[00:00:00] x.com. I'm gonna start calling it X. Have you started calling it x? No, I can't. I can't. I like the mental, it's just too much history. Yeah. It's like you physically can't do it, right? Like I can't, well, what do you call a tweet? You just call it a, a stroke, like a mini X. I don't understand. It's retarded. I don't know.

[00:00:18] It's like, I mean, calling it a post is honestly fine. Like it's definitely boring and it kind of loses the whole, like what makes Twitter? Twitter? It's like part of what makes Twitter, Twitter, it's, it's kind of like a tweet is like a whole concept. It's a thing, you know.

[00:01:21] Kalen: dude, it's nice to finally fricking chat, man.

[00:01:25] Dylan: Well, we've been chatting. It's just the funny part. We've been chatting for a while. We, yeah, we chat all the time. Uh,

[00:01:33] Kalen: x.com. I'm gonna start calling it X. Have you started calling it X? No. Twitter. For how long have you been on Twitter?

[00:01:40] Dylan: Since 2009.

[00:01:44] Kalen: 2009. Oh, shit.

[00:01:47] Okay. I don't remember the year for me. Wait a minute.

[00:01:51] Dylan: I just remember it was my freshman year of college. So it was 2009.

[00:01:55] Kalen: I think I'm like 14. Damn dude. Anyways, , so do you call it X or

[00:02:00] Dylan: I can't. I can't. I like the mental, it's just too much history. Yeah. It's like you physically

[00:02:05] Kalen: can't do it, right?

[00:02:06] Like, I can't,

[00:02:08] Dylan: well, what do you call a tweet? You just call it a a, a stroke, like a mini X. I don't understand. It's retarded. I don't know. It's like,

[00:02:16] Kalen: I mean, calling it a post is honestly fine. Like it's definitely boring and it kind of loses the whole like, what makes Twitter?

[00:02:24] Twitter, it's like part of what makes Twitter, Twitter, it's, it's kind of like a tweet is like a whole concept. It's like it's a thing, you know?

[00:02:31] Dylan: Yeah. It made up that whole category. I think that was a terrible, terrible branding move. Like, I get that you own the domain x.com. Awesome. Right, but save that for Grok.

[00:02:42] Like he could have made x the grok brand.

[00:02:45] Kalen: I actually had never considered that. I I don't think Grok was in his head though, when he started, well, it was probably somewhere in his head, but like when he did the X thing, I think that was a good couple years before this whole, all this AI stuff ramped up.

[00:03:02] Dylan: Yeah. The timeline's really kind of iffy. I, I don't remember if he bought Twitter before the divorce with OpenAI. 'cause he was like, he was an investor in OpenAI. He was, and then he

[00:03:13] Kalen: was,

[00:03:14] Dylan: yeah. I just don't have it like the right events in order to remember if GR was after x.

[00:03:21] Kalen: Yeah, I, it, I just think the name, I don't, I just feel like the name Gro, he just came up with like, in the last, I don't know, like year ever since.

[00:03:30] He's like, I think once, 'cause this whole time he is like, am I gonna have to do an AI thing? And he is like, trying to avoid it, And then at some point it became clear to him that he had no choice but to do it. I think was is my under, like, that's how I'm assuming it went down.

[00:03:45] And I feel like that was the point where he is like, all right, lemme can pick a name. Whereas like the x.com thing has been simmering for him for like forever. And then like, yeah, I know it. I am, I feel like it's, he really has added a ton of features. Like I'm using it and I'm going to myself, I have a paid subscription.

[00:04:06] I have no ads. I have editing, but all the dumb stuff, , editing a tweet, undoing a tweet, and now I have an AI thing, and it's like, and I know he really wants to make it like the everything app, like in China, whatever that app is where you can pay your rent, pay your bill,

[00:04:22] Dylan: WeChat. I think he's gonna WeChat.

[00:04:24] Kalen: Yeah, WeChat. I'm sure he's gonna pull that off. So I, and then when I saw the merch store and the fact it was on Shopify, it looked so nice. I was like, you know what, , I like the brand too, like, I like the way it looks and stuff like that. But even though I decided, I was like, I'm gonna start calling it X the next time I like said something about Twitter, I just, I couldn't do it.

[00:04:45] I was like, I can't, it's Twitter. Like I can't

[00:04:50] Dylan: tweet, tweet, pay, tweet, shop, whatever.

[00:04:56] Kalen: Oh, so dumb.

[00:04:57] Dylan: Have you, I wonder,

[00:04:59] Kalen: yeah, sorry.

[00:04:59] Dylan: I was gonna ask like, have you figured out how to filter out the garbage? Low quality viral content that they try to push on your feed? Like, have you found a way to, to feed that? You know,

[00:05:09] Kalen: you know, it's funny you mentioned that, dude. 'cause I haven't seen that for, and I, maybe I'm scrolling less because the way I would see it before is that it wouldn't be till I sort of scrolled through all the like, Shopify type stuff that it almost seemed like it fell back into another mode or something.

[00:05:27] And I might've been scrolling left lately, but I did click not interested a, like a billion times and then like, it still kept coming for like a, a month or so, and I hadn't even thought about it until just now that I don't really see that stuff, on there anymore.

[00:05:42] Dylan: I felt like at one point I was like finding the ocean, like trying to build a sandcastle on in the ocean.

[00:05:46] It's like you say, not interested, it'll just give you a different form of crap content. I never liked basketball. Why is it giving me NBA insider like meme stuff? I don't know like.. If I click not interested on that, it's gonna find the next like F1 series. I don't know, like it's going to find the next like low quality to me, low quality thing.

[00:06:07] Yeah. Like I'm sure like middle aged men in America. Love this stuff. Nice

[00:06:13] Kalen: white man.

[00:06:13] Dylan: Yeah. And it's, it's correct, like the persona's, correct. It's just like, dude, I'm a weirdo. Like I don't know what to tell you. Yeah.

[00:06:22] Kalen: I'm breaking your algorithms. It is like, you can say not interested in the subcategory, but the category of dumb shit.

[00:06:28] You can't like, yeah, say that PFC fights or whatever, I don't know. But now I have like the opposite problem, which is that when I, so I used to use TikTok and then I stopped and I started using the Twitter video, , thing like just like my, like 30 minutes of downtime during the day where I'm just do watching stupid shit, like intentionally.

[00:06:49] When TikTok got banned or whatever happened, I switched, to the Twitter video. And I actually liked that it was sort of just like not work related. Like it wasn't, very well correlated with my like, regular Twitter feed. , And I liked that. But now it is, it's like all business stuff.

[00:07:06] It's all startup stuff. It's all, so if I'm trying to unplug, I don't want to use, so then I start using TikTok again

[00:07:12] Dylan: for that. Oh, so you can unplug. Maybe. I haven't figured that part out yet. You know, see you like a a personal Twitter mode and like a work Twitter mode. 'cause they're so similar.

[00:07:23] It's so unique. Love it. Like LinkedIn. LinkedIn is like this is going to be work crap. You just know it is like Twitter, right? Yeah. It's like, it's a little bit ambiguous. It could be either or, right. Do

[00:07:36] Kalen: you, yeah. So do you follow like a range of stuff on Twitter? Or is it all Shopify stuff?

[00:07:42] Developer stuff?

[00:07:43] Dylan: I don't like do a good job of curating my follows. I kind of rely on the algorithm to be like, alright, somebody in my network like this tweet. Therefore yeah, it's probably something cool. , And then if I like see that name pop over up, like over and over again, I might follow them.

[00:07:59] Kalen: Is it all like tech work related or is it like, do you Pretty much like nascar, but yeah. Okay.

[00:08:04] Dylan: No, yeah, pretty much. Yeah. Like if I wanna follow that kind of stuff, I'll go to like Reddit, I think. Like personal, right? Like NFL or whatever. Are you a Reddit guy? Oh, I love Reddit.

[00:08:15] I've been in Reddit longer than Twitter. Yeah. It's bad.

[00:08:19] Kalen: Nice.

[00:08:19] Dylan: I want Wish.

[00:08:21] Kalen: No, I wish I was a Reddit guy. 'cause like I'll comment every once in a while on some random thing and then I, like, I never respond like, but like, I wish, because like if I have a friend of mine that's just deep dive Reddit all the time and it, it's like you're probably getting more done, you know, like more productive in some ways.

[00:08:38] Dylan: Yeah. Maybe gives you that feeling of that. And then like the more I read into it, I'm like. Lot of these people have no idea what they're talking about. Like Yeah, if especially you're looking on like the JavaScript specific ones or like, you start getting into one field really deeply and then you start to look at the posts about that subject.

[00:08:57] You're like, actually a lot of this stuff is not correct. Imagine that applied to all of the different subreddits. There's a lot of stuff that's like, not,

[00:09:07] Kalen: that's interesting.

[00:09:08] I don't know if I've ever looked at a topic on Reddit that like I knew a lot about already, although like I'd go on the Shopify, Reddit, but like I was just there to like sort of promote things and like as non-promotional way as possible, but I wasn't really like interested in it,

[00:09:23] Dylan: but Oh yeah, they are like, that's so strict.

[00:09:26] Like they'll have bots that detect URLs and, yeah.

[00:09:31] Kalen: Yeah.

[00:09:32] Dylan: I wasn't.

[00:09:33] Kalen: No, and I wasn't like doing dumb shit like that. It was just like trying to legitimately be helpful with replies. And then if it was something was very relevant to like Mesa that, you know, the product I was promoting, then I'd like mention it directly or something,

[00:09:46] Dylan: but Oh yeah,

[00:09:47] Kalen: nothing, not like ultra spam guy mode or anything

[00:09:51] Dylan: like that.

[00:09:51] Yeah. Yeah. You could sniff that out on shop the Shopify, like, do you do Shopify community forums? , Not the dev ones, but the merchant ones.

[00:09:59] Kalen: I did for a while. I had a good little stretch there. And I haven't done shit for a long time,

[00:10:04] Dylan: which,

[00:10:04] Kalen: mm-hmm. What do you,

[00:10:06] Dylan: uh, I try to, it's just so, loud and it's so fast.

[00:10:11] Like people definitely have botified it, at least it seems like it's Spotify or at least notifications are going off. And,

[00:10:18] Kalen: they migrated down one to discourse just recently, right? Mm-hmm.

[00:10:22] Dylan: So you could, like every single discourse category is A-J-S-O-N and RSS feed. Right? So you could do the same thing, get notifications on the keywords, but like, even then, if you have a life of any kind, it seems like you can't be helpful, or at least you can't.

[00:10:40] I don't know. Yeah,

[00:10:41] Kalen: it's, yeah, I mean, and I did it very intentionally because I was trying to, build up, some cred, you know what that type of thing, like it has to be something that, you know, you just very intentionally do with kind of like a focused goal.

[00:10:55] And otherwise you're gonna end up like, I still get a lot of leads for like, flow related stuff from the posts that I did. And the flow stuff. I didn't even, I, because most of like the year, year and a half, I was doing like MESA related posts and still get a few for those.

[00:11:08] But then I did flow related stuff for the, like six months or nine months. I dunno. And I get pings on those, , from time to time. That's cool. , Yeah. But I try to avoid doing flow consulting work now because, with all the cloud code stuff like.

[00:11:27] Clicking around manually in the interface just seems insane to me.

[00:11:31] Dylan: Like

[00:11:31] Kalen: I Think if flow is great, if you can pop in, spend a few minutes, get something working that solves your use case, it's absolutely magical. And like as an app, like to create them so that your customers can do those things, is, is a magical, but like if you're consulting and you're spending like 10 billing, like five hours for some complicated workflow, and it's like, and now I'm thinking I could do this in 20, 20 minutes with.

[00:11:54] Cloud code because, it just, the nature of it, and like testing is like if you've ever built a complicated flow and you're like copying and pasting code in hit, rerun, refresh, like doing that 20 times, like I built out all these CLI tools to test run codes locally and stuff like that.

[00:12:13] So anyways, and I built a little, open source thing to called shop worker to basically do automations. So now like when I get an email about flow, I'll be like, uh, well I'm doing most of my consulting with this thing now. Like, lemme know if you're interested, type thing.

[00:12:26] Dylan: Wow, okay. I mean, I've been, I've been following , you and David talking on this pod and hearing about the shop worker over time following your tweets.

[00:12:34] Yeah, for sure. Oh, because I've, I been super curious about it. 'cause I, I also believe that's like the step I've talked to Paul at last, unite, not the last one, but the year before. Right. And I was like, so why? Like, what would it take for anybody to like programmatically create flows? He's like, eh, not possible check zones.

[00:12:51] We make sure no one could do that. 'cause we only have validation on one way for us to computationally do validation right from a programmatic workflow. Be a nightmare for us. Please don't do this. Like, just don't try. Um Right. And I was like, but I had the same thought as you. It was like, yeah, there's gotta be a way to do this.

[00:13:09] And so

[00:13:11] Kalen: I, I may have potentially not confirming or verifying may have, possibly, it may have been in a dream and I'm recalling it weirdly that I saw instance of Shopify with the checks on removed. Um Mm. And that, if that were to ever happen, it would be pretty cool.

[00:13:30] Dylan: I thought, I saw Paul's mention something on LinkedIn about like, they're going that direction.

[00:13:35] They're trying to. Oh, did he say that? over, I'm on vacation for this past week. I skimmed over it. It was okay. Some kind of conversation. He didn't post it like as on his timeline, but it, I saw him interact with somebody else saying like, they're experimenting with some new direction.

[00:13:50] It seemed like they were talking about prompting workflows.

[00:13:53] Kalen: Yeah. So, well, if you said it publicly, then it's cool. Yeah. Basically there's a version of it that, has, the checks are removed and, , and then, I mean, it wouldn't be a surprise to anybody if. Sidekick got flow support, right?

[00:14:07] Like, yeah. Everybody's like, that's probably gonna happen, at some point, relatively soon. And I think like, I mean, who knows? Like maybe it's gonna be totally perfect. And I mean, you would expect that to be the main place you'd want to go to do an,, prompt for a flow.

[00:14:23] But like, it, I still think it'd be cool if third parties could build out their own things right? Against, like, building out the JSON. Because like, one of the surprising things with all this AI explosion is like, sometimes the third party does a better job, maybe even just temporarily. And then you can learn from them like the way cursor, did their thing as a third party.

[00:14:44] They, like they out competed the first party, like Microsoft S Code, you know what I mean?

[00:14:49] Dylan: Yeah. Like walk me through what you think by, um, I'm trying to understand.

[00:14:54] Kalen: IJI just think that if the js ON if you can upload A-J-S-O-N without a checksum, and then you can generate that JSON, like you could build your own little, , thing to generate JSO using mm-hmm.

[00:15:04] Using PO code or mm-hmm. You know what I mean? Mm-hmm. And that maybe would be a lot better than whatever comes native into sidekick.

[00:15:12] Dylan: Yes. Yeah. Like, yeah.

[00:15:14] It seems like Shopify will always create the core thing for free. It will cover 80% or 70% of the use cases.

[00:15:22] Right. But that remaining bit, that's where the opportunity lies for a partner to be like, all right, , you have an ERP. That's right. Super complicated. Guess what? Right. I got you covered all these edge cases that Shopify can't focus on because they're catering to a much wider market. It's still an opportunity like template, like printing orders, that what's the Shopify printing order app called?

[00:15:49] Is it called Order Printer? , I think it's called,

[00:15:51] Kalen: sounds somewhat familiar. I don't know if I've seen that word. It's sometimes I feel like I haven't seen very many apps.

[00:15:58] Dylan: I feel like the same, I feel like I've only touched like three parts of Shopify. I've been in it for five years.

[00:16:06] Kalen: I'm like, so I mean, like, somebody will ask me something and it sounds like it should be really basic, and I'm like, I don't know. I'm only two and a half years in. Right, right.

[00:16:15] Dylan: Like, what do you mean inventory? I'm like, does inventory just works? Right. They're like, no. Right. There's all kinds of management and like locations and I'm like, oh, and production sometimes maybe there's not production.

[00:16:24] Maybe it's dropship, like digital versus physical. There's a lot of different moving parts , and I'm like, oh, so it's not just that. It's a lot to it. Yeah.

[00:16:33] Kalen: Yeah. Yeah. Like if you haven't messed with the product API as much or the inventory API as much. Something that like surprised me is that like, did you know that you have to like activate the inventory before you can set inventory levels in the API.

[00:16:45] You mean like publish it? Not you publish it, it is actually called activate, like the mutation is like activate inventory or something like that. , And it's just like when I ran across, I was like, that's weird. Like you can't just send inventory levels in a thing, like in a,

[00:17:00] Dylan: to a draft product basically.

[00:17:02] Yeah. Uh,

[00:17:04] Kalen: yeah.

[00:17:04] Dylan: Well it's still a draft, like visibility draft, but

[00:17:07] Kalen: Yeah,

[00:17:08] Dylan: it sounds like it's a grandfathered thing. Like a grandfathered status has been around since the beginning and they're like, well. Well, it

[00:17:14] Kalen: could be, but, but like inventory does need to be activated in a, in a specific location. 'cause like if you have two locations, the inventory, it's, it's not like the inventory is zero by default.

[00:17:26] It doesn't exist. Which Right. Is probably more like what the model, you know, underlying

[00:17:30] Dylan: model. Yeah. Inventory's, null not zero. Right, it's unavailable. Um,

[00:17:36] Kalen: right. See, I'm

[00:17:37] Dylan: already learning stuff

[00:17:38] Kalen: with you guys. Yeah. Like I, I just, since you mentioned like that you don't do much with products and stuff like that, I figured like that, you know, but anyways, dude, I, so it's funny, I went through your Twitter, beforehand to just like, to just, you know, like find the, you know, things in case I ran outta things to talk about or whatever.

[00:17:55] And you have very consistently solid tweets. Like, because like I'll see your tweets on my feed, but they're just mixed into the feed. But looking at them all in a row, I'm like, these are all good. Even somebody with cool tweets, it'll be like maybe like 50% of 'em are cool. 50% of 'em are like, ah, dude, shut up.

[00:18:13] like, , That's probably how I am. But I looked through yours and I was like, every single one is just great. It's just great little tweet.

[00:18:21] Dylan: I'm flattered. My friends and family don't believe you. My brother's like, he even pulled out like a recent, like, I think it was like Thanksgiving or something.

[00:18:28] He is like reading through his, the tweets. 'cause they have, they're not in tech. They have no idea what I'm talking about. They're like, right dude, you're speaking mumbo jumbo. Two people like it. Like just stop talking.

[00:18:40] Kalen: Dude. It's so funny because I know exactly how that feels too, to be like, oh my gosh, I got a hundred views.

[00:18:47] Like this is the, what am I wasting my life on? And then like , when you realize even if there's one actual human being like there, let alone if there's like. 30 that are seeing it. You know what I mean? Yeah. It's like, it means everything, you know? And it's really like, you know, it's stupid to focus on the numbers, but that's the most obvious.

[00:19:06] Like for like a reg, for a normie, basically. Yes. Right? Yes. Like I get that.

[00:19:11] Dylan: Yes. I wish I would've like done that sooner. 'cause like there was like things to me that were cool when I was younger. I didn't really like tweet about it. I'm like, who cares? But Right. You never know what's gonna happen.

[00:19:22] You never know when you're gonna try at Riverside for the first time with the buddy you've been talking with for a, for a while. Right. It's the same things. You have just no clue. Yeah.

[00:19:29] Kalen: Yeah, it's just sort of a reflex for me now. It's like my brain goes down a certain path of like, this is new, or this is funny, or this is like whatever.

[00:19:38] And it just, I just want to share it. Like I just, I need to put it out into the world, I don't know if that's some weird, problem with me or something, or I don't know what it is. And then before you get it out, it's like your head doesn't like continue.

[00:19:52] It's like you're stuck somehow in some line of thinking. And then once I get it out, I'm like, okay, it's out. Now I can think about the next thing.

[00:19:59] Dylan: Yeah. I'm the same way. Like once, I feel like almost like catharsis when you like get it out there, you're like, okay, that's resolved. I can at least, yes. Get that off my mind.

[00:20:09] Onto whatever,

[00:20:11] Kalen: yes, exactly. Yeah, that's exactly it. But anyway, so I put together, I didn't, I wasn't intending to just talk about all my own stuff for like, I put a whole, whole list of, starting with your five year mark. I saw the, uh, congratulations a little golf. Oh, congratulations on the, on the five year mark.

[00:20:31] Thanks, bud. , That's cool. I was like, Dylan's a good dude. I'm really happy for him also. I'm very jealous and, you know, fuck him. But at the same time, that's my thing now, that's my thing is I wanna get this app thing going. Like, but when I like, I'm literally genuinely happy for someone and the next thought is, ah, fucker.

[00:20:51] You know, it's like,

[00:20:54] Dylan: that's just how I feel.

[00:20:55] I'm gonna be jealous of you once shop worker. 'cause that's just, yeah. It's just a wide, it's such a widespread use case and it's like so complex. It's hard. It's gonna be hard to replicate. I just think, I, I think it's gonna be a winner for you.

[00:21:10] I really do. Thanks. I think you should publish it. I think you should. Yeah. I think it's time to publish it. Yeah.

[00:21:17] Kalen: All right. All right. Thank you. Thank you. I appreciate it. Um, well, it is technically published. It's on gi it's on GitHub,

[00:21:23] Dylan: but, well, I mean, like if it's possible to publish it as a Shopify app.

[00:21:27] Kalen: Oh, as an app, right, right, right.

[00:21:28] Yeah. So I was working on that. I was doing a Docker thing and a with a clot because I was like, well, if I publishes an app, it needs to be usable by people. So then I was like, I need to just let them prompt to create stuff. And so then I was like, okay, but as , because there's two ways to use it, you, you could just use it locally in your CLI, that's like the way I wanna use it.

[00:21:50] And I was like, okay, let me jerry rig a CLI inside of a container.

[00:21:54] Dylan: Mm-hmm. Mm-hmm.

[00:21:55] Kalen: That the app talks to. And I kind of got it like most of the way there. And then like I had like, this is what I'll do things. I'll be like, is anybody interested in this? I got a couple likes. Cool. One person dmd me. I was like, cool.

[00:22:06] And he is, I was like, which store do you want to have an access on? Okay. Okay, whatever. And then I was like, okay, let me actually deploy this. 'cause I had it working locally. I hadn't deployed all the containerized stuff yet to like fly.io. Mm-hmm. And so then. I was like, I emailed him back or I messaged him.

[00:22:23] He didn't get back to me and I was like, okay, if he doesn't get back to me, I'm just, I'm not gonna, like, I want there to be a real person on the line, like a real bite in order to like, and then I'll spend eight hours fixing it or whatever. Oh. But if there's not a single person that actually is replying to my message about like, then I'll just kinda, and then maybe it'll circle back.

[00:22:44] Like, maybe it'll circle back later. So I'll get another bite but I kind of do things like kind of progressively like that, which is probably stupid.

[00:22:51] Dylan: No, that's smart. I think the right way to do it is to get, five to 10, 10 seems like a lot, but even maybe even five, three to get like deep connections with at least three people, three merchants that absolutely need what you're looking, what you're building.

[00:23:08] Kalen: Right.

[00:23:08] Dylan: Have them try it before it's published. Then you have the reviews lined up. Soon as outta the gate, right. To get the reviews. I think I could be wrong, but I think there's probably some kind of internal promotion algorithm. If you have a high velocity of reviews right after publishing, you have a higher chance of getting that front page.

[00:23:29] And then it's like a moonshot. Like, I can't even imagine the growth you'd hit. I've never been on the front, the homepage. But I think it would be really big.

[00:23:38] Kalen: You're, but you're in the, customer accounts, like promoted section or like, I'm in the security and security and privacy, I believe, and I think one other, but definitely security and privacy or, and maybe fraud.

[00:23:53] Maybe fraud. Top 50 fraud. Okay. Okay.

[00:23:56] Dylan: But it's, it's still a niche On top of that,

[00:23:59] Kalen: the app store category.

[00:24:01] Dylan: No, I'm just saying like, even if I'm in the top category for fraud, that applies to all stores. Not all stores qualify for using my solution for fraud. Fraud. Oh, gotcha.

[00:24:10] It's not like fraud insurance that applies to every single merchant. It's a tool. . Only applies to certain types of fraud that are very, very expensive. Or high risk. So, so dude, how did you get into this? Oh man, , okay, so let's go back to 2016. I'm freshly ish, graduated.

[00:24:30] I started freelancing. , But I had this crazy idea with my wife way

[00:24:34] Kalen: back from the beginning.

[00:24:35] Dylan: This is like 10 years ago. I had this crazy idea with my wife to build a tiny house. Part of that was okay.

[00:24:42] Kalen: We're like the same person. We're literally like the same person. It's

[00:24:45] Dylan: really

[00:24:46] Kalen: me out. Yeah. I won't, I'm a big tiny house guy.

[00:24:49] Okay. Sorry.

[00:24:50] Dylan: Okay, well I can tell you about that in a different episode if you want. But anyway, that led to me trying to find a remote. Like I couldn't concentrate on doing that and following contract work flying to New York, it was crazy. So I found a local company that was doing a startup that was doing RV rentals.

[00:25:07] It's called RV Share now. It's the world's largest RV marketplace, bro.

[00:25:11] Kalen: I've gone on RV share a billion times. Really? I'm like, yeah. Whenever I'm like, I need to rent an rv. And, and I think I might've even used them once. , That's crazy. You worked on that.

[00:25:22] Dylan: Yeah. So I was engineer number one. I was early, early employee there.

[00:25:27] And it's based outta Akron, Ohio in my backyard. Now, it's, now it's based outta Austin. But we were purchased into my tenure. So I started working there. I built their search. API built like some like affiliate stuff. But part of the perk of working there is that you get a, basically a yearly stipend to spend on an RV trip.

[00:25:47] So I was like, great, I'm gonna go out to Seattle. I wanna go do like the Oregon coast up and down on like a class B and like just camp on the beach with my girlfriend at the time. But every single RV owner I spoke to was like, eh, we don't book through the platform, just, text me here. We'll just take cash and I'll save you 25%.

[00:26:05] 'cause their commission's 25%. And I was like, oh shoot, if this is happening in Seattle and, Portland, I wonder how widespread this problem is. So the next hackathon I built like a fraud app that would troll through the messages and find keywords that display the intent of going off platform, so like PayPal, cash direct, oh, et cetera.

[00:26:26] And I found somebody in sales that solve this problem too firsthand. He is like, dude, we have to, we could probably make a, like a bunch of money for the company if we just basically messaged these owners and told 'em, knock it off. They get kicked off. 'cause they're stealing our marketing revenue, our right marketing spend.

[00:26:42] Right? So that's how Scrub Cop was born is how we called it. So Scrub Cop just found these like suspicious people, owners and then just mm-hmm. We went through the queue and messaged them, right? And then someone else is like, Dylan, this looks like fraud over here too. I'm like, yeah, that's self booking fraud.

[00:26:59] People are self booking. Siphoning credit cards, stolen credit cards and making that 20%. Or you know, like 20% became a fraud guy. Became a fraud guy. I just like saw the fraud. I tried to make tools in this app. So cool. So eventually that became my full-time job and I implemented Id verification on both the renter and owner side at one point, because we found that, I went through the whole list.

[00:27:21] I went through IP detection, I went through VIN detection, I went through like. Trying to make an, I even made like a machine learning algorithm, which is the best tech at the time, based on all these signals to like try to determine fraud on the renter side and owner side pictures because they started putting their phone numbers and stuff and pictures and phone numbers, the way they'd circumvented the messaging system to bypass my, like there was all this crazy stuff like cat and mouse game.

[00:27:45] ID Verification was great because once you're caught once, it's extremely difficult to , get another ID that matches your face, that matches your billing details and pass again, like you can commit fraud once and then you're out I mean, at best case scenario if you're that brazen. yeah, I did that for years.

[00:28:06] I finally solved the problem pretty well by 2019. Then 2020 hit, I had a lot of downtime like everybody else and I was like looking around the platforms that were easy for developers to pick up.

[00:28:17] Kalen: And when you were doing all this full-time fraud stuff, was it that all at RV share?

[00:28:21] Dylan: Mm-hmm. Yep. All RV share.

[00:28:23] Gotcha. So I was, I finished the tiny house. I was traveling, with my wife. She was travel nursing, so we would just, you know, spend three to six months at a place. I was just working at a coffee shops and coworking spaces, that kind of stuff. Yeah, like just made an app that employees could log into ban or warn or silently banned eventually.

[00:28:46] , There was all kinds of stuff, but That's so crazy. So that whole like fraud world put me in that mindset that was marketplace fraud, slightly different. E-commerce fraud. Didn't know much about. I had a feeling that people were ignoring it or it wasn't, like, not ignored, but like not well understood.

[00:29:06] Right, right. So I took the best tool that I knew, brought it to Shopify, just kind of on a hunch, like mm-hmm. The right way, right approach is what we've been talking about, which is like, you know, find merchants that are vested in understanding that this solution solves a problem. They have. By the time I was just,, you know, engineer with zero e-commerce experience and just built it, published, it took a long time to find the market, or understanding even like the right words to use.

[00:29:36] Like, I didn't understand merchants language at all.

[00:29:38] Kalen: It's called Real id, right?

[00:29:40] Dylan: Yeah, it's called, it's, it was, that was the name out of the gate was Real id.

[00:29:43] Kalen: Oh, okay. Well, you had that, you had the good, a good brand right from the get.

[00:29:48] Dylan: Got lucky because RV share, I always, I learned from RV share, like, so their main competitors, outdoorsy and Austin. Mm-hmm. And outdoorsy just couldn't rank well because RV Share the title captures Right. , The intent, which is to rent an rv, you're sharing your RV outdoorsy, you're like, what does that mean?

[00:30:06] is not as good. So I learned that like little marketing lesson from, , the Masterminds of RV share.

[00:30:14] Kalen: That's cool, dude. , Damn. And just living the dream

[00:30:18] Dylan: ever since. Oh. You know? Yeah. I guess I'm, I'm very lucky. I'm very fortunate. Yeah.

[00:30:25] Kalen: That's awesome. I was talking to a friend of mine and he was like saying, he's like, do you wanna like build an app or you know, he's like a freelancer.

[00:30:32] And I'm like, yeah. And he's like, man, he's like, all the app guys are like, they're always like, just desperate for reviews and stuff like that, you know? That's true. And I'm like, there is a, and it's funny 'cause everybody I ran into in additions, , there was this anxiety related to something with the app.

[00:30:50] , A is AI gonna kill it is, , I do I want Shopify to acquire it or, or, or do I want somebody to acquire it? And like, , should they acquire it now? Or like, you know, should, is now the time. Like, there's so much pressure. It's like a bit of a pressure cooker even though , like you could also just be super like lifestyle business about it probably too.

[00:31:10] Dylan: I know both ends of the spectrum. Like I know people that. Incorporate into Shopify is so, well, they hardly have any infrastructure and they literally don't look at their email for like two weeks, which we, I could not comprehend the other end exactly what you're talking about, where it's like we have cutthroat competition.

[00:31:29] Kalen: Right?

[00:31:30] Dylan: Or maybe your merchants are like a high, like you have a high percentage of revenue to one merchant or three merchants in particular. Like they take up like 60% of your revenue or 70%. So you're kind of like on call, all the time. Or maybe you're like right at the cusp of like 20 reviews. It just takes one, one star to bring your average down to like three point.

[00:31:53] Who cares? 'cause at that point you're at three,

[00:31:56] Kalen: at three point, who cares? That's the show title

[00:32:02] Dylan: The reviews man. Yeah. I don't know. It depends. , If you have an app that applies to every single merchant. That means therefore there's a lot of competition for it and the reviews are paramount. But if you're like super niche,

[00:32:17] Kalen: right,

[00:32:18] Dylan: I don't think it matters nearly as much. Yeah, there's downsides to bat that too.

[00:32:22] It doesn't grow nearly as fast. I mean, it took me five years. Pros out there are probably like laughing, you know, like it took me like one year or two year, two years to get there at most. It took me a long time.

[00:32:33] Kalen: Right? Yeah. Everybody kinda is on their own journey, you know? It's like, it's dangerous to compare, and to feel whatever, you know, like negatively about it, But it's also human nature, you know? So it's like if you can find a way, to just kind of draw inspiration from people that are like a little ahead of you without it like eating you up, you know? And I think that's like the productive way to, to do it. Yeah. That's a good point. Grass is always greener and in some way.

[00:33:01] I mean, not always. Some, not always. Like, sometimes it's just, I don't know, to just like, you know,

[00:33:07] Dylan: I'm starting to think like more services, like what you're doing, because like, that's funny. Yeah. I, it's the grass is at least you have like. Closer contact with your customers, you know,

[00:33:22] Kalen: but, , you're not just saying that you literally, were thinking that you'd prefer to be doing services,

[00:33:27] Dylan: I think to be successful, you can't just provide a product alone. I don't know, like even Shopify does this. They offer professional services and they offer a product. I think the services kind of give you Right, the baseline understanding of what merchants need. Mm-hmm. And then you build a product that's highly automated that fits it very, very well.

[00:33:51] Right. If you just try to go for product first, ?

[00:33:53] Kalen: Yeah. But no, that totally makes sense. But like, you personally are not thinking like, oh, maybe the grass is greener doing services, like Yeah, I'm, I'm sure.

[00:34:00] Dylan: If you don't rely on reviews, it sounds like way, way more chill.

[00:34:06] But I assume like the, the hardest part when I remember freelancing years ago was like knowing where your next client is coming from. Yeah. And it's hard to get a retainer. It's hard to get like, committed hours for a long period of time. I assume that's the, those the two big

[00:34:21] Kalen: thingss.

[00:34:21] Horrible. Yeah.

[00:34:22] Dylan: That part sucks. Okay.

[00:34:24] Kalen: Yeah, well, so I, I initially was trying to, very intentionally doing like smaller jobs for like a larger number of clients. I was juggling like t like 10, 12 clients at the same time, and which I wanted to do because I wanted to have like, you know, this, my risk diversified and I really wanted to build it, uh, make it a product I service and things like that.

[00:34:46] But then I just, you know, then an agency got me to do some work, and then I was doing app development. I was like, I like this better, and I don't like getting on a bunch of calls. I'd rather work with like, people that I know. So I've mostly been just doing most of my work for one agency. And it's, I guess it's tech, it's sort of like a job ish in that sense, but technically, you know, still,

[00:35:08] Dylan: yeah, contact switching is so expensive.

[00:35:11] Kalen: Okay. I have so many questions to ask you. It really is, , did you post about doing work Trees recently or something like that?

[00:35:19] Dylan: Yeah. Dude Quad. Yeah. My buddy and I are the same problem where, okay. Even if you have one Shopify app, if you try to take advantage of work trees to develop multiple clawed features, at the same time you still run into the problem of webhooks responding to the same like static URLs.

[00:35:37] Yeah. Right. Yeah. So my thought was like, you need like a meta manager that knows based on some like naming system. Like you tell in your prompt like, okay, you're assigned to just take the next available slot. If it was like really, really smart, almost like an MCP integration. And then it would be like, okay, feature branch, multiple currencies is assigned to.

[00:36:02] I had Halo named terms, so like the grunt branch or the grunt environment that's pre-made. So we can't touch that one. The PR is still open, so I'll go to the next one, which is the marine, the marine environment, and I'll open. Mm-hmm. You know, PR there, the app is available and now you know like, okay, we got three functional en environment or X number of functional environments.

[00:36:24] You still do the legwork to create them, but you can at least like expand your number of work trees. Times that amount, right? Yeah.

[00:36:34] Kalen: Right. Yeah. I wanna, start like I'm doing, you kind of have to do two things at once if you're working in cloud code, like a minimum, you don't have to.

[00:36:43] I mean, you can just sit there and read the code. And actually I was doing a little bit at the other day and reading all the code and I was like, you know, this is good. 'cause it's like I'm a little closer to it, but normally I like to switch back to something else and then, you know, and then I was, started to try to do three at a time and, , or even, and, and then, you know, you see videos of people doing six at a time, but I think the six at a time people, I think that it's full on vibe coding, like paying zero attention whatsoever.

[00:37:10] I could be wrong.

[00:37:11] Dylan: Do you think there's influencers paid by affiliate revenue anyway? Or

[00:37:14] Kalen: like,

[00:37:14] Dylan: you know,

[00:37:15] Kalen: could, yeah, it could be.

[00:37:16] Dylan: They're selling a course.

[00:37:18] Kalen: That could be complete bs. But I, I do, like, I keep thinking of this analogy of like chess masters that can play 30 games blind

[00:37:25] Dylan: mm-hmm.

[00:37:26] Kalen: Simultaneously. And I, because I've never been a believer in multitasking, I just, I, I've, like, for the last decade plus, I've just, I'm all about, like, I've disabled my notifications, like I'm all about deep focus. Mm-hmm. But I just feel like it's changing now because the thing is gonna do a better job than me for the next two minutes.

[00:37:50] So

[00:37:50] Dylan: what you

[00:37:51] Kalen: do

[00:37:51] Dylan: Yeah. For

[00:37:52] Kalen: the next two minutes.

[00:37:53] Dylan: That's true. I definitely feel that. And if you give that break, then it's letting your mind like drift off to like Reddit, oh, for example, it's.

[00:38:03] Kalen: Yeah, it's like, 'cause I found myself going to Twitter too much and you know, I usually try to go a few times a day, but if I'm, clicking a Twitter like every 10 minutes, that's not good.

[00:38:13] And I was like, I gotta find something else to do, like in my little downtime minutes, you know?

[00:38:18] Dylan: Yeah. So I have a team X, window for each thing going on at once. So like the N eight N integrations I was working on, that was going on at the same time as like real ID development at the same time as, , integration adio.

[00:38:32] So like, I'm doing multiple projects that are symbiotic, but they're not touching the same code base so that way I can maximize the time. Right, and that's,

[00:38:42] Kalen: yeah, that's probably a more sane way to do it in general.

[00:38:46] Dylan: Uh, and a real like, if you really wanna focus and get that one app the best and the quickest that the work trees has to be the way to do it.

[00:38:56] Kalen: Yeah. I mean, sometimes what I'll do is I'll do, I'll just be working outta the same, directory, but I'll fire up two different, if I know I'm gonna be touching two different, files or parts of the code base, I'll just run 'em in parallel or, you know,

[00:39:08] Dylan: Interesting. And they don't collide or like,

[00:39:10] Kalen: No, not if they're working on other, on different files.

[00:39:13] No. Interesting. Sometimes there's a common library you weren't thinking about, and then they're, you know, then they're both touching a common library and all confused,

[00:39:21] Dylan: oh man. The scientist minded me is like, oh, you can't do that. Like, your experiments need to be, pure.

[00:39:27] But Right. At a certain point you become like, pretty familiar with the code base and you're like, yeah, these two things do not touch. And your automated tests will prove that in the individual prs anyway.

[00:39:41] Kalen: Yeah. Yeah. And yeah, it's like, I'm spending more time now thinking about like how to understand the code at a, at a high level and then how to verify it.

[00:39:51] How to verify that things are working better.

[00:39:53] Dylan: Mm-hmm.

[00:39:54] Kalen: Do you have a lot of automated tests for your, for your app?

[00:39:57] Dylan: Absolutely. Yeah. I have a ton of integration tests. The acceptance tests are kind of tricky. I'm glad that Shopify just released the ability to, allow a crawler to visit the online store.

[00:40:10] you know, Allowed. I think it's for only, for a short period of time that does not cover the backend of like, logging in, acting as a merchant. Gotcha. Clicking around. Gotcha. I get why like, you, that's a huge security concern. So I'm thinking about after I finish this next project, next side project would be something called like.

[00:40:30] Platoon bridge or mock bridge where basically it would pretend to be the, I frame like the outer I frame of the app bridge and mock it as best as it can provide the same API methods, but allow you to write acceptance tests with a session token that's like, you know, that's fake, but at least your app in development mode could, interact with it like as if it was the real app, as if that merchant's logged in.

[00:40:53] But they're really not. It's just some local note server that's pretending to be Shopify. Right. You know? That's smart. It'd be kind of cool.

[00:41:01] Kalen: That's cool. I did do some playwright stuff the other day with a Shopify admin and it was, it was running like, it didn't run into any blocks or anything like that.

[00:41:10] Dylan: Oh, you didn't have any cap shows or anything like that?

[00:41:12] Kalen: Well, I definitely had to man, log in and then two factor off. , And then after that it just, it like reused , the session or whatever.

[00:41:22] Dylan: Mm. Okay.

[00:41:22] Kalen: And it, I don't know how the chrome it, it would open a separate chrome window when it ran.

[00:41:27] Dylan: Yeah. You can have it share the same cookies, the same cookie jar. Right, right. So I'm overthinking it, of course. Usually as usual.

[00:41:38] Kalen: I don't know if you could realistically rely on that though. Like I feel like it might, you know, but what you're talking about is a more reliable solution

[00:41:47] Dylan: because for, you're always gonna be playing catch up with the latest average changes.

[00:41:50] Like they may add, oh breaking API changes and stuff. It would be a constant.

[00:41:56] Kalen: Dude, you like pain man. Like most people never want to like, they see the app like once every couple years and they just get it to install and they hit a couple annoying errors and they hope to never have to think about it again.

[00:42:10] And you're like, I'm gonna ride a super complicated interface to the app Bridge and maintain

[00:42:15] Dylan: it across versions. I've evolved with the app 'cause when I started the app was brand new and I had to like basically write an off library for it, so mm-hmm I've had the chance to like evolve with it. 'cause at the time they, their template was next js, but the server was coa so they had next JS in the front end and then a COA backend server.

[00:42:37] It was so painfully slow to develop. I was just like, pause, I'm just gonna build a pure next JS implementation because next 11 just came out at that time and it included the server side pages, router. Mm-hmm. Um, so I spent. Too long. I spent too long making a template library. like I've learned with it.

[00:42:59] Yeah. If I was a newbie now, I would not try this. It would not be fun. Yeah.

[00:43:03] Kalen: Right. So you understand it intricately. Like when I run into some annoying average error, like I've gotten blocked on them before, I'm like, I don't understand what's happening. I don't wanna understand what's happening. Like this should just work and be abstracted away.

[00:43:18] But it's, it's been one of those painful debug sessions.

[00:43:22] Dylan: I'm not a master, I didn't write it, but they had really detailed docs back in the day because they expected you to write your own off back then. Right. And they gave you like these flow charts and like understanding when the OAuth stops, when the session and the average starts.

[00:43:39] It was very, very intricate. And since then, they're like, yeah, we can't, be reasonably expecting devs to like know all this, just to get their foot in the door. That's kind of crazy. We're gonna make it even simpler, which is the right move, for sure.

[00:43:51] Kalen: Right, right, right. Absolutely. , So, okay, so you posted that you were working on the hardest, most dynamic application ever.

[00:44:01] What is this

[00:44:02] Dylan: in my history? Yes. I am not like a tenant developer to you app. Yeah, exactly. Calendly Revisited. No. The problem that I faced is over the years I've created all these scripts that manage the app and do things that as the app becomes more complex, your service becomes more complex.

[00:44:24] You need to have these. Ways of like checking accounts or updating accounts or registering web hooks or registering scripts, what have you. It could be Shopify related, they could be database related. They could be some third party service, like an email service provider. So like you end up making like literally hundreds of scripts.

[00:44:46] The problem with that is you're now tied to your laptop and the RV RVing tiny house me that wants to go to the beach, right? Or anywhere. Like all of a sudden you're married to your laptop and this is your full-time thing, which is right, fine. That's like part of the deal. But at the same time, there has to be a better way of okay, here's my remote web interface.

[00:45:11] I don't wanna rebuild it from scratch. I don't wanna use a tool like Retool, which is going back to your Shopify flow description of like a no-code tool where it's like, right. Lots of pointing and clicking. Debugging is a pain. It's expecting you to be not that smart. My idea was it's the worst.

[00:45:28] It's the worst. So my idea was like, why not?

[00:45:30] Kalen: Running into some stupid limitation that like you can only loop over X amount of what? Like, I, I

[00:45:37] Dylan: can't. Yeah. We're, we are too developer. We're just too developer for that solution. We don't have the patience to like sit there and like click, click, click wait, click says, wait, no, we could type or least have Claude or whatever do it.

[00:45:52] We're just so spoiled and, right. Like that. But anyway, like I was thinking, well the CLI thing I, things I've been writing have like prams and they have peram descriptions like shop name. Timeframe from to whatever those can translate into web components. Why not build a plugin for my command tool of choice that reads all the parameters, all the commands, and creates a UI for a web server?

[00:46:19] Then you could host the web server and you can customize it to, I mean, they can log in. You can customize it to fit the layout you want. So like a Shopify plugin, it has the context of the shop that you want to interact with, and all of a sudden it has all of these easy, mobile friendly options to retrieve the current subscription status.

[00:46:41] View hooks, view script, tags view. I mean mean like look up a product, look up a specific order. All these things that are easy to do with a CLI once you have, database in their tokens, right? It's just like a pain in the butt to like do that on the go. Real life, like on a phone or something like that?

[00:46:59] On a phone. On a phone, because your phone is with you all the time. Customers can ping you all the time,

[00:47:05] Kalen: right?

[00:47:06] Dylan: Yes. You could go hire a support agent somewhere. You could do that. Yes. It's be more expensive if someone just built an open source thing. Open source thing that you could self-serve, do a better job.

[00:47:20] Because you know the system really, really, you're the builder, the system making so that you could solve the problem even quicker. . So that's my grand vision. I've, I That's cool. I've started building it, like I've built the, web server and the UI part, but what I want to do is make it like Shopify friendly, where you can pick the merchant and then all those options like that are common.

[00:47:42] Getting the subscription status, like the normal API type things, those are like mm-hmm. Native, and then you can add your own CLI scripts. The framework I use is called, O-C-I-L-F. It's by Salesforce. It's a node CLI tool. , But you can basically register all these scripts with the parameters and options and flags, blah, blah, blah.

[00:48:03] And then you can bubble those up. So if you add your own custom scripts that are not Shopify related, it would be smart enough to be like, okay, these are kind of like in the junk drawer of util and whatever. Right? But there's so much more to it. 'cause then you have to worry about the data, like direct database integrations or API integrations, what's safer, right?

[00:48:22] Do you want an audit trail of these things, these actions that are admin level actions? So like, yeah, it gets complex. Do you want a third party app to integrate your database directly? Eh, I don't know. Maybe,

[00:48:33] Kalen: maybe not. Yeah. Yeah. Ah, that's cool. So you could basically like, ' build out a bunch of those exact same cli I use,, commander is the node library use.

[00:48:43] Dylan: Oh, that's a good one too.

[00:48:45] Kalen: Yeah. And then you can feed it back in, like, I'm running a Claude session and it's testing something and I say, no, it's wrong. And I copy and paste the details so it knows why it's wrong. And I'm like, wait a minute, why am I copying, pasting something from a product page?

[00:48:58] If I build, build A CLI out, I can tell it to use a CLI and they can figure it all and multiply that times whatever, and then they can figure it all out on its own. Or even just for me to test something, like, it's just so much, I, I love it so much. Mm-hmm. Like cli.

[00:49:12] Dylan: Yeah. And you have like a common interface between all of 'em, like dash s It was clearly shop name.

[00:49:18] Yeah. And the pro. But the problem with using CLI only is. It's kind of a pain to like, you're gonna have multiple follow up commands. Like, all right, get the shop details. All right, wait, wait, I need to look at this specific order. All right. Passing the same shop name again, if it was a web ui, at least you can have that context, kind of like in the background, like cached in a way.

[00:49:41] And then it's easy to like, easy to just do this quick one-off kind of operations. So that's, that's my grand vision. I,

[00:49:47] Kalen: that's cool, man. , It's funny, man, like, I keep thinking about like, what are all the invisible like effects of and would you say that like Claude Co and stuff is helping giving you more bandwidth to work on these different projects you've mentioned?

[00:50:02] Dylan: Yeah, definitely. I'm still very cautious when it comes to ai. I don't try to like drive the ball, like I'm just doing like putts and like. Little pitches i'm treating it like the dumbest junior en engineer possible, but this time I let it be like, okay, you cook up the inter, like the web interface with all these components that translate CLI paras into react components.

[00:50:24] Like you go hog crazy on that, I don't care, whatever. Right? Right. And it did pretty well. But when it comes to like the off and things like that, I just don't , trust it as much.

[00:50:35] Kalen: Right,

[00:50:35] Dylan: right, right. But that, that piece right there would've taken so long to like make, to make a, a mapping of, right.

[00:50:42] The various different types of data types to a react component. Right. That would've taken months, I mean like not months, but it taken a long time.

[00:50:50] Kalen: It's certain types of things in particular you go like, I would've spent so much time on that, I just wouldn't have done it. Like, I wouldn't have added the documentation, help text links on this modal, 'cause I would've just made it a basic modal and I would've had the, the bandwidth to think about those little extras,

[00:51:06] Dylan: like progress indicators.

[00:51:08] Yeah.

[00:51:10] Kalen: Like a million things. It's like, I'm like, yeah, that's a great idea actually. ,

[00:51:14] Dylan: Using a lot for the pod too. Like, editing for you? Like beyond just like coding, like is that possible? No.

[00:51:20] Kalen: you know, There's a bunch of tools. , I've tested out some of them. I might not be using the latest and greatest stuff, I don't know.

[00:51:26] But shout out to our editor, rj. He's awesome. He's been great with it. , It took a little back and forth of like, oh, don't do it this way, do it that way, and da da. But he just like, handles it now. And it's kind of cool. But I did, build out a little tool for making clips, which, but it's, it's not fully automated, you know, it's like, I'll have it give a GPT prompt for like, what are some interesting ideas from this transcript?

[00:51:48] Or what are some interesting clips? And then I can look at them and go, okay, I like that idea. And then, and then I can pass in that timestamp that it gave me into another thing that'll generate like a little video. So I've done a little,

[00:52:00] Dylan: oh, that's full, but, so it, like,, I can generate the video itself, like using.

[00:52:05] A Peg F fm. FFM. Peg can make a video too. It's not just for like, yeah. Cutting and all that stuff. Okay. Right.

[00:52:12] Kalen: Yeah, it's wild, dude. 'cause I, I, I used FM peg, like, I don't know, maybe 10 years ago I had some use case. I mean, that was at the time when it's like everybody's making these video sites and it's like, I don't know how they're doing all this.

[00:52:24] And I ran into some issue with it, troubleshooting it, and I just kind of threw in the towel, it's hard, fast forward. Yeah. It's just one of those things, it's like, if your life depended on it, could you figure it out? Yes, of course. But like, the, the priority for whatever the reason is you're looking at, it just wasn't high enough to justify going from like, I'm gonna spend four hours banging my head against the wall to like, I'm gonna spend 25, and then this time around like I mean, I still spent quite a bit of time on it and like, if somebody wasn't, wanting to figure it out, I could easily see them kind of giving up on it. But it was just cool that it could like. Just figure shit out, put all the parameters into FM peg, you know, , the way and then a couple times it has a mistake, you know, you tweak it.

[00:53:07] But it's just cool.

[00:53:08] Dylan: There was another one that was like a little bit older. It's literally react, you can make react, produce video. Really? It's like react code. Oh yeah, that sounds familiar. Yeah, yeah, yeah, yeah, yeah, yeah. I've had this crazy idea once of making like those tick mermaid charts and passing it to this thing make them like mm-hmm.

[00:53:25] , Actually animated,

[00:53:27] Kalen: nice.

[00:53:28] Dylan: I forget what it's called. I mean,

[00:53:29] Kalen: mean's a cool idea. Dude,

[00:53:31] Dylan: back, back then it would've taken it too long, but now

[00:53:34] Kalen: yeah, it would've taken a lifetime back then, dude.

[00:53:37] Dylan: Yeah. Video generator. It's by Facebook. Facebook made it. Remotion. That's what it's called. Make videos programmatically.

[00:53:45] That's wild.

[00:53:46] Kalen: yeah, but I was gonna say, going back to the thing, the tool you're building to like help you, you know, basically deal with your SaaS business, like on the go, it's like, I've been thinking like, what are all the invisible effects of ai? Because it's like, you see the splashy headlines, you see the like, whatever these big acquisitions and things like that.

[00:54:04] And then in your personal life, like you see how you're more productive. Like you're like, oh my gosh, I'm getting done three or four or now I can work on this side project that I wouldn't have had the bandwidth to work on. And then also multiply that times like all the bad guys, all the black hats are also getting done like 10 times more.

[00:54:23] So like there's gonna be a lot like more bad shit happening. And like, that's just like a little example with you. Like obviously, you've been incredibly productive to date, you got, an app that's does well, but it's like now you have the bandwidth to make this thing so that like you can relax a little bit more, when you're traveling and like it's.

[00:54:42] Just cool. you know, it's just fucking cool.

[00:54:45] Dylan: it's , capacity increase, not just for your time, but also the complexity of things you can take on, which is really cool.

[00:54:51] Kalen: Right, exactly. Exactly. And in some ways, I don't even know if it's complexity per se. Like for example, I did like a little iPhone app using Swift , and it's like I never would've done that before.

[00:55:04] And It's not necessarily that it's complex. It would've been hard and time consuming for me just because I'm not used to the language. Mm-hmm. I'm gonna have to learn all the little things of, no, you don't do a bracket this way, you do it that way. All the muscle memory stuff. But I'm like, it's kind of like from first principles, like it's kind of a good thing that there isn't all this friction in the way of like using the language you wouldn't have used.

[00:55:27] Or like if somebody builds FMP, they're building it as a library to be consumed. Like the only reason there's ever problems with it is because there's like leaky abstractions in it and things don't work the way they're supposed to. But like, it shouldn't be hard to use tools that people create to be used,

[00:55:46] Dylan: now's the thing. Like, dude, it's the future. Like right now, we've been optimizing for, for tools that people use, right? Like React is the most person friendly way to make complex applications on the front end, but the, maybe the future isn't people friendly. So therefore like the lowest level language and the most .

[00:56:09] Like hard to use interface, but it doesn't matter because the context window is gonna be so large, you're gonna be optimizing for performance. ,

[00:56:17] Kalen: It's so hard to predict how things are gonna play out in terms of that, in terms of what language gonna be, , what's pay gonna be like, what's blah, blah.

[00:56:25] It's so hard to play those things out.

[00:56:30] Dylan: Mm-hmm. Yeah.

[00:56:31] Kalen: You

[00:56:31] Dylan: know,

[00:56:32] Kalen: it's

[00:56:32] Dylan: like, yeah, it could be a fork. Or not, it could be a hard fork in the road, or it could be React was the last ever human friendly UI component library. You know, Like we start or spelt or like whatever the, the current generation is like, this was the last one we're done.

[00:56:48] Now it's back to the bare metal. Interesting. You know, like, let's go back to vanilla because AI can read it,

[00:56:56] Kalen: And I think this can be true for a very, very long time, even if AI is doing, like, let's say AI is doing, I dunno, 80% of my code, I mean, you know, sometimes you have to spend, I don't know what the percentage is, but when it gets to a hundred percent, , it still is gonna be very important for people to be able to understand the code.

[00:57:15] Unless it goes the route of just verifying the code, like

[00:57:18] Dylan: mm-hmm.

[00:57:19] Kalen: Maybe there's gonna have to be some high level and not even high level, like the correct level abstraction for someone to go into the system and go, okay, I need to understand how this is working.

[00:57:31] That's working. So it's better that it was react than if it was all just vanilla.

[00:57:35] Dylan: Yeah. I guess the future I'm describing is if what in the future it does go to a hundred percent ai. Then you're literally just gonna feed it. Bug reports and bug reports. As long as all the tests pass given these new bug reports, it fixes the bug.

[00:57:51] Know what I mean? Like what does it matter what know goes underneath?

[00:57:54] Kalen: You know what would be crazy if like a new language like document format emerged that was just a product requirements document. Like Yeah. That became like the actual language that like mapped into some super lower level language or something like that.

[00:58:12] Dylan: Yeah, like bubble. All those people got it wrong. It's Atlassians, right? We're gonna , JIRA tickets gets all the way down. It's just gonna be as a merchant, I click this, I click that, I click and it just like, it generates all the stuff. And that's the future. Congratulations product managers, you've won.

[00:58:34] Congratulations. You're on top of the food chain.

[00:58:38] Kalen: And then I'm like, wait. I'm like, are the product managers gonna be doing it all? Or are are the developers gonna have to be, become the product manager? You know, like, uh, yeah. It's so hard to predict how it's gonna go.

[00:58:50] Dylan: you just kind of feel like one as you're like dictating these multiple clot inta instances around of like, all right, kid, these are the three things I want you to do. Very specific. Yeah. Please don't, do it this way. Do it this way. Yeah. Look at this code. you're Like, basically instructing a very, very junior engineer, it could just type way faster than you ever could.

[00:59:13] Yeah, that's the way I treat it most of the time.

[00:59:15] Kalen: Yeah. It's so weird because I don't work with, other, developers too much. And, but, about, I don't know, eight months ago or so, I was on a project where there was a frontend developer doing some work, and I had architected out the thing I wanted to do.

[00:59:28] I was building out the API they were supposed to do on the like liquid and alpine JS side, like in integrating with the a, , the API. So I was like, okay, cool. I'd exploit. I'd be like, cool, here's what you should do. And they would just mess things up, like it was so frustrating. , So then I'd sometimes dip in and do it all myself for one thing, but then I didn't do it for all the whole thing.

[00:59:49] So it was just fast forward today, I'm working on that same , code base now, and it's such a strange feeling to be like talking to Claude about doing stuff in that same code base that my last memory was this other developer. It's the only instance of that for me, where it's like this code base, I have this direct memory of like an actual developer and I'm like, it is so much nicer.

[01:00:12] Working with really lot is. Yeah. Yeah. And it's like, it's a weird, it's a very bizarre feeling.

[01:00:21] Dylan: Oh man. I don't know. I think it's opportunities too for the juniors because it's not so hard anymore because of that, what you're saying. Oh

[01:00:31] Kalen: yeah. Like, yeah. It's a good thing in a way. I mean.

[01:00:33] A hundred percent I think, I feel like if I was, and I heard somebody say it, this effects like if you're just getting outta college, you're super AI native. Whether that's with code, not, with other types of tools. Like if you're three months ahead of the curve right now, like let alone a year ahead of the curve, which is probably most of us in tech compared to the general public, if not two years outta the curve.

[01:00:55] Like the amount of value you can have is off the charts, so

[01:01:00] Dylan: Yeah, for sure. I mean, AI engineer is like, the new is a new job title, which is, oh, is it fictitious? You know, I mean, I'm getting like random recruiter emails about being an AI engineer, so I assume it's a real thing now.

[01:01:14] Kalen: And not in the sense of like working on machine learning. Like in the sense of being a cloud code engineer

[01:01:20] Dylan: type thing. Yeah, yeah. Like quad code engineer or finding ways to. Basically facilitate AI in some process before that involved a human. Categorizations. Categorizations probably the biggest one.

[01:01:34] Like any kind of business process that in the past, like someone had to give some kind of like, that's like as a router basically. , That probably is subject to automation of some sort. Just depends, , yeah. I've noticed if I tell the ai like, gimme your confidence level, it'll always gimme 95. Just like, okay, you have no concept of that.

[01:01:57] It's so bad at that.

[01:02:01] Kalen: It's so bad at certain things. It's crazy.

[01:02:04] It's so weird to me when I think about that. I am writing this huge paragraph of things that I'm trying to visualize in my head how this is a vector database. , I still don't quite understand them, but I saw like a visualization of them and I was like, oh, I think I get it a little bit more now.

[01:02:18] So I try to imagine that, and I'm like, how like, does it, it's insane, dude.

[01:02:25] Dylan: Yeah. It's so many dimensions. I mean, I think the part of the, like hiccups me is like understanding which vector store there's different options to use, like co-sign versus, I'm not even sure of all of them. Like how do you know which one to even use for the problem at hand?

[01:02:42] Kalen: Ah, dude, I've tried a couple. CloudFlare has a little vector, thing that's kind of easy to spin up. I tried some other one. Neither of them did a good job for like podcast search and then I used Notebook LM for it. It was incredible.

[01:02:58] blew me away how good it was. ,

[01:03:00] Dylan: Not just, just like, not just like looking up some kind of like topic and asking to make a podcast, like drop dragging, dropping a podcast in.

[01:03:08] Kalen: Yeah, I drag, I dragged all my episode transcripts in and like, so the, when I was building my own little vector thing, what I basically, an example of a question I wanna ask was, what are some Shopify apps we've talked about? Because I know it's one of those things like we, talked about throughout.

[01:03:22] I can remember a few top of mind. I'm sure there's other I've sort of forgotten about. It's like a cool little list. , And so that was the question I'd ask and it would just give me like five results that like, I know it's missing tons of stuff. Mm. And then I did it on Notebook lm, it gave me five pages of, it must have given me, I don't know, a hundred mentions, which in my head I was like.

[01:03:45] I don't think we've mentioned more than maybe like 10 apps tops, but like, it, there was a ton of them and I was like,

[01:03:52] Dylan: oh yeah, I forgot about that. Oh yeah, I forgot about that.

[01:03:56] Kalen: It just blew me away. Like It's a cool thing to be able to search across like conversation context,

[01:04:01] Dylan: you know?

[01:04:01] Huh. That's really cool. I haven't Yeah. Thought to try that. Like, I tried it originally it was like maybe a make me a podcast about topic, about my industry, blah, blah, blah. Right. Um, is it okay? It's kind of cheesy.

[01:04:15] Kalen: Yeah. I did one on the Shopify forums and it kind of blew me away, but I was like, I'm not gonna actually listen to this.

[01:04:22] You know what I mean? Right.

[01:04:23] Dylan: But I think it gimme the change log.

[01:04:26] Kalen: Yeah. Yeah. I should definitely be listening to the change log. That's a great idea. Is that what you did? Did you,

[01:04:32] Dylan: no, I just literally. As we're talking about like yeah, the community thought about the change log is probably the next best thing to Yeah.

[01:04:40] Tune into. That's a

[01:04:41] Kalen: gr that's a great idea because with the community forums one, like, it would talk about some posts, which, like if you look at the homepage of the forums, you're really not interested in all 20 of them. But like there's probably maybe three to five that are that like, a good audio podcast style overview would be interesting.

[01:05:00] But with the chains log, almost everything would be good. Would be good. It just feels like it could flow into a podcast. Yeah. Really well. But it, the way they conversationalize stuff is it is incredible. I mean, and I can imagine a future where I actually listen to that shit like every single day and that's going to be so strange.

[01:05:20] Dylan: Yeah. I would take that over Most podcasts, I listen to most of 'em are pretty garbage. When you actually take a step back and being like, , this is the news you wanna report right now. Like, does this help me? How I understand

[01:05:35] Kalen: this is a waste of human potential.

[01:05:38] Dylan: I think I want the, the community forums more to be like a notification if there's an outage that's unreported, , check out extensions are down.

[01:05:45] Like, I need to know that right now to let my people know, guys, it's not me guys. It's not me.

[01:05:51] Kalen: Right, how fast do you reply to emails? , Do you have to get back to him right away?

[01:05:56] Dylan: I have this like weird thing with customer support that I try to be on top of it. Very quickly.

[01:06:02] Like, I will answer any email within an hour at Tops, hence me fixating on this. Like, m it's probably, no one else is probably wanting this. They're like, whatcha talking about? Like, why would you wanna be tuned in on your phone? Like, well dude, if I'm at the beach with my family or at the theme park or whatever, like Yeah.

[01:06:20] It's by one device, Yeah. . And I want to solve people's problems quickly. It also leads to op review opportunities. To me it's like, it's a business, thing. Yeah. It's not like I want to, it's that.

[01:06:32] Kalen: Yeah.

[01:06:32] Dylan: Yeah. You know, to get serious. Yeah.

[01:06:35] Kalen: Dude. I mean if I'm a customer of an app, that is what I want.

[01:06:40] Like, it's impossible not to want that. But like, as a person, I, I would wanna try to avoid having a. Do that. But you know, that's my, that's, see that's the thing is that the people that are successful with apps, they're, they're not lazy. you know, you're putting in the work.

[01:06:55] And I, I do think their whole review si situation incentivizes, like, you have to, like, I feel like you build more long-term value, like, without having to worry about the, like, reviews in that sense. But it's such a strong incentive that like, it makes sense for you to prioritize it as heavily as you do.

[01:07:13] Dylan: And then, yeah. It's the best system they've come up with so far. Yeah. It's pretty antiquated, this, this thing could be automated. It could be based off of like average retention. It could be based off of average revenue. They have all these metrics under the hood. It's just that the review system, there's no reason to go off of it.

[01:07:35] And like you said, like it's a powerful incentive for apps to provide support that's above and beyond. Yeah, I'm cool with it. I like providing it, like, I like solving people's problems. I think it's really fun. because again, my fan friends and family have no idea what the heck I'm talking about.

[01:07:50] So when someone's into it as, as much as I am, like, great, cool. I get to help somebody else. It's just a good feeling.

[01:07:59] Kalen: yeah, that's, I, um, I built an app back when I was doing Magento stuff and it, and it, uh, it, you know, it was my full-time thing for a few years but I, like, I got to the point where I would have like, I'd be like, respond within 24 hours kind of a thing, and then mm-hmm.

[01:08:13] I'd do my emails in batches and stuff and, if a bug came up or something,, I'd have to jump on it even if I was on vacation or something like that. But.

[01:08:21] Dylan: How did that work out? Do you notice any like, big difference between switching to that policy? Like was there anything significant?

[01:08:26] Kalen: you know, It was the kind of thing that was like, at the time it was hard to make that decision 'cause I was like, oh, people are gonna get upset, blah, blah, blah.

[01:08:33] But once I did it, I was realized like it was actually fine. It like, didn't really matter. But the big difference there is that we didn't, I mean, we had, there was a Magento like app store. I, I don't think I even got most of my business through that at all. , Because it was different. Like it was self-hosted, so you weren't as dependent on that.

[01:08:50] Mm-hmm. And like now, I fast forward to now like, I want to, that's my goal. I want to get an app going again . But part of me wonders if I have, I really don't want to do support stuff so quickly. , I think I'm not like built for that. And so I wonder if I have the, composition, like psychological composition to like be successful in the Shopify app store because it, it does seem like you fucking gotta be very, very fast.

[01:09:19] Dylan: I don't know, like my app integrates with checkout and it also merchants that have require for every single order or it blocks for fulfillment, and that impacts their customers. Yeah. So, so my app is kind of like mission critical cloud. So other apps aren't, , like the, like some developers I know, like the one I was like the ones I was telling you about, where they can just like not check their email for weeks, right.

[01:09:41] They are like a blog. Integration where like, oh no, the styling isn't exactly the correct. Right, because you have some competing theme thing, right? Is that a mission critical problem? Most likely not. Like one day. One day response time. Totally fine. Gotcha. But if you're touching things that are directly in operations, you could probably charge a whole lot more money.

[01:10:05] I hope you do. Because the burden of support and the responsiveness expectation is like all caps, email, what's your phone number like is going on. , It's just different expectations.

[01:10:22] Kalen: No, that, that makes complete sense. And , I hadn't realized that you were in the critical path of checkout. For some.

[01:10:29] For some, yeah.

[01:10:31] Dylan: That

[01:10:31] Kalen: some

[01:10:31] Dylan: It's before checkout. Some it's after, right.

[01:10:34] Kalen: Oh, man. Okay. Gotcha. . Because that's the whole thing with Shopify is they take away that whole, like with Magento, it's like, checkout's always down. Like every, everybody that did any, work at all was, you know, responsible for checkout issues.

[01:10:45] And it's like, it's kind of beautiful to not have to deal with that, but like, you're in that critical path, which is like, that's a lot to, that's a lot to bite off, I

[01:10:56] Dylan: didn't realize that's probably a reason why

[01:10:57] Kalen: you don't have a ton of competitors. So, you know.

[01:11:01] Dylan: Yeah. It's, it's the very small amount of merchants that need this level of security.

[01:11:06] But when they do, it's oftentimes for, yeah, they have to have it, or, right. It's the stakes are way too high. Like it's a thousand dollars plus piece of furniture or, drones or wigs, hair, wigs, whatever. Or

[01:11:23] Kalen: like THC, weed.

[01:11:25] Dylan: That. Yeah, that's the low, that's the high volume, low cost per ticket., But that's mission critical too.

[01:11:32] Um, right. Yeah. It just depends. Yeah. But I wanna hear more about, like, I remember, I remember speaking with you at like Unite, you mentioned, , the Magento days and like I've heard your, the pod, like in the past you mentioned that you had this streak of having this Magento app. Like what did it do?

[01:11:49] What what? yeah, Tell me the story behind it.

[01:11:51] Kalen: Oh, yeah. It was an email, automation app. So it was like, uh, automated band cart review, request emails and stuff like that. Yeah, I was doing it for, it was called Mage Mail, and I was doing it for, three years or so. , You know, as I was making,

[01:12:05] like significantly more money than I, you know, had made, you know, with, uh, as an employee and it was working out pretty well. And, only hired like a, you know, part-time customer support that was really all I ever did. I wasn't good at really growing it. And I think I got kind of overwhelmed by, technical debt builds up and then you're like, you know, you're like, ah, this part is like, this part of the system is a mess.

[01:12:29] I don't know when I'm gonna clean it up or whatever. Like, okay, let me try to just make a brand new version and get halfway through it and not finish it and not like, , so I ended up having a small acquisition on that. And then I worked on like a hiring marketplace type of a thing for a few years.

[01:12:46] Yeah, so I was like self-employed, like the hiring marketplace was like a. It was like, it was meant to be like Upwork for Magento type of thing. And then his call Commerce hero and then it, , and then it kind of morphed into like just recruiting by email, which was kind of weird because I was like, I can't, getting more and more people wanting hire full-time versus contract roles.

[01:13:05] And then like, I just would like, post a LinkedIn message, get some, , referrals, email them over and just build a credit card manually. And then it was like. At one point doing like, even better than my last app did. Amazing. And I got really like lazy. Like I, I was working like an hour a day, I wouldn't check email,

[01:13:29] and, uh, I would just check email and then that was it. And I stopped coding. Like, I just like stopped wanting to code. Like I wasn't, keeping my skills up. I had no interest in it. I was like learning to surf in Costa Rica and stuff like that. What? That's cool. Um, but then I don't know what kind of happened, but I started like, the business was kind of going down and I wasn't like, I, I wasn't putting.

[01:13:52] There's no thing as pa passive income. You know it, you know, like if you're not putting into it one way or another to grow, like it's gonna kind of die eventually. And that's kind of just what happened. It just played out over a few years. And then I would be like, okay, I need to sit down and, you know, get back focused on this again and get, get this curve going back and the, uh, the upward direction.

[01:14:11] I just didn't care enough. Like I wasn't motivated enough to do it, and I was kind of feeling like depressed. And it was weird, man. Like I, my motivation was like lower and lower and, and this was even when I was making like pretty good money with it. But the trend was going down, you know, it was going down and it had gone down from like 50 grand a month down, , 40, 30.

[01:14:29] And it's like you're making solid money, but you see the direction it's going and you feel like, I felt like I lost the ability to like, like be motivated enough to fucking ramp it back up. Mm-hmm. I think it might've been related to like testosterone. I ended up getting on, uh, testosterone replacement therapy and, uh, which, you know, it helped, that helps with a number of things.

[01:14:52] It helps with , your drive like motivation. It's not just like muscle growth, like it helps with that too. But, I feel like it really helped me. 'cause I think I was fucking like getting to like progressively more depressed and demotivated. , And that was kind of how it showed up in the business.

[01:15:11] Dylan: Yeah. , Like our profession's so funny. Like we were in front of the desk for, you know, hours. Most, like a good part of your waking day is in front of a computer. Right. And you have to make up for that lack of activity, that lack of like pushing your body in one way or another somehow. Right. If you don't Right.

[01:15:31] You know, like

[01:15:33] Kalen: Right.

[01:15:34] Dylan: It's just not good.

[01:15:35] Kalen: You really do. Yeah. I, I, yeah. For so long I would just, work, you know, you're just, you're at like hour number seven at the computer. You're just running on fumes and like, I'm very consistent now of taking those exercise breaks and it just, it consistently helps me like reset my head and get focused again.

[01:15:53] And it's like the idea of living without that seems just insane to me now,

[01:15:59] Dylan: mm-hmm. But these young guns can do it, they can sit that computer for like 12 hours, 15 hours for like weeks of men. They have no hip issues or hip flexors or, you know, like knees, these younguns. Right.

[01:16:13] Kalen: Exactly.

[01:16:14] Dylan: , I

[01:16:14] Kalen: know that's cur how old are you by the way?

[01:16:17] Dylan: I am just turned 35. , Okay. So, yeah. So you are a young gun. I don't feel like anymore. I don't feel like, so, but down here at, there, at the beach, I skim boarded when I was like, you know, 12, 13, and I was pretty into it. Ah, and I picked, I, the house we rented had one in the garage and I was like, oh boy, let's break this thing out.

[01:16:39] The first run I try, I like get my, my, , like pinky toe slash like pinky finger, like underneath my foot. And I think I should have broke it. I, I might've broken it. I don't know. I've, I've definitely messed up a ligament, , oh shit, for sure. But I've just never felt that I've felt that before, ever, you know, all the falls I've taken, I'm like, oh man, I am getting old that it hurts this bad.

[01:17:04] I can't believe I did this, but I kept on doing whatever. It's fine. I'll figure it out when I get home. But like. Yeah, I was, I'm like, oh man, I'm fragile now. Like I can't just think that's a good, right. Yeah. I'll still, I think, I think I've maybe got like three more years of this left in me, but I wish I would've known from like 20, 21 to now this window is only so big, man, you can only do so much next or like skateboarding or whatever, and then it's right.

[01:17:30] And then the stake are pretty high. What, how

[01:17:32] Kalen: did you do,, Did you do that type of stuff? Like

[01:17:34] Dylan: MX? Yeah. Yeah. I did VMX. , I raced VMX. I broke both my arms. Yeah. 'cause there's a Akron BMX, Akron, BM X's, um, that race track, my dad and I did it together. Yeah. Now my son's doing it, so I'm taking him down. It's pretty awesome.

[01:17:49] So I take him down, dude. That's so cool, man. Yeah. And then snowboarding, skateboarding, ski boarding. I'm surfing. Nice. Surfing is so funny to me 'cause you're like driving with your left foot, which is the exact opposite of snowboarding. So I've only done it once.

[01:18:05] Kalen: You go, you go snowboarding a different direction than surfing.

[01:18:09] Dylan: Well, I mean, it's like when you're, like, when you're turning or carving, you're like, I feel like you're pushing with your back foot. Like you're pushing and pulling. Oh, I see what you mean. When surfing I, I surfing feels like to me, like , you're using your left foot if you're regular footed, like you're kind of like pushing and pulling, , just the opposite foot than what I, is natural to me.

[01:18:30] Kalen: Dude, it's so funny 'cause I literally just like, I'll, I'll visualize remember like when I surfed and stuff like that and, and like, for like, from time to time. But I just tried to visualize like how my feet feel , when I'm surfing and I can't remember like the actual, because it's been like emotion half or two.

[01:18:48] You like, I can't remember the sensation on my feet because it's been like, I've been using my one, I use my one wheel a lot.

[01:18:55] Dylan: Oh, you mean like , the, like you're leaning forward like at the one wheel thing?

[01:18:58] Kalen: You should try that if you never try. They're fucking amazing.

[01:19:01] It's the closest thing to snowboarding on cement. I, that I could ever imagine.

[01:19:05] Dylan: I tried it once. I didn't take it to the limit.

[01:19:07] Kalen: I could see how It's really fun, dude. Yeah. But I, I like, I, I use that a lot. I can feel the sensation right now if I try to think about it, but it's been so long since I surf, which sucks.

[01:19:16] Like, I, I, oh God, I want to just, I just wanna disappear dude, for like months

[01:19:22] Dylan: then like do they have, in Austin? Does I, I think you're in Austin. Austin, right?

[01:19:25] Kalen: Yeah.

[01:19:26] Dylan: Oh, is, do they have like indoor, um, you know, like the wave, like the wave generator things

[01:19:30] Kalen: there? You know, I think there used to be one in Austin that closed down and there's one in Waco, but I haven't been up there.

[01:19:36] But I'm planning to, , planning to get up there. , Did you meet William Belk at additions. , he's cool man. He's actually just moved to Dallas from, uh, like, uh, California or something. And anyways, we talked about meeting up at the Waco, 'cause it's like halfway between us.

[01:19:53] , We talked about meeting up there to go surfing. Just made me think about it.

[01:19:57] Dylan: That'd be awesome. I mean, Houston's probably still pretty far from you. I'd imagine. That's the closest beach, Sean, right,

[01:20:04] Kalen: right, right. And the waves there are, you know, they're, they're not, yeah.

[01:20:09] Dylan: Oh, no. Okay.

[01:20:10] Kalen: Yeah. Unfortunately they're like not great.

[01:20:13] Dylan: Well, I have Lake Erie, so like we can, you can surf it in the winter. That's the only time you can surf. Really? Yeah. There's winter surfers in the Great Lakes. They should see the pictures. They have like icicles coming off their beards. They're coming outta the water. That's

[01:20:26] Kalen: insane.

[01:20:28] Dylan: How big are the waves?

[01:20:30] They can get up to like four or five feet, depending. No shit.

[01:20:34] Kalen: In a lake.

[01:20:35] Dylan: Yeah. And it's a lake. that's the size of, you know, a sea. It's huge. , That's crazy. You can also, I mean, like after a, during a storm you can surf it. It's just like very dangerous. Like the riptides are still there.

[01:20:48] They're real, and on top of that, like, this is kind of gross, but all of the, um, sewer systems, if they get overwhelmed and it, during a storm, they dump into the lake right where you are, surfing. Wow. So if you fall, paddleboards safe, paddleboards is more safe in general. Right,

[01:21:06] Kalen: right.

[01:21:06] Yeah.

[01:21:06] Dylan: I've been

[01:21:06] Kalen: paddleboarding lately and it's, it's fun, but, you know, it's just, it's fucking slow, you know, like. Core, I wanna move faster. Right? Yeah. Good core. That's, that's what I'm mostly excited about is my, my lats and my core.

[01:21:20] Dylan: Like, that's why, you know, you're getting old, like, oh yeah, that's great.

[01:21:26] Yeah.

[01:21:27] Kalen: But I, I do enjoy going out. I just wish I could go about four times as fast.

[01:21:32] Dylan: What about those hydrofoils things? Those like surfboards that like ride above

[01:21:36] Kalen: Dude, I know a guy that has a rental place and I, I wanna rent one of those, to, uh, to use out here on the lake. , So dude, that, that would be so fun,

[01:21:45] Dylan: dude.

[01:21:45] There you go. Do that one day. Yeah, it's close, . That's fine.

[01:21:48] Kalen: Yeah. I think it's like three 50 or 400 bucks an hour or something like that.

[01:21:52] Dylan: They're probably brand new, a thousand. So if you like it, then it's pretty easy to like get a new one.

[01:21:58] Kalen: I'm pretty sure that I looked it up and I'm pretty sure they were like eight grand minimum.

[01:22:03] Oh really? For fucking, yeah, dude foil, electric foil boards. Yeah, dude. Oh, I mean, I don't know, maybe, maybe they're cheaper.

[01:22:12] Dylan: It'd be direct to Audi Express or something that's a little bit cheaper. Right, right.

[01:22:18] Kalen: Yeah. It is crazy how some of these things that they're doing direct sale from TikTok or whatever, like they're charging like, I don't know, like three or for like, like 4% of the price are like, less than 10% of the retail price.

[01:22:32] And it's like, how do they have these margins? This is crazy.

[01:22:35] Dylan: Yeah. Yeah. This whole paradigm, it's kind of built on the retail being at the, , like in the language and the locale of us. But in the marketing of us,

[01:22:50] Kalen: yeah,

[01:22:51] Dylan: it's, I'm just surprised it's taken that long for direct to come up, it's like the internet's a new thing.

[01:22:56] Kalen: Yeah, it is kind of surprising actually. , I guess it just takes, and it's funny 'cause even now, like there'll be some like, , TikTok or it's like a Chinese lady, like singing a song and she's like in a factory that sells tires or something like that. And she's like, we sell tires.

[01:23:14] it's, and I, you, You cannot tell if it's a joke or not, but like, regardless of whether it's a joke, it, it, like, I'm sure it works 'cause it's driving eyeballs to them.

[01:23:25] Like they're at tire manufacturers trying to sell direct or something crazy like that. So it's, it's insane.

[01:23:33] Dylan: I love those honest marketing ploys, like the, the real estate, the newest trend of real estate agents going into houses being like, you can't afford this. Why are you watching this video? This piece of shit wants to charge you $700,000 for this kitchen has been updated since the nineties.

[01:23:47] You're like, whatever. You know, they're just going through just like ragging on it. Of course, price sells next day. It doesn't matter. Right. It sell.

[01:23:55] Kalen: Right. I think I saw one of those. I think I did see one of those. Yeah, that was a trip. Yeah. I've seen some weird, like real estate related little TikTok videos where there'll be like an agent doing like some goofy thing or something like non-standard.

[01:24:11] Anyways, man, I think we're about at our time, dude, but, this is cool man. What do you, what do you got going on for, uh, the rest of the weekend?

[01:24:19] Dylan: I mean, tonight's our last night, so we're gonna hang on the beach for a little bit and then fly back tomorrow. Back to reality.