projecting for me to get to the location was way longer

[00:17:56] Kris: than the time for Yeah. A a Yeah.

[00:17:59] John: And I think a lot of that is just because it is more careful. Yeah. But I'm like, you know, again, if you cut out whatever 80 or 800,000 or whatever the number of deaths are in a year, Like, and, you know, what's it called? DUI. Like, there there'd be no DUI. Like, someone could just be passed out in the car. Yeah. And it's safe. Nobody else is at risk. Like, I just, it's gonna it's gonna happen when people kind of start to get over it. So, I guess my hot take is just like, how many years? Yeah. Until this is like... I think there will be some real, you know, until price gets... Keeps getting collapsed, people are gonna pick and there'll be some small percentage, 5% or whatever that'll pick, like, you know, combustion engine just because they're like, whatever. I don't trust the man or Yeah. Something like that. But I think I think in two to five years, it's gonna be... It'll it'll be in South Bend. It'll be everywhere. And like... And they're just so much cheaper to make. Like the... I think the cyber cap is like under 30 k.

[00:19:00] Kris: Wow.

[00:19:01] John: And the the Waymo is a Jaguar with a 100 and, like, 50,000 plus in extra accessories on it. And I'm like, how... I don't see how that can compete.

[00:19:12] Kris: Yeah. Yeah.

[00:19:13] John: But, like, little things, like, you would... You... You're driving, you know, and there's all these parked cars. And, like, as a human driver, you would leave, like, a foot or two gap. Yep. So you'd almost be over the line a little bit to try and make space. And the Waymo is just a lying, you know, three inch gap between the mirrors. Because it's like, I'm a robot. I know exactly how far away I am. I have no hesitation. I'm not a driver who's on the opposite side of the road who's looking over there while they're trying to pay attention to the front. And so it would just like whip. I was like, wow. Okay. That's interesting.

[00:19:44] Kris: So, yeah.

[00:19:45] John: So that's my hot take. And maybe that's not that hot, but like two to five years, like I'll be the first in line if I can get a car that will just drive me. I don't care who makes it.

[00:19:54] Kris: Yeah, that's cool. So the two, one we've talked about a little bit, but it's crept up again and it's still hot is like using AI for creative digital art.

[00:20:09] John: Yeah. Is this about my other podcast? You just wait until we're on air to air it. I see how it is. No. No. I'm just kidding.

[00:20:20] Kris: No. No. I I was actually I I wasn't thinking of that. I know. I know. Was just kidding. But Yeah. It's it's... You know, we we got some additional art for the screen print shop, like, the other day.

[00:20:29] John: Yeah.

[00:20:30] Kris: And they are like, I wanna see what you can do with it. I I would say more than one time in the last, like, week. But Yeah. And I'm just... It's it's it's a flat file that has way too many gradients, has way too much stuff on it. Yeah. It's like... It it's so cumbersome to make it into a file that is, like, usable. Yeah. And I don't know. It it... It's just... Yeah. I don't know. It's super frustrating and and, like, to educate people, it's like, but do you want the business or not? Like, as as the as the screen print, you know, business? And we do. And so it's like, it takes way longer to get that sale closed and to production than someone who's coming with, like, a vector artwork or even a PNG, you know, that that we can ease... That they made, but they maybe don't have the original file. Getting that to production is so much quicker. Yeah. One thing we probably need to start doing is charging hourly artwork. Yeah. For if people will come with that because it does take way longer. Yep. And it's typically like a worse design.

[00:21:32] John: Yeah.

[00:21:33] Kris: Yep. It's it's... Yeah.

[00:21:34] John: It's the same thing in... For box out. Like, are using AI to make graphics like that, and they're horrible. Mhmm. Like, it's... And you can just... You you can tell they all have the same smell and look.

[00:21:44] Kris: Yeah. Have you... Don't... When you scroll on Facebook, like, anyone who's posting an event that I I just... I'm like, I just wanna start... I don't have, like, unfollowing people because Yeah. It's just annoying to see. Because there's there's no thought that went into the structure or the information. Yeah.

[00:22:04] John: Yeah. I mean, typically, what what we've noticed from the sports side is just way too much information. The graphics are way too busy.

[00:22:11] Kris: Yep. Yep.

[00:22:11] John: Way too many things going on. And it's just like, if you took, like, eight things out of it, it'd be an okay graphic. Mhmm. Like, stylistically, it's not bad. It's just there's... People generally don't understand the idea of contrast

[00:22:28] Kris: White space.

[00:22:29] John: And white space, and like three things like that.

[00:22:32] Kris: Two fonts. Yep. Like, and I and I I say that from, like, I'm not against using it to help do something. So using it to help remove a background. Yeah. Totally great to make it way quicker than having to trace around, you know, an object which you used to have to do. But when it's just creating art that you can't do the next step on efficiently, I think it just becomes super frustrating. Yeah. I totally agree. I can see it.

[00:23:04] John: Simultaneously, there's like these incredible thing, like the videos. Yeah. The videos are just like Cloud Opus 5.5 came out. You can literally just with a prompt be like, here's my website. Make a thirty second intro video for me. And the end result tends to look about the same. I mean, it'll use your styles, but it's the same animations, effects, and and things that it tries to do. So I think, you know, the difference between that just going out and, like, something really good that goes out is probably, you know, a little bit more taste of being like, we don't need to flurry world, all these things around on every single and slow it down because I can't even read it, you know, like, but you can do this kind of stuff now. And, you know, I was even talking with Yoder yesterday and he was like, oh yeah, like I actually used it to ship like a video for a client and they loved it, you know?

[00:24:00] Kris: So I take it on the opposite side of that. Yeah. Not, not is I had a designer. So I have worked with designers, like contract designers

[00:24:10] John: and

[00:24:10] Kris: they used it on something and didn't tell me, it went to the client and then it got produced and it was not good because I didn't know, didn't catch it. And then people who were seeing on the other end of the client

[00:24:24] John: saw

[00:24:25] Kris: it and it's caused a lot of issues. So I think there's, I think there's as a helpful tool, 100%, but like when, I don't know, it's, it's, it can just also be negative on client work. I've had the experience on the negative side, unfortunately. But I think I maybe have had positive too. It's just the negative ones, you know, stick with you more because they're like, ugh. Like, I I wish I would have caught it. I wish the designer didn't do it. I should have told me that he did it so that I

[00:24:55] John: Or can disclosed. Yeah. If they just disclosed this is what I did. Then I could know and be like, we can't do Like,

[00:25:00] Kris: let's let's change it back to what it was supposed to be.

[00:25:03] John: Yeah. I think disclosure is gonna be like... I mean, Claude's even doing this weird watermark thing now.

[00:25:08] Kris: Yeah. I heard... I saw that.

[00:25:09] John: I haven't really followed it that much, but I think disclosure is gonna be a big thing so that you can know, is it AI? Is it not? That way you know what you're getting in. Like, I think... And the thing I told Yoder yesterday when we were talking is I was like, I think it makes total sense to, like, use Claude and GPT for stuff like that. But what you wanna do is, like, what what needs to happen is you can't then just do your same old pricing. I don't think that's fair to the client.

[00:25:32] Kris: Mhmm.

[00:25:32] John: Because now, like, you made that in, like, zero time.

[00:25:35] Kris: Now it's

[00:25:36] John: also not worth... Like, if it takes you five minutes

[00:25:38] Kris: Oh, I can hide video.

[00:25:40] John: Yeah.

[00:25:40] Kris: I got a new one.

[00:25:41] John: My thing is if if you used to charge, whatever, a thousand dollars for this Yeah. And AI made it, and it would take you two hours, and now you can do it in five minutes.

[00:25:50] Kris: Yep.

[00:25:50] John: Well, yeah, that doesn't mean you charge $5 Right. Or whatever. But there's some middle ground where it's like, I have experience and taste, and I have applied that to this video that AI has made for me. And it was way easier for me to make it. So there's some point in the middle on that that makes sense from, like, a pricing standpoint.

[00:26:11] Kris: Yeah. Yeah. Oh, so I don't think we've talked about this yet here, but using Claude or whatever you wanna... You pick your system to input a bunch of information and then send a, create a report and send it and not check it before you send it. You want other people to review it. I think that is, I've experienced that in probably the last six months, multiple times. And as soon as I get to the, like, there's one error, I'll keep going. But once I see two errors, I'm like, I'm done reading this because it it obviously the person who compiled it didn't fact check it. Mhmm. And, but they want me to spend time reading it word for word because I want to. I'm like, oh, this is a thing that I should evaluate and get feedback on. I'm a very nice person, and I like people to like me. So it's hard to have a straight rule that respond to people with like, if you haven't read this, like, I'm not gonna read it until the errors are corrected, but I want to create something like that because it's just happening too often. Or to then pick out the five points you want me to specifically look at, because then I know you at least went back and like, here's the things of the 15 pages when ordinarily that report would be like two pages. Yes. And I forget who someone said this to me recently too, a different context, but you know, if I had more time, I would have written a shorter letter. Yeah. Like, I feel like that's what

[00:27:41] John: Mark

[00:27:41] Kris: Twain. Yeah. I feel like that's what some people who are using AI to create these huge reports need to do is spend more time, make it smaller, cut the fat.

[00:27:51] John: It's not some people it's it's everybody. So it's everybody before who like maybe didn't do their job or barely did their job.

[00:27:57] Kris: And

[00:27:58] John: now they're like, oh, I can make it look like I'm really doing my job.

[00:28:01] Kris: Yeah.

[00:28:01] John: But actually now your job is different. Your job is to make sure that things are accurate. And like, always think of, you know, my good friend, passed away now, Jason Rudolph. His whole thing was I need to formulate my communication to make it as easy for you to digest as possible.

[00:28:19] Kris: I agree.

[00:28:20] John: Whether you're using AI or

[00:28:21] Kris: not,

[00:28:21] John: is your goal.

[00:28:22] Kris: Because I would even say I was doing some packaging, package pricing for potential customer, and I went back through a math check. So, like, I I was adding up a couple different things to create package APC, and I think one of them was off by like $20. And I was like, oh, no. Like, it was it was correct. Like, I put in the right inputs. Yeah. But if I were to send that to someone, that would have been embarrassing. And I don't know that everyone goes back and

[00:28:48] John: checks all that. It goes back to the very first time we talked about AI on this, is verification. The number one thing, the number one skill right now is verification. So it's like if you come up with better ways and ideas to verify work, you get more done. And if you don't, you send slop out and people will get frustrated like that. So that's a good hot take.

[00:29:07] Kris: I don't know if we have a segment on it, but we can do our update on how I've been using Codecs the last week. Is there a segment?

[00:29:15] John: Let's do Embers. Small updates, little wins. Yeah, logs on the fire. Here we go.

[00:29:21] Kris: We this is maybe two weeks ago, right? We set up or we didn't even set up to my computer. You sent me a text that told me how to set up.

[00:29:30] John: Well, what happened is you were like, hey. I wanna I wanna get... Was like, you should use Codex. You can do it. And you were like, cool. Help me set it up. And I was like, alright. Let's meet. Then we couldn't meet. Then we couldn't meet again. And so I was like... And I could tell you were waiting. And I was like, Oh, he doesn't have to wait. Like, he can just... Here's your prompt.

[00:29:47] Kris: Do this and try it. And you were like, Okay, I'll give it a shot. Because historically... Yeah. As we talked about last time, historically, it took eight hours. Yep. But it's like failing a 100 times before he gets to work.

[00:29:56] John: But now it's not.

[00:29:57] Kris: So started with just doing some revisions on some public facing pages for speaker deck. I would say that went pretty well because we don't have a ton of branding. Like it's just basic, like styles that we're using. And then, then I started to do some other, like, reviews of some things I knew were frustrating me. Yeah. Just in the UI in the app, and then trying to do a little feature stuff. Yeah. And we put some stuff out there. I think it... I I tried to have either you or Eric, like, look at it before I did, and just because I'm not sure all the things it's touching. And I I try to have it do reviews multiple times. Because I know you've mentioned that. Yeah. So before... I I don't take the first one. I have to go back and then do the different G Stack reviews. Yep. Which what's interesting is, like, I'm always... Because you told me that from, like, day two. Yeah. I'm always doing that. I'm always doing, like, office hours to start. I'm then doing a review. Sometimes I'll do an engineering review. I'll only do a CSO if I feel like there's more, like, password protection stuff. Yeah. Sometimes it will just say it doesn't need it. Yep. But I know a lot of people who have been using it to code and don't even know what that is yet. And like, oh, I don't even understand it. It's too risky. Why not have these reviews happening? We do... I've pushed out some small features. Nothing that's super, like, database heavy of changing that because I know there's some things, like, tagging and stuff we wanna do.

[00:31:30] John: Yeah.

[00:31:32] Kris: So we did that on speaker deck. Also then went back because our screen print shop website, I built before we acquired it. Yeah. And I hadn't really touched it. I was like, oh, let me see if I can get it to do the WordPress site.

[00:31:42] John: Oh, yeah.

[00:31:43] Kris: And so that was actually pretty easy to get it to do it and revamped a number of pages, updated some technical, like SEO stuff. Okay. Did then a, We, we, for the last, like six months, we've been able to do like banner printing for like roll up banners, you know, vinyl banners. Not one person has ever submitted our inquiry form for that page. Yeah. Because I I it was a terrible like the page was pretty bad. So, I I revised it and the next week, we had someone submit a quote.

[00:32:14] John: That's awesome.

[00:32:14] Kris: And it was the one that went through. So, it wasn't even just someone that was like like a bot or something to get a lot of forms for that, but it was someone who needed some banners and quick turnaround. We did it, and I was like, oh, dang. Like, it's Site is looking good, man.

[00:32:28] John: Yeah. Yeah. I'm click... I'm clicking around I

[00:32:30] Kris: fixed a a bunch of just, like, responsive stuff, fixed some accessibility stuff that was always bothering me, but I didn't want to go back to my developers because I don't want them to charge me to So fix I knew some of the accessibility and some of the mobile issues. And then I went through some of our forms to I wanted to reduce, and I know we can do another phase of it, like, reduce the back and forth after someone submits an initial form. The forms I created two years ago haven't checked them. So, like, the information we're gathering now is a little bit more specific so that we can hopefully get them a little further of getting them some sort of estimate before we need to, like, ask them a bunch of questions. Yeah, it's been, I think, I don't know if there was another site I tried to use it on. I tried to use on the momentum site, but we're on Squarespace and it's, it, it needs, it will do a like click in it will, it can adjust content, but then needs to go in like into square space to do it. I did a one page because we're using a landing page for some ads.

[00:33:32] John: Do they not have an MCP yet?

[00:33:34] Kris: I don't know. I didn't, I didn't go that far to look actually.

[00:33:36] John: So that's what I would do is just anytime you go to do stuff like that, I would just search, like, Squarespace MCP. Because if they... They do. They offer, MCP for AI assistance. It's scoped very narrow.

[00:33:48] Kris: Okay.

[00:33:48] John: So it it doesn't help with managing or editing the site. They're... I'm telling you right now, Squarespace is dead in the water if they don't do this. Yeah. That's the biggest thing I'm... I would even, at

[00:33:57] Kris: this point, move it to WordPress because I'm like

[00:34:00] John: Yeah.

[00:34:00] Kris: At least then I can have all this stuff.

[00:34:02] John: Honestly, you... At this point, you can move it to... If you don't ever need the client to update it, you can just move it to flat files. The fastest thing now is flat files, which is insane.

[00:34:11] Kris: But, like

[00:34:12] John: Mhmm. To actually work on. Not Right. You know, it used to be that was,

[00:34:15] Kris: like The slowest thing. I mean, dream dream weaver. Yeah. Yeah. Exactly. You ever

[00:34:21] John: use that? Oh, a 100%. That was what I learned on. That was that was my first editor was Dreamweaver. Yeah. So I've definitely went down that route. So for the nontechnical people who listen, what was the trick for making it work with WordPress? Like, what are the things that you did? Do you remember?

[00:34:35] Kris: So I use SiteGround for hosting. If anyone wants a referral URL, I got plenty of them. But so that I use for all the clients. It was really just creating SSH key and then putting that somewhere on the computer and

[00:34:54] John: telling

[00:34:55] Kris: Codex where it is. And after that... So I don't have staging site on...

[00:35:00] John: Okay.

[00:35:01] Kris: But it wanted us to have one. I was like, is... Yeah. If it was a client thing, I'd probably would say, yeah, create this staging site because SiteGround can do it. In this case, was like, just do it live. But it was really just getting that connection. And then, again, I would say, I think part of the why I 'm not saying I'm great at using the tool, but I think why I can be more successful quicker is because doing the consulting work, I was dealing with designers and contractors that were outsourced. So I had to be very specific, I learned how to be very specific on what I wanted to have happen. And so to creating documentation on that before just saying, create me a website. And so I can be very specific on here's the scope I want, here's what I want done, and it will come back. And there might be back and forth, but at least, like, don't get, I do get frustrated sometimes, but it's, I I have a skill to create good documentation for what we're trying to do. And so I feel like it's the same as giving it to some of the outsourced designers. Yeah. Or like developers, I should say. And yeah, it's... So Matt's who created

[00:36:04] John: Ruby.

[00:36:04] Kris: Yeah.

[00:36:04] John: He was up on stage for like a fireside chat kind And of he he was like... He's been, you know, vibe coding a bunch. And he was like but really, he's like, I've been vibe coding for fifteen years. And everyone starts laughing, you know? And he's like, no. I mean it because the way Ruby core worked was basically he would have ideas. He would give the ideas to, like, the implementers. Mhmm. They would implement them. They would bring back, and he'd like, what about this and that? And they would go back and fix it. Yeah. So, like, he was doing it with humans.

[00:36:34] Kris: Right. Agree. Me too. Yeah. Yep.

[00:36:36] John: So now doing it with computers, it's like, oh, it's even better. Yeah. Yeah. Of course, I I like to do this and I'm skilled at it.

[00:36:42] Kris: Right.

[00:36:42] John: Because I've been doing this for fifteen years, you know?

[00:36:45] Kris: Yeah.

[00:36:45] John: I thought that was really, like a pretty deep insight actually, because I'm the opposite. I I did all

[00:36:52] Kris: the You were doing all the other computer. People who are doing project management type work are probably be able to use a tool quicker Yeah. Efficiently, but just from the sense of that's the whole thing is managing different people and inputs to get the right output is what Yeah. I think whole job.

[00:37:11] John: The tricky part is like, they have always seen software as a black box. So they're like, well, here's my inputs, and here's my outputs. Mhmm. So, yep, you made it work. It's good. Mhmm. Which totally makes sense. But I think there are limits to that right now with the current models. At some point, those limits will go away. Yeah. And it will be so good that somehow it'll infer all the context you don't even have. You haven't given it, you know? But like, for example, one of my RV buddies, Kyle, he... Kyle Chowning, he's like in, let's say Tennessee or something like that. And so he runs like a dance studio and a marketing firm and a bunch of different things. And he, last night he was like, Hey, how long should CI take? Like for your apps, like how long does it take? And I was like, well, let me look. So I was like, well, I know BoxOut's the slow. So I looked at box out and box out typically finishes within like eight minutes. Okay. So I was like, you know, and I know some of my, my faster ones are like two to three minutes. So I was like, typical range is like three to eight minutes. And he was like, oh no. And I was like, what? And he's like, well, it's taken sixty minutes right now for... Because he's been working on a piece of software for, like, a year with

[00:38:16] Kris: Claude.

[00:38:17] John: Yeah. And then now because of me, he's starting to use Yeah. GBT a little bit. But, like, he was like, that seems really bad. Right? And I was like, oh, yeah. Like, that was like, well, here's the thing. There's a ton of ways to fix that. So for him, you know, it's very much a black box. He doesn't know what's good, what's bad.

[00:38:29] Kris: He

[00:38:30] John: sets a thing. It does a thing. For what... Everything he knows to check, it works. And so he's like, cool. Let's ship it. And he's like... So... Which is awesome. I think the fact that he's empowered to do that is

[00:38:40] Kris: incredible

[00:38:41] John: to, like, make his business more efficient and better. But the downside is, like, this kind of stuff happens. And then there's real engineering that has to... That you you don't even know what to say. Correct. Maybe you could just say, you know, and you just haven't thought of it yet. Like, how do I make my CI faster?

[00:38:57] Kris: Yeah.

[00:38:57] John: You know? Yeah. But maybe Claude will figure it out. Maybe it won't. It depends on, you know, a lot of things. He had so many minutes burning that, like, he went through, like, fifty thousand minutes on GitHub actions. That was, like, his budgeted amount. And now he... This month, he's up to 200,000, which is, a grand

[00:39:15] Kris: Yeah.

[00:39:16] John: In CI bills. And I was like, oh, dude. You can just

[00:39:18] Kris: Yeah. I I think that's the thing. Like, when people don't... Yeah. When people don't know what what the... Yeah.

[00:39:24] John: And for the the editor, you're gonna wanna take that stomach girl out or leave it in. Whatever. But, like, what was... That was mine also. Don't wanna embarrass Kris and make everyone think it was his stomach growling. I evidently didn't eat enough breakfast. But, like, what I what I think is awesome is I love that he can do it. I love that he's working on it, and he keeps pushing forward. And I love that even though he doesn't know, he's not... Like, the... Again, the biggest difference between, I think, people who accelerate and people who don't is being curious Yeah. And asking questions. And not being afraid to be dumb. Like all the time I'm like, wait, how do you do that? What does that mean? Like, just just naturally... I wanna know and I don't care if it looks stupid if I don't know what NGMI means or whatever, you know? And so I I think that's kind of the thing that, like, you know, their project managers, whatever, they have their inputs and outputs, but they don't know all the outputs to check. Right. And so... And the models are not good enough yet to... For you to not need some... Right. At some point, some... You know? So he's like, yeah. He's like, well, I've got, 2,800,000 lines of code. Like, was that normal for a project? And I was like, you know what? I don't know. So I... What? What do think? I just think that's... It's it's

[00:40:36] Kris: just interesting. Like, don't know what they answer either, but it's just... I I I wouldn't even think to ask that question, let alone, like... Yeah.

[00:40:43] John: Well, I was like, okay.

[00:40:45] Kris: But but but not in the coder would, like... Yeah.

[00:40:47] John: I was like, let's let's just look. Yeah. So I went into Codex, and I was like, of these eight apps, you know, Flipper, BoxOut, etcetera, etcetera, how many lines of code are they? And it was like, okay. BoxOut's 330,000. Fireside's 330 ish thousand. Flipper is 73,000, which Flipper is... Has historically been hand curated basically by me. Yeah. So it is succinct.

[00:41:12] Kris: Right.

[00:41:12] John: You know? Right. It's... It has vibes in it, but Yeah. It doesn't have as many vibes. You know, Box has five people working on it. Fireside has a few people working And on so so it makes sense that those are bigger. But I was like, oh, that's actually really interesting that they're like... So first off, they're not... Box Out is 12, 13, 14 years old

[00:41:30] Kris: Right.

[00:41:30] John: And it doesn't even have a million lines of code. And his app is like a year old and it's got like 3,000,000. And I was like, okay, that's fascinating that like... And he was like, oh no. He's... Then he started to panic. And I like, you don't need to panic. Like, you can just tell your AI, like, simplify.

[00:41:44] Kris: Right.

[00:41:45] John: And it will start... Like, said, like, tell it right now. I just... I'm just giving him prompts, you know?

[00:41:49] Kris: Yeah. I wonder if there's I wonder if there's a lot of, like, CSS stuff that's, like, duplicated, you know, versus using styles that are reused. Like, is it putting inline CSS everywhere? As an example, it could be. And it could

[00:42:00] John: be, like, his lines of code is counting CSS and JavaScript and all this stuff, mine is just counting Rails lines. Yeah. You know? I mean, there's lots of little things that could be...

[00:42:07] Kris: Well, I'm still... I'm from... We're wondering from the simplification standpoint. Yes. Exactly.

[00:42:10] John: Yes. And that could definitely be a thing that would help with it, and that would increase it. But then I was like, oh, actually what I'm really curious about is about a year ago, we started using AI a lot more and really like December is when we really started doing it. But I was like, well, what's the difference between a year ago and now? And so I looked and like BoxOut a year ago was 130,000 lines of code. It's now 330,000 lines of code. So 150% increase in lines of code. Fireside, even more epic. So Fireside is 338,000 lines of code. A year ago, it was a 104. And again, lines of code is not like... Yeah. That's not a metric that really...

[00:42:47] Kris: Right. It matters.

[00:42:48] John: Yeah. Don't optimize for that.

[00:42:49] Kris: Yeah.

[00:42:49] John: But fascinating to see the explosion. So, like, fireside is literally 223% more code than it was a year ago, a year and a half ago. Yeah. No. A year ago. August. Yeah.

[00:42:58] Kris: Yeah. No. Yeah. The... Yeah. Was just... Was... I'm trying to get to... We had our... Another anniversary last week when you're out of town.

[00:43:06] John: That's right. Yeah. And I... You know what? And I put anniversaries like that on my calendar so I don't miss them, and I still missed it. And you reminded me. And I was like, no. That can't be right. And I looked, and I was like... And it didn't show it. And I was like, no. He's wrong. And then I realized it was because I had, like, four other things.

[00:43:20] Kris: Oh, too many things. Yeah.

[00:43:21] John: Yep. I scrolled, I was like, oh, crap. This is right. Yeah. How did I forget this? So, yeah, fireside anniversary last week. Two years officially.

[00:43:27] Kris: Yeah.

[00:43:28] John: Have, all new studio.

[00:43:31] Kris: Mhmm.

[00:43:31] John: We got a new website that we're starting to iterate on. Yep. It's... But it's live. And new podcast? New podcast. Another one. Yep. And a second...

[00:43:40] Kris: Well, I I meant yours. Yours. Yeah. Yeah.

[00:43:42] John: Our new one. So, yeah. So, we have the Friday deploy. So, that's... It's... We're just waiting on Apple. Apple messed up the submission and I submitted support request, and I have heard nothing. So crickets. So if anyone knows Apple, you know, and I'm just kidding. So, you know, all of that, we have all new servers. Every single piece of infrastructure that serves fireside is brand new compared to two years ago. Nice. Which is kind of crazy. And then on top of that, just efficiency and, a ton of things. So... We've got some really good stuff in the, coming real soon. Yeah. So... Well, but anyways, that's the very long way of saying, like, super fascinated by this, you starting to be technical in those ways. And how you found it. I think that's a great

[00:44:30] Kris: Amber.

[00:44:31] John: The fact that if you're a technical person, you can just step in and start doing this thing.

[00:44:36] Kris: I don't want to move on from it, but when we created... I would say that my development team, when we were working with a client doing some really hard application development, this was maybe four years ago, I wonder how much we could have progressed that software further. Because it was new technology, and so there was a lot of challenges with, and there was lot of proprietary technology that we are trying to work with. And I don't think the app exists as it was trying to. But I'm like, could we have... There were just so many render issues. It was three d modeling and a bunch of other stuff. I I'm... Like, not... We can't go back. Yeah. But that'd be... I'd be very curious. Could we have gotten it further?

[00:45:31] John: You could have. A 100%. Yeah. Could have went further than you can imagine. Yeah. Like, because it's so good at at that three d modeling and some of that in particular. Yep. Either of them. Know. You could have picked either one and it would have, in a day, you could probably build what you couldn't even imagine.

[00:45:48] Kris: Yeah. I think though, even if we had it, it would have been new the client may not have let us connect it because we had to connect to client systems in order to get all the data and all the So I think it would still need like to be today or even a year from now in order for that work to exist the way that we would want to be able to use the tools.

[00:46:09] John: But

[00:46:10] Kris: I don't know. Some of those things we've worked on over the last like five, ten years, I'm like, when we had really hard technical problems, could the thing have existed? Probably could, yeah. But it might not need it anymore, I don't know.

[00:46:24] John: Do you feel that? Do you feel the subtle fan of the flames? So one recommendation. I've already given one recommendation, but I do have another. But so recommendation book tool idea. I guess you kinda gave yours. I mean, use Kodex or

[00:46:38] Kris: use Yeah, I get another recommend. Well, I'll I'll re recommend. Yeah. Because it came up this week, and so we had we hosted we host customers and clients all the time here. Mhmm. Some are just in the space to rent it out for you know a meeting or a leadership team. Whatever it might be. And think we've... Have we talked about unreasonable hospitality? Yeah. Because, like, I read it, like, when it first came out. We probably talked about it then.

[00:47:06] John: I think... Whatever. Yeah.

[00:47:08] Kris: I... And and that might have been before we opened or just as we were opening last summer when it came out. Yeah. And I was like, oh, like, that's what that's what I want us to be. And the group had come in and they were actually using that book for their leadership team. And they were so, like, appreciative of the way that we, let's say, we're personifying that book in the space, the way we greet people, way we check on them, all that. And we don't always get that feedback. Like, I know we're trying to achieve that, but I'm, it's hard to, we don't always get that feedback loop of what people actually experience. And so I'm back on recommending that book because I think we're we're not... I I've I've talked to the team about I want to be the the Michelin star of like these sort of spaces. And what things need to be true in order for that to happen. And I think it's just like, wouldn't read it again because I'm like, when I read it in the space we were in at the time was very different in terms of the progress of space. And I think I would get a lot of different things from it going back through it. And I just finished the... I know we talked about probably, like, a month ago, the Leaders Eat Last book. Yeah. And so I have... I need to to do another one, and that... I might go back to back to listening to that one again.

[00:48:25] John: Nice. My recommendation is Cloud Opus 5.5 came out recently.

[00:48:31] Kris: Okay.

[00:48:32] John: And the whole internet is exploding with basically using it to make twenty, thirty second intros

[00:48:38] Kris: as movies.

[00:48:40] John: The first stab is impressive enough you're like, okay, this is pretty good. But if you actually iterate with it, you can make a really good intro video for product, website, whatever. Yoder sent me one. I talked to him at basketball. He sent me one after basketball, and I was like, this is great, explaining his business. And I've seen, you know, several people in our chat room and stuff doing the same thing. And it's like, it's, they're mind blowingly good for how little effort you put into them.

[00:49:05] Kris: Yeah.

[00:49:06] John: So yeah, try it out. Just go to Opus, you know, five five and basically just ask it to make, you know, a short 15 whatever 30 video that explains your product or website. Give it access to the code if you can or just the website. Tell it to keep your branding the same and it'll just crank it out. It's pretty cool.

[00:49:25] Kris: Nice.

[00:49:25] John: Yeah. That's it.

[00:49:28] Kris: That's it. This is a good one.

[00:49:31] John: Oh, that was... That... Let... This brought the heat. I've... Yeah. That one was a good one. We just needed just... I just need to go to Austin and randomly get inspired, and you just needed to have all the good things happen that happened to you. So...

[00:49:46] Kris: Well, if you haven't liked or subscribed, you know, give us a comment. Be the second comment.

[00:49:51] John: Smash that. The second? There's...

[00:49:53] Kris: I don't know.

[00:49:54] John: Oh, I was gonna say, do we have a comment?

[00:49:57] Kris: Soon. I think we've had some on some of our shorts. Nice. Yeah. When you're talking about codecs and AI and all that.

[00:50:03] John: I'll

[00:50:03] Kris: have to look at those. Yeah. But cool. We'll hopefully catch you in another week.