[00:00:00] Kalen: I actually saw this tweet. It's a little bit messed up, but I think they fired like a hundred designers at Google Cloud and, oh really? Somebody tweeted and said they had designers at Google Cloud.

[00:00:21] This episode is brought to you by smile.io, the loyalty platform trusted by thousands of the fastest growing Shopify Plus businesses to increase repeat purchases, reward VIPs, and build lasting customer relationships. Smile helps businesses become more profitable while discounting their products less.

[00:00:43] You gotta be careful with the discounts. You don't want to push too hard on the discount pedal. If you're ready to turn transactional sales into lifetime customers, start building a more profitable business with smile.io today.

[00:01:03] David: Dude, happy Friday.

[00:01:05] Kalen: Happy Friday my friend. Happy freaking Friday. We made it. How uh, is it as Fridays go? Is it a good one? So you sound chipper.

[00:01:15] David: Yeah, it's good Friday, a little chipper. I was up late last night hanging out with Claude, and I feel like I got some cool stuff done, so that always makes me feel better.

[00:01:25] Kalen: Isn't that funny how if you're not feeling productive, you just. You could be like relaxed, you could be money, could be fine, like everything, but if you don't feel like you're doing stuff and then when you do, you're like, that feels great. And then you're like, yeah, I

[00:01:39] David: hate

[00:01:40] Kalen: it.

[00:01:40] David: Why can't I just be lazy all the time and feel good?

[00:01:44] I know.

[00:01:45] Kalen: That's cool, man. What, uh, what were you, uh, cracking away on?

[00:01:50] David: Uh, I was trying to, uh, like for way too long I was copying analytics, like as a CSV and a Claude and being like, Hey, help me, help me figure out what's going on. So I just bit the, like, I don't know why it took me so long to do this, but it was just like, you know, cloud code can use the GCloud command and run queries directly and start data warehouse.

[00:02:14] And so that's what, that's what was going on. Nice. It's kind of hilarious though 'cause like every time Claude does a new poll and like gets some data, it's like,

[00:02:23] Kalen: oh,

[00:02:24] David: I see what the real problem is. It's just like, it's like, wait everything else that we already looked at, doesn't matter. This new thing is the thing we need to focus on.

[00:02:36] It's kind of a, a little a DHD, but

[00:02:38] Kalen: I know, dude,

[00:02:40] David: I gotta start playing with Claude MD and all the stuff. 'cause I wanna try and make this Oh yeah, like headless, you've done some, like just let it ride on a virtual server somewhere. Right? And it just runs. Oh yeah. I wanna try and do something like that. Nice.

[00:02:54] Kalen: Nice. No, the GCloud thing is frigging magical. Like, especially because I hadn't used Google Cloud. I hadn't used AWS, you know, I'd always used like hosted things and stuff and, and then I got had to do, so anytime I go into the Google Cloud interface. It is so confusing.

[00:03:17] David: Yeah, there's a lot of stuff in there.

[00:03:19] Kalen: Oh my God. And like I actually saw this tweet. It's a little bit messed up, but I think they fired like a hundred designers at Google Cloud and Oh really? Somebody tweeted and said they had designers at Google Cloud.

[00:03:37] A hundred of them. No disrespect. I'm sure anybody that was a designer at Google Cloud is gonna be fine. They're gonna find another gig, so we're not too worried about 'em. Good luck out there. Good luck out on those streets. But it really is horrendous. Like it's so, it's so confusing. And then, um, I bet

[00:03:57] David: it's hard.

[00:03:57] Like it's, they're like, find a way to cram 1000 things into this one view. Yeah,

[00:04:02] Kalen: for sure. No, no, no. There's, yeah, there's so much complexity there. Yeah.

[00:04:07] David: Yeah. Like I was in, so Yeah. The GCloud, yeah. Like that command is like, it's, it does

[00:04:11] Kalen: everything. It's awesome. It's so cool. And it'll just like figure, and it's, it's stringing together like GCloud commands with like 19 command line arguments.

[00:04:23] That, like you just know, typing that in would take 20 minutes and it just like,

[00:04:30] David: it'd take me 20 minutes to research all the options. Oh

[00:04:33] Kalen: dude. Yeah. I mean, but you know, the whole point of all these things like Google Cloud, it's supposed to be an abstraction that makes it easy to do X, Y, and Z, but then all the stuff, all the configuration gets in the way.

[00:04:49] So it, it's like, I don't know, you kind of feel like you're cheating, but then I kind of feel like this is how it should have been all along. Like someone builds this super complicated hosting, a bunch of hosting abstractions, they're doing that to make your life easier, like to take them off your plate, you know?

[00:05:07] Yeah. The experience matters. Yeah. Nice man. So you ran some GCloud stuff to what, what was it doing? It was configuring some of your analytics stuff.

[00:05:18] David: It's just connecting directly to our data warehouse and running like, uh, I'm trying to tie orders to Google Analytics events to track like orders by like new versus repeat orders by channel.

[00:05:30] Oh, right, right. And the interesting thing about using Claude Code is like, I'm starting to realize all those stuff that I've skipped to like just get it to work. Mm-hmm. Like there's, uh, like you start talking to it and it does things a certain way and then you're like, I actually want it to be this other way.

[00:05:47] Or you can group these things together, you're like this. And then every once in a while I'm like, all right, let's like put all the stuff that we just did into the Claude MD so that you don't forget. Yep. And I'm just thinking like, man, I gotta get good at making Claude MDs. I gotta figure out, like, there's things that happen, like kind of, I don't know what the word is, but like you'll all of a sudden find yourself in the situation where, for example, it would be running, uh, SQL queries directly with the GCloud app.

[00:06:17] Right. Or, uh, CLI. And then I'd be like, actually I want to kind of like track over time. Where this data is coming from. So let's keep, like in this code base, let's keep a list of all the SQL that we've run, and then name the, the outputs, the exact same thing, just like csb. Nice. So I can go back over time and look at them.

[00:06:36] Nice. And then eventually what it started doing is instead of writing GCloud. Commands with all the SQL in it, it would write the SQL file and then run the SQL like GCloud run and then reference the file and then output the, so it's, it started to get like more efficient over time. And I'm trying to figure out what's like, what's the best way to realize that that's something that happened and you should capture it and what's the best way to capture it.

[00:07:00] Right, right.

[00:07:01] Kalen: Yeah, dude, that guy Stewart had tweeted one time Revo dude, that he created an a Claude agent, that that's whole purpose was to update the cloud files after doing stuff. And then he would say, okay, I want you to remember what we just did for X, Y, and Z. And then I started doing that. It's pretty cool, but it's still not perfect.

[00:07:26] And one of the crazy things is I feel like Claude is actually pretty bad at updating Claude files. Mm-hmm. Because you could do 'em per subdirectory and things like that. And it's like. Compared to how good it is at all sorts of crazy complicated code. Like just getting it to like put something in the right subdirectory.

[00:07:47] It'll take me like a couple times and, and then also it'll do things like you'll go, you know, you'll say, okay, I want you to remember to do this, and it'll add like 30 lines to the clot MD file, or 50. It'll be like, here's an example of what to do. Here's an example of what, what not to, which is probably good, but then yeah.

[00:08:06] If it does that. And then if you do that a few more times, it's probably too bloated. And then you need, you need it kind of trimmed down. 'cause I'm sure it can only handle so much context, like really well.

[00:08:17] David: Totally. Or by default it'll read like the top 100 lines of that file and not actually know what to do.

[00:08:23] Kalen: Is that what it does? Yeah.

[00:08:26] David: Yeah. I've seen it do that a bunch of times. It'll be like, all right, let me look at this. And it runs head 100 or whatever. Oh, right, right.

[00:08:33] Kalen: That's crazy. And then it's one of those things, when you do it, it's like it gives you such a huge improvement, but then sometimes it's hard or it's hard to remember to do it, or you kind of don't know exactly how to do it while you're coding and stuff.

[00:08:46] If you don't do it in the moment mm-hmm. You're definitely not gonna do it later.

[00:08:49] David: Right. Um, yeah. Which is in itself like a new kind of discipline, I guess, where you have to like,

[00:08:56] Kalen: yeah. Yeah.

[00:08:57] David: I'm, I'm making fast progress and it feels good, but I also need to like. Chill and make sure that we're, yeah, like this magical context that we have right now.

[00:09:05] I need to figure out how to capture it. Yeah.

[00:09:07] Kalen: It's so weird how like there's new things that are hard to do that like saying they're hard. Sounds so silly. But they are somehow. And so like people that are doing all of those things like optimally are, you know, having much better results. And I, I kind of think that's just a glimpse into the future where like you're just gonna have more and more leverage and there's still gonna be people who take better advantage of the leverage that the tools give them.

[00:09:37] And then there's gonna be people that just like when we're just being lazy about it and then like writing the same dumb pattern three days in a row. Like, you know what I mean? Like I, but I do think everybody, it raises

[00:09:48] David: the skill ceiling. Yeah. And so like everyone needs to kind of move up a little bit, which meets everyone needs to learn some things.

[00:09:54] And if you don't then

[00:09:56] Kalen: Yeah.

[00:09:57] David: Low average bro.

[00:09:57] Kalen: Yeah, exactly. It's crazy, man. Like I've been doing a Google Cloud deploy recently, uh, for a new app and it was pretty good. It knocked out the deploy, but then it was like slow. And then there's so many different services, you know, there's different ways to do it.

[00:10:16] Artifact registry for like the container image and you can, it'll use one thing for the bill, uh, service or pipeline or whatever. Then turns out there's another thing that I actually got a reply from the freaking head of Google Cloud run on Twitter and Oh, that's cool. Uh, you know, 'cause I just posted, I was like, is it normal for like, like three minutes?

[00:10:39] Then he actually replied and said that they're actively working on, on improving the deploy times. But he said that specifically there's one way to do it where it's deploying the from the source code versus doing an image, a docker image, which is, I think the more common way is to do the docker image.

[00:10:57] That's what I'm doing. So there's another mode that does it directly with the source code somehow. I don't know. And um, that's

[00:11:05] David: interesting. Yeah. Like you bypass the anything about containers or something.

[00:11:09] Kalen: Yeah. And so you don't have any docker file at all? I, which I just started reading about it a little bit and how it works, and I'm just like, okay.

[00:11:17] And I think I actually did try it once with Claude. I think it was one of the things that was suggested, but it didn't work. So I was like, okay, let me circle back to this. But I think I definitely need to circle back to it now, because, you know mm-hmm. It seems like that's gonna be the official way to do it.

[00:11:32] David: Yeah. It sounds less complicated.

[00:11:34] Kalen: Yeah, I'm, but yeah, I guess it, yeah, I hope hopeful it'll be less complic.

[00:11:38] David: Maybe it's not.

[00:11:39] Kalen: Yeah, I know, dude. And then I had to spin up a local Docker, because I usually use SQ L light locally, and which works great with CloudFlare because they have a SQL Light database, but Google Cloud doesn't have a SQL Light option.

[00:11:57] Do they have their own Google version of SQL Light? I don't know. I is. I did a quick search. It seemed like they just don't have it, but So you wanna do Postgres and then one of my migration scripts was failing because it's not valid for Postgres. So then I had to change the code and I gotta change the code, then I have to change my whole local environment.

[00:12:15] So I was like, I gotta, instead of doing local Shopify app dev, now I have to do a Docker. So I started doing a Docker, and then it was like, well, why don't you just use Docker for the database and continue to do your app dev? Your thing. And I'm like, okay, fine. That works for me. And it's been, it's been working.

[00:12:31] Like Claude was Who said that? Yeah, yeah, yeah, yeah. Anytime I say, okay, I read something, I learned something, I saw this. I'm talking about chat, GPT and Claude. That is the, well, or if I mention a tweet, but yeah, I'm not reading Wikipedia. I'm not like reading any technical documentation whatsoever. It's fucking amazing.

[00:12:59] But it is pretty sweet. It's pretty cool. But you know, you're still responsible for it's right or wrong. And if it's gets something off, then you know, you gotta double check it. You gotta spot check it. You got a sanity check it. Mm-hmm. You know, like it's, yeah. There's still a lot of ways I don't, I do find myself

[00:13:14] David: doing that a lot.

[00:13:15] Yeah. Like it'll, it'll output some stuff and I'm like. Uh, these two things that you made kind of conflict with each other. Yeah. Can help me resolve that? Yeah. That's pretty good at figuring it out once you, once you point that out. But like, you gotta know that that's wrong, Claude. Yeah. You gotta know that.

[00:13:31] Yeah.

[00:13:32] Kalen: Yeah. I know it's funny sometimes of like typing and I'm sort of trying to visualize the vector space or whatever of how, of how this all works. Like I've seen a couple visualizations of it and I'll, I'll be like, oh, okay, I gotta get it. And it's like a multidimensional vector space or something.

[00:13:49] Yeah. And I'm just like with like lightning shooting through it. Yeah, exactly. And I'm, I'm writing a prompt and it has, it has, you know, it's a run on sentence. It has like, but actually maybe this, and then da, da da, da da. And I, and I'm trying to visualize this as a vector and I'm fucking, it's, it makes no sense.

[00:14:11] It, it's, yeah. Unreal. No, it doesn't.

[00:14:13] David: Yeah. It's crazy. Yeah. It's just like unfathomable size of this brain thing. It's so crazy. I've been watching the Sora stuff that's been coming out. Okay. And like now people like, you know how they always try and jailbreak models? Yeah. I saw a guy who's trying to jailbreak Sora.

[00:14:30] Right. And they do the same kind of stuff where they like make up these, well, they're not made up tokens, but they're like tokens that they know OpenAI uses, like open bracket pipe, SOS like pipe, and then close bracket. Mm-hmm. That's like, uh, start of stream or something like that. Mm-hmm. So that's how they kind of like unlock it mm-hmm.

[00:14:51] And make it do stuff. Mm-hmm. So someone told Sora in this way to like output their system prompt.

[00:14:58] Kalen: Mm-hmm.

[00:14:59] David: And so Sora made the video and it was like on a chalkboard. It like wrote its own system prompt. It's like, this is just so wild that like the stuff that people are doing poking at these things and it just like deals with it.

[00:15:12] Kalen: Sora did an image where it put its system prompt on a chalkboard. Yeah. Because like, have you seen the Sora two stuff that's,

[00:15:21] David: that

[00:15:21] Kalen: came out this week? I barely seen a few, like, I don't know, image video clips, but I haven't really dug in too much into it.

[00:15:30] David: They're incredible. The best one I've seen so far is SpongeBob getting pulled over and he is like, ah, I can't get a ticket.

[00:15:36] And he just like speeds off and pulls Chase after.

[00:15:42] Kalen: But is it like, it looks

[00:15:43] David: so good. It does. That's awesome. It's like, it looks almost perfect, but yeah, they did that to, so they like tried to get sort of output Its system prompt and the way it did it is by writing it on a chalkboard and, and it's like kind of scribbly in the middle, but you can see like the starting and the ending are a little bit more clear.

[00:16:00] Kalen: That's so crazy. God, that's insane. So I did try to click a link and log in, but it was like a wait list thing or whatever. Are you just seeing them online or are you Yeah, I'm just seeing 'em on

[00:16:12] David: Twitter. I wish I could get in.

[00:16:14] Kalen: Okay. But

[00:16:15] David: I have not gotten a, an invite code yet.

[00:16:18] Kalen: Dude, it's so crazy how fast this is all going, dude.

[00:16:22] I know. It's like there's now there's another like insanely crazy OpenAI, you know, feature release, release, product release. It's like,

[00:16:32] David: yeah.

[00:16:33] Kalen: That's just so

[00:16:34] David: crazy. And it's like a whole, like it's a new social media app, which we'll see how it goes, but it seems like there's a bunch of activity in it.

[00:16:41] Kalen: Dude, I kept thinking like, I was trying to figure out if there's a way for chat CBT to become like a, a feed somehow.

[00:16:49] And I didn't even know what that meant or what it would look like, but it just felt like you wanna spend more time there and it can figure stuff out or, or maybe you see a feed of what prompts your friends are doing. I don't know. That's probably dumb. Oh yeah. But like that's so crazy that they did a social media app.

[00:17:06] I can't imagine scrolling on AI generated stuff and actually being into it, but like it probably once it happened, like you don't realize it until you do it and you're like, okay, I'm gonna just use this from now on.

[00:17:20] David: It's weird. It's like it's not a replacement, it's like a different thing. Like you, you know, it's AI generated, but also it's pretty hilarious, right.

[00:17:27] What people come up with. Right, right, right. It's a new kind of

[00:17:31] Kalen: thing. That's the thing. It's like a new canvas for creativity and it's like mm-hmm. Just like memes were like, memes are not, like, it's not a painting that you have to spend hours and hours drawing. I mean, it's a copy paste. True. But it became this canvas where like a really funny meme is just like amazing and like there's people that, that's their skill, you know?

[00:17:54] David: Yeah, it definitely is. And you can like, really, you can actually impact people's, like emotions with memes, right? Like mostly just laughing, right? I mean, that's pretty cool. Or sometimes cool, feeling

[00:18:04] Kalen: sad. I mean, I, I think, I don't know, maybe, but yeah. Yeah, they, I don't know. They capture some like, like cross section of your memory in a unique way when they take a, like, you know, a movie you knew or character, a scene and they smash it together with something else.

[00:18:21] It's like, is doing something in your brain like that makes it,

[00:18:25] David: it's pretty, it's pretty hilarious. Yeah. I just sent my friends one that someone posted of Lord of the Rings, but more Gen Z

[00:18:35] Lord of the Riz. That's so funny.

[00:18:40] Kalen: That's hilarious. I'll send it to you. So I saw this tweet. This is so funny. It was like from an interview that Toby was in, and it says that for the last, I think 10 years or maybe longer, he's been using a key logger to record every keystroke and taking an auto screenshot of his screen every 15 minutes.

[00:19:04] David: That's incredible. Isn't that crazy? That's like the most nerd thing to do ever. Yeah.

[00:19:10] Kalen: Yeah. I only heard of one other person that did that. He was, he's the guy I think that created Mathematica or something like that. I forget his name, but he's like a super programmer genius. But he's been doing his keystroke recordings for like 30 years, or 50, some insane number.

[00:19:26] I think Wolfram is his name. Steven Wolfram. Wolfram or something like that. Yeah. He's like a mega genius person. Did Wolfram Alpha. Yeah. Yeah. Yeah. Dude, I was browsing Linus Val's GitHub account today. Tell him what you got. I just, no, I just saw somebody tweeted. They're like, how is this guy still working so hard?

[00:19:47] And it was a screenshot of his profile with. The charts that show you how much you're committing The heat charts. Yeah. Of how much you're committing. And it's like, it's obvious that guy's a super freak. He's just gonna work on crazy shit like for the rest of his life in terms of

[00:20:02] David: COVID. Yeah. It's his hobby.

[00:20:03] Kalen: Yeah. So I click on it. Obviously he has Linux, like one of his repos is Linux. Yeah. Which is like, that's so, and then you need Linux. And then I go, wait a minute. He also invented Git. So like I'm on GitHub. Yeah. Which is one of the biggest things in the world for programmers. And I look at his profile, its like, oh, by the way, he actually invented this whole thing.

[00:20:30] Yeah. It's

[00:20:32] David: so crazy. Is he like committing stuff on the daily or what's he, what's he been up to? So,

[00:20:38] Kalen: okay, so there was another project called Sub Something and I was like, what is this? Some library for Linux, something like that. It's a software for tracking, uh, dives, like scuba dives and it lets you track.

[00:20:49] Where you are, your location. I, I'm not sure what all it does, but I'm like, oh, he must have gotten into scuba diving and then he fucking built this app for it and then he had like some kind of guitar pedal hardware integration library or something like that. Just like

[00:21:09] David: did

[00:21:10] Kalen: he

[00:21:10] David: Wild stuff. Adding the firmware

[00:21:11] Kalen: on

[00:21:12] David: his, that's awesome.

[00:21:14] Yeah. Yeah. Sounds like he's a tinkerer. Pretty cool. Oh

[00:21:17] Kalen: yeah. I mean probably greatest of all time. Yeah. Apparently he's a bit of a bit of a jerk, but it's like, what are you gonna do, man? You can't be that brilliant without having a few screws loose. You know what I mean?

[00:21:30] David: Yeah, no, I agree. It's like what I was thinking about that like some of the people who are doing crazy stuff or like the coolest stuff are also just a little bit weird and like sometimes that makes it hard for them to make things happen because they can't get along with other people, so it's.

[00:21:45] Yeah. Unfortunate.

[00:21:47] Kalen: Which by the way, I don't have that problem at all. I get along with great with everybody. I'm not hard to get along with my life at all. I'm realizing recently that like I don't get along with anybody. Like I'm always fighting with my wife. Like whenever I have a, like a coworker or like a person, a client that I spend more than like 10 minutes per day with, I end up getting into awkward arguments with like, like if I see someone once in a while, it's like our two week thing is.

[00:22:21] Incredible. You know what I mean? Because I have no bandwidth to be like annoying. But like we did that one thing where we spent an extra hour planning out that app idea. And then I got all tense and I got all like, that's true. I did right. I was like, so do you wanna do it or not? What's the feature set?

[00:22:40] And I was like, I'm like, dude, you're being such a dick. Like, but it's like I can't help mys, like if I'm into something, you know what I mean? If I'm trying to figure something out, get something done, you're like really

[00:22:50] David: into it.

[00:22:50] Kalen: I can't help my, you know, I can't be like chill. I just can't. And I wish I could.

[00:22:56] That's how you know you're a genius. No, it is. It's something's off, but it's, it's not off in the right direction. That's, that's, that's correlation. Not, not cau issues, but yeah. Nah, that's crazy. Oh, are there any AI tools for doing wire frames? Because I've, I tested one. One person mentioned one. I don't know if Figma lets you AI wire, like prompt wire frames, but it feels like this should be a thing for, like, I had a feature ca set a feature come up and I'm not, I don't normally don't wanna do wire frames.

[00:23:31] I used to do this back in the day, but like sometimes you just want to see like five screens, what's going on where, and that seems like the type of thing you should be able to prompt really well somehow these days.

[00:23:43] David: Yeah. I gave V zero, I think is what it's called, a try. It's from. Verse, they have this thing that you can prompt and it'll just make a, a mockup.

[00:23:52] Kalen: Uh,

[00:23:53] David: it was all right. I think a lot of people use that, but yeah, Figma does have one. I haven't tried it yet, but

[00:23:58] Kalen: does V zero just do mockups or does it build out actual apps?

[00:24:02] David: Oh, I don't know. I thought it was a mockup thing.

[00:24:04] Kalen: I'll have to double check on that. 'cause I, I had always thought it was a, an app builder.

[00:24:09] I did find one that was kind of cool. I mean, I'm sure they'll get there. It's just sometimes you don't need it really. I mean, and people that wanna wire frame every single feature. It's a little much, but there's certain things Yeah. That you're kind of talking 'em through and you're like, we really need to just have a thing that we can look at so we can all get our brains in sync.

[00:24:30] David: I feel like Figma does that the best. 'cause they, you can also like tie together different states and like pretend that it's actually an interactive thing. Mm-hmm. Well, I mean it actually is, right?

[00:24:41] Kalen: You're like, except you don't even have to pretend

[00:24:45] David: it is interactive. It is, yeah. But it's not like on the site or whatever.

[00:24:50] Yeah. But yeah, I gotta go see if Figma, 'cause I saw some tweet go by about Figma having that, but I haven't given it a shot.

[00:24:56] Kalen: Yeah, I know. I think there's a Figma MCP and I'm like, what does that mean exactly? Maybe that's exactly it. Maybe there's just a CLO MCP and it talked to a Figma file. If so, that would be perfect.

[00:25:10] David: There are some like page builders that go, like if you've got a Figma artifact, you can pipe it into, I think it's called Instant, and it'll create mm-hmm. Like a Shopify page out of it. Mm-hmm. Mm-hmm. Which I think we tried it and it wasn't super smooth.

[00:25:27] Kalen: Ooh, the negative, negative review for

[00:25:30] David: Oh yeah, negative review

[00:25:31] Kalen: for instant.

[00:25:32] He's a Twitter friend of ours. Oh, is he? I think his name's co something with a CI think he's a Nordic gentleman. Yeah. I remember seeing when he launched it, maybe like a year ago or something like that.

[00:25:44] David: Yeah. There's some cool, cool page,

[00:25:46] Kalen: but you're harsh and you just harsh in his vibe.

[00:25:49] David: I'm harsh and on it, I'm harsh and on it.

[00:25:51] I don't remember what the reason was. I mean, but I remember it like didn't really fit in well with our workflow, but that is the dream. Like let me generate a Figma and then let me put this Figma just in Shopify. Just do it.

[00:26:06] Kalen: But do we really need that step? If we could just prompt Shopify and see like a preview of it.

[00:26:13] Oh yeah. What? I mean, do we even, that'd be the best. Do we even need Figma in this mix? Ultimately?

[00:26:19] David: I mean, if it's fast enough to create a usable artifact, then yeah, you could totally skip wire frames. Like why would you need a wire? The reason you need a wire frame is so you can talk through something.

[00:26:30] Yeah. And like be helped to imagine what it would be like.

[00:26:33] Kalen: Yeah.

[00:26:33] David: But I feel like, like you're right. You don't really need to imagine like you just make the thing.

[00:26:39] Kalen: Yeah.

[00:26:39] David: Maybe it's not as robust as it would be in production, but like, yeah, just build the HTML and go.

[00:26:45] Kalen: Yeah. It's sort, I guess, yeah. It could be like a, could be like a prompt chaining thing.

[00:26:50] Like sometimes you want a prompt to mm-hmm. Give you something like lower fidelity and then pass that into another prompt. Sometimes that's more efficient, but Yeah, I gotta figure that out.

[00:27:01] David: I feel like we've talked about N eight N before, right? Yeah. Like the flow builder thing.

[00:27:05] Kalen: Yep. Yep.

[00:27:06] David: Have you incorporated that into your workflow at all?

[00:27:09] Kalen: Have not. I think I finally logged in, created an account, haven't done much with it. I don't like clicking around things like I want, I wanna Mm. It's like the GCloud, CLI. It's like that dude. It's like if I can prompt something to do stuff versus having to go click around on something, that's kind of my, my, I actually looked at and it, and if I could, because it's open source and I was looking at if I could build some kind of a, because I don't think it has, it has a lot of AI features.

[00:27:36] Like you can plug in an o uh, open AI prompt into a flow, but I don't think it lets you AI prompt flows, which seems like the main thing you want, like if you have this gigantic 50 step flow and now you gotta reorganize it, you gotta click 30 different things and drag 'em. Or like, if you could, that seems like the type of thing you should be able to prompt really well, you know?

[00:27:58] Yeah.

[00:27:58] David: And you should be able to, if they have like an API, then they should also have. A command line interface or something. Yeah. And then it's profitable,

[00:28:07] Kalen: right? Right. But then, um, the license, like you're, I think you're not allowed, you can like charge for it if you're sort of like an agency doing, I don't know, it's a very strange license, but basically my understanding was that trying to productize something like that would not be allowed, so.

[00:28:26] Gotcha.

[00:28:27] David: It's like, so it's more for like internal use or like contract work?

[00:28:30] Kalen: Contract work, yeah. Like if I'm doing contract work for you, I can go, Hey, I'll host N eight N for you on Digital Ocean because Gotcha. I know Digital Ocean and I'm your guy and I'm on top of it. So you don't have to host it with their main hosting whatever, or maybe you want on-prem or something like that.

[00:28:47] I don't know.

[00:28:49] David: Yeah, dude, I'm, I'm so impressed with those. Like I discovered earlier this week that this mini site I made for Kuana in 2020 is still there. Like it's still, I built it on Amazon Light Sale, like right when light Sale came out, uhhuh and then I basically forgot about it. Uhhuh, someone was, someone was like, do we have like a warranty registration?

[00:29:14] I'm like, yeah, I made one five years ago. Like, oh, it still works. And like, oh, we need to update this 'cause it's all wrong, but it's still hosted fine and like it's still wrong.

[00:29:26] Kalen: Nice. That's for Kuana, the warranty registration. Oh, nice. So has it been getting used and

[00:29:34] David: stuff? Somehow? I saw some people have submitted stuff like within the last couple months, so that's a problem.

[00:29:41] It's all,

[00:29:44] Kalen: it's always so wild to me. If something is still working, like within even a month after I last touched it, like there's nothing, nothing weird has happened. There's not a single error log. Like there has to be something and there usually I know, and there usually is. But

[00:30:05] David: I mean, to be fair, it was like, it's an H tml and JavaScript thing.

[00:30:08] That's just like a single page application, but Right. Servers go down. Yeah. And like things happen. Yeah. But light sale, just, I'm sure the servers that were running it have switched out multiple times, but I haven't had to be a part of that process

[00:30:23] Kalen: a while. I mean, even the JavaScript dude over five years, I'd expect something to break.

[00:30:28] Yeah. Something break. You know? That's cool. Yeah, I've had a few of those where like some, I think, like we were saying, these new infrastructure companies and JavaScript, it's just somehow stays up better than PH.

[00:30:44] David: Oh yeah. PP too.

[00:30:46] Kalen: Yeah. Quick shout out to Gil. I posted that the checkbox label in this one screen and the Shopify admin isn't clickable.

[00:30:56] It honestly bothers the hell out of me. Like I, 'cause I have to click on it a lot for testing. And then so Gil like went in and fixed it, which is

[00:31:06] David: ing, and then you immediately asked for something way more complicated.

[00:31:10] Kalen: No, no, no, because, because he literally goes, what's next? He goes, Hey, I did this. What's next?

[00:31:19] And I was shocked. First of all, that was an open invite. I was shocked that he did it at all. And then the fact that he said, what's next? I was like, dude, this is my chance. And because I feel like. The SKUs are already indexed. They're indexed. The variant IDs are indexed. It's all indexed.

[00:31:38] David: No, I agree.

[00:31:39] Kalen: I see it comes up like, although it must be in, maybe it's not indexed to the, uh, I don't know.

[00:31:46] Uh, maybe the SKU is only indexed to the product ID and the variant ID is only indexed to the product ID or I, I don't know. But that seems like the type of thing they could fix and it would make such a gigantic difference.

[00:31:58] David: Yeah. 'cause the, the only difference in the URL is that it's got like a slash variant slash the variant id.

[00:32:04] Right. That's it. So, so why can't we just go right there? It's such

[00:32:06] Kalen: an easy, it's such an easy URL.

[00:32:11] Even typing in the literal variant id, sometimes I'll do, 'cause it's faster than going into the URL and like, just at least make that clickable. At least that if I type in the variant ID itself, bro, hook me up,

[00:32:28] David: bro. Yeah, I, I still haven't heard from anyone on the, uh, the bulk editor situation with like the columns that keep coming back.

[00:32:38] What columns did we talk about that? No, maybe that was just a Twitter thing. Like when you go in and you bulk edit variants. Mm-hmm. Every single time you go in to edit. It brings up the inventory for that variant across every single location.

[00:32:54] Kalen: Mm-hmm.

[00:32:54] David: Which, like, if you have stores, that means it's unusable for you until you go in and remove all of those columns.

[00:33:01] What, so like you have to go in and remove the columns and then put in a column that you want. Like usually I put in like three meta fields that I want to update on. All, all like, right. All of these variants. Right. And that's also not sticky, right. So like every time I go into edit, I have to remove the columns, right.

[00:33:18] And then add in the columns. I wanted,

[00:33:19] Kalen: by the way, going back very briefly to the native PIM idea. Okay. Why? Why can't we just take whatever frustrations you have with the bulk editor? Automate those somehow as a phase one. That and that alone, you agree to pay $15 a month for it and we're off to the races because 15 is the magic team off

[00:33:46] David: to the races baby.

[00:33:48] Yeah, well I feel like, like specifically for the bulk editor thing, there's already a nice spoke editor. Isn't there? I just like, haven't bought it. I thought you, uh, oh, yeah. You knew about it. Espresso. Like espresso.

[00:34:00] Kalen: Yeah. Yeah, yeah, yeah, yeah, yeah. But for one, for one thing, fuck his app. But secondarily, secondarily, I'm trying to make this No, but, but like, yeah, of course you, you'd wanna try that one out.

[00:34:14] But I'm assuming that there, there'll be things in your workflow that like, could be like optimized, you know, like there's a segment of merchants that are like you in whatever way, you know, maybe it's fashion, maybe like, and you could build something, like, there's gonna be something like there's no way that that app is gonna perfectly fit your workflow.

[00:34:35] Like it might help, you know, with certain things. But like, yeah, if we build something to just dial your workflow in, like anything that would have any addressable market, like something fashion specific, like SKUs, how you get whatever the like I know, whatever. There's gotta be a feature set there to just

[00:34:52] David: probably,

[00:34:53] Kalen: yeah.

[00:34:54] David: Yeah, maybe we should, maybe I should, uh, just like go through all the stuff that sucks.

[00:34:58] Kalen: Yeah,

[00:34:59] David: exactly. Like, um, what was, uh, there was another thing like, okay, so we, we assign accessories, right? Here's a weird one, right. That probably isn't used across a lot of merchants.

[00:35:09] Kalen: Okay.

[00:35:10] David: But we assign accessories. And so accessories are a variant list on the variant.

[00:35:17] So the variant itself Okay. Has a variant meta field that's a list of variants. Okay. And so that, that's because we want to, like, for the black system tote, we want to assign all accessories that are also black. Oh, right. Every time I have a new color, I have to go in and I have to be like, okay, what is the usual set of accessories that we have?

[00:35:39] And also, do we have the color? Do we have a matching color in all these accessories? And if we don't right, what, what different color should I use?

[00:35:46] Kalen: Like you could have, you could have some auto assigned logic, like that feels like the way to do that somewhat generically. And you just have to set up that auto assigned logic once, like when a new variant is at, you know, find the, like link other variants with the same option.

[00:36:02] You know, there has to be Yeah, a way to do that.

[00:36:05] David: Yeah. That would be interesting.

[00:36:06] Kalen: But yeah, that'd be cool. And for, and for those following along at home, if you're trying to, to start a sass, you've gotta go in for the hard pitch. You gotta go in for the kill. So that was just a demonstration to the viewers at home.

[00:36:19] You gotta ask for the $15, you know, you can't stop at Yeah, that sounds like a good idea. No, no, no, no, no. You gotta take it further. You gotta go for the hard. Oh, okay. Yeah, you go for, gotta go for the hard pitch. David shot me down hashtag biz

[00:36:33] David: dev.

[00:36:34] Kalen: Biz dev. David shot me down for the 15, which is okay.

[00:36:37] David: Maybe they, dude, I, I will pay $15 a month for that, but

[00:36:41] Kalen: that's not gonna be

[00:36:41] David: worth it for you.

[00:36:42] Kalen: Maybe they make a billion dollars a year. He can't find 15 in the cushions. That's okay. That's not a big deal. But you, someday I'll have

[00:36:49] David: that budget. The

[00:36:50] Kalen: important thing, you gotta get the reps in, you gotta ask for that 15 over, over, and over until none of your friends wanna talk to you anymore.

[00:37:00] David: Yeah. And that's how you know.

[00:37:05] You're doing it Well,

[00:37:06] Kalen: that's how you know you're hustling. Okay. What's next? Okay, here's another idea. Here's another idea. Not for you specifically, but a product ID in general. Okay. So like AI native, like Klaviyo, basically. I feel like all the apps are gonna get reinvented for something that's very AI native, like everybody's tacking AI on and stuff like that.

[00:37:29] But like, for example, going, uh, and this goes back to the idea of prompting flows. Like in, I, I've done a couple Klaviyo flows where it's like, first of all, their flows suck. Like you can't do branching, like mm-hmm. Well you, I guess you can't do branching, but there's lots of stuff you can't, like there's, even compared to Shopify flow, so many limitations.

[00:37:47] David: Shopify flow has spoiled us, I feel like. Right. I was trying to do like some, I was trying to do some actions in Asana today. Right. And Oh yeah, it is horrible dude.

[00:37:58] Kalen: Yeah, Shopify is just honestly fucking really good at a lot of stuff. Like yeah, they just are. So, I don't know if you can prompt Klaviyo flows.

[00:38:08] I'm assuming you can't, I feel like a lot of the email apps, like you can prompt to like create an email or whatever, but what about the flow itself? What about the, like, Hey, let's start sending a newsletter every fourth Thursday, or, you know what I mean? Yeah. And I thought that could be a cool little idea.

[00:38:25] David: Yeah. A lot of them have those things like, uh, like, uh, don't want to build your own flow, ask AI to help you, and then you do it and then it like isn't what you wanted. Exactly.

[00:38:36] Kalen: Exactly. Because I'm, I feel like that has to be a core feature Set. Set. Has that Yeah. Yeah. Like I feel like that's not gonna be trivial to really do really nice job of.

[00:38:45] David: Unless you have a ton of like this, uh, the AI is like fine tuned on whatever your documentation is or whatever. Mm-hmm. And just knows how to do it all well, but it never seems like that's the case.

[00:38:54] Kalen: Yeah. Yeah. It seems like there's something to do. There's some work there. Like I don't know exactly where the work is, like picking models or I don't know.

[00:39:02] Or maybe the models will just eventually get better, make it trivial, but feels like there's like not an insane amount of work, but like a reasonable amount of work to make something that would be like really good

[00:39:14] David: you would think.

[00:39:15] Kalen: Yeah,

[00:39:15] David: but none of them are good. So

[00:39:19] Kalen: it's time to build baby.

[00:39:21] David: Maybe that's why Shopify doesn't have one yet.

[00:39:23] Kalen: Well, like yeah, they have like email app and, I dunno, the email app is weird. I don't know. People must use it, but. Oh, I

[00:39:30] David: mean

[00:39:30] Kalen: like AI

[00:39:31] David: flow builder,

[00:39:32] Kalen: right? Right. They got the AI flow builder now, but for example, you can't prompt to update a flow only to create one. Oh

[00:39:41] David: yeah. Wait. Shopify has one where you can create a flow with ai.

[00:39:45] Sweet. Wait, is that not out yet? I need to give that a

[00:39:48] Kalen: shot if it's not out yet. It's just a theory that I have. Um,

[00:39:59] and theoretically I think if they did do that would be possible. If they did do that, they would potentially only implement the create and not the update because that would be a natural mistake to make. Right.

[00:40:16] David: That would, that would be some feedback that I would give if I had tried it

[00:40:22] Kalen: hypothetically.

[00:40:25] That's so funny. Oh my God. So like what feature set would you need for this? Like for you guys? Okay. Don't worry. I'm not gonna ask for 15 again. Lets go go back to the $15. Not to worry that we're not, we're not asking again. You got, you go in once and then you back off for a month or two and then you circle back.

[00:40:44] So I'll circle back with you in a couple months. Like in terms of what you guys would need the way, do you use Klaviyo or what do you use? Yeah, we use Klaviyo. What would you need in terms of your feature set for something to be sort of like better than Klaviyo, but still have all the features that it had?

[00:41:00] Like do you do a bunch of flows? Do you like, you know what I mean?

[00:41:06] David: We do

[00:41:06] Kalen: do a bunch

[00:41:06] David: of flows. I think one thing that's weird, oh, someone's

[00:41:12] Kalen: singing. Oh, can you hear my daughter singing? Let me, let me tell her. No, it's okay. I love it. Hey Olivia, you're coming on the podcast that.

[00:41:26] That's freaking hilarious.

[00:41:30] Let's, she's having a good time dude, by the way, rj, don't edit that out. 'cause that's actually kind of hilarious. I want to hear that. That's funny. Shout out to rj. What were you saying right now? What was I

[00:41:42] David: saying? We were talking about Klaviyo flows. Oh, right, right, right, right, right. Things that could be better.

[00:41:48] Yeah. A support for looking at variant meta fields. Like done The one that you would think is,

[00:41:56] Kalen: there you go. I just implemented it. I just prompted it. It's done. What else you got? What else you got? She got,

[00:42:03] David: okay. I need more than 5,000 variants, please.

[00:42:09] Kalen: Okay. Variant meta fields. Okay. Okay. Makes sense. Makes sense.

[00:42:13] David: Yeah. There's a like, uh, specifically the problem is actually, I can't remember exactly what it is, but it's something like back in stock emails. We could make, like we have to do exclusions a lot of the time mm-hmm. For like an entire product.

[00:42:28] Kalen: Mm-hmm.

[00:42:28] David: Or like a, a variant id. Mm-hmm.

[00:42:30] Kalen: Right.

[00:42:31] David: And it would be much simpler if we could just say in Klaviyo, in the flow, don't send back in stock emails for variants that have, is hidden, equals true.

[00:42:40] Or something like that. Right, right, right. But you, you like can't really get at. Very medical. It's funny how much, like you could probably do some like web hook or something, but

[00:42:48] Kalen: it's, it's funny how much that your variant hidden thing comes up. Like,

[00:42:53] David: I know it's such an issue, dude. Do you know where else?

[00:42:56] Yeah. Like my boss was like, what's an MCP? Yeah. And I was like, oh, let me tell you about the storefront MCP and how it's gonna make us sell all the variants that we have fitted.

[00:43:06] Kalen: So

[00:43:09] David: it just goes everywhere. Uh, I think I still haven't heard anything from Shopify about like, how are we gonna be

[00:43:16] Kalen: able to adjust the, I could be experience, I could be hallucinating this, but I thought that Toby tweeted something related to this this week, or replied to somebody else's tweet.

[00:43:28] Oh God, I wish I had it handy.

[00:43:30] David: I did see something that said like, more news for merchants soon, but like, I just like, man, it was like, I've been, I've been asking about this specifically. Can you just tell me and I'll keep it a secret? Yeah. I wanna know. Oh

[00:43:44] Kalen: man. I could have sworn I saw something. Just, you have to find something to talk about while I scroll through this,

[00:43:52] David: because

[00:43:53] Kalen: where is this thing?

[00:43:55] David: What did I work on today? What did you work on today? That's a great question. Uh, we had some good AB tests. I'm not gonna say what they were, just to keep a little bit of a competitive edge. Wow. But it feels good.

[00:44:07] Kalen: Okay. Okay. Fair enough. If you want the real information, you gotta pay up. Just insert our upsell.

[00:44:14] That's right.

[00:44:15] David: Yeah. Buy my course. But no, it, that's another reason why I feel good today, I think is like holiday's coming and we don't always have wins when we're AB testing, but this past week we had like two really solid wins. So I'm, I'm excited about that. Right. Gonna do some extra sales this holiday season based on hypotheses we had.

[00:44:37] Kalen: Nice.

[00:44:38] David: Okay. Really, basically what it is, is one of them, we stole a feature from Bloomingdale's, and the other one we just put messaging in checkout. So there's

[00:44:47] Kalen: the, there's the thing. Nice. Okay. So Taylor tweeted about SEO, hidden Metfield should hide products. Okay. And Toby said, we're gonna clean this up by soon.

[00:44:59] Adding an unlisted publishing status. Now that's at the product level, I think. And then this guy, Jonathan, well that doesn't work. The dream would be for it to extend to variance. Very often we'll launch a new cosmetics shade to build a wait list, but it instantly flows through to other sales channels is sold out.

[00:45:16] And then Toby said, yes, we are looking at this. Do, do finally. So,

[00:45:21] David: fuck Yeah, that's it, right? Oh, that sounds great. Yeah, that's, that is what I need. That's it. Yeah.

[00:45:27] Kalen: I fucking, I was like, I, he said something. Thank you. Because I remember seeing it, like my mind exploding, so who knows when it'll get done. Oh, Toby's talking about it, dude.

[00:45:37] Then

[00:45:38] David: yeah, that'll, you know what I mean? That will be my Shopify Christmas gift. Like, just give me the ability to do things with variance. Yeah.

[00:45:46] Kalen: Yeah. Dude does make things complicated. Probably the indexing is expensive on that. That's what I always imagine. Like, why don't they do something that doesn't seem that complicated, but

[00:45:58] David: is it an indexing problem?

[00:46:00] Kalen: I have no, I mean, I mean, if they, if they had an index, they could just filter, just add a fucking single wear, wear clause and call it a day. True. I mean, I don't know. It's probably insanely more complicated. I also

[00:46:15] David: don't understand why they changed, uh, like on the product edit page, now it's two clicks to get into a variant.

[00:46:21] I feel like we've complained about that before. Mm-hmm. But why would you make that change? I don't get it. Yeah. Yeah. That's, that's important to, I guess, 'cause you can have multiple with the same option value or something, I don't know. Mm-hmm.

[00:46:35] Kalen: Mm-hmm. Yeah. I don't know. Dude, do you see these tweets about Toronto Builder Sundays?

[00:46:42] David: How much do you, I was wondering, like you just got so distracted. You went, you went straight on Twitter. You did not leave here. Wait, wait, wait, wait, wait. Whatcha talking about

[00:47:00] which to be fair, I've done that before. Uh, it's hard to, but no, like the. What'd you say? Oh, um, see, this is a great podcast today.

[00:47:14] Kalen: This is quality as fuck. Um, good thing nobody listens. Oh, Toronto Builder Sundays.

[00:47:24] David: Oh yeah, I've seen that. And then they have like some in New York. I'm super jealous, dude.

[00:47:28] Kalen: How much do we hate them for having it over there versus somewhere we could get to?

[00:47:33] Yeah, come on. Geez. Bring the show to us. Look, Toronto's not the center of the world guys. You know? Yeah. Just 'cause Shopify's there. Who cares? Do it in Austin.

[00:47:47] David: I feel like that's a, you gotta do it in Austin, Caleb. Like, I bet if you asked for money from Shopify to do a builder Sunday,

[00:47:55] Kalen: they gotta be putting so much money into that.

[00:47:57] Or they probably have. It's their own office. Right. So they just invite people over. Yeah, dude. You know how much money their office costs, man. You know how cool their office is. This isn't just some regular, I do little meetup with some pizza, dude. Like the offices are incredible. Just,

[00:48:16] David: well then I guess they need some nice space in Austin,

[00:48:20] Kalen: like I guess up Liam.

[00:48:21] List the dollars. Fucking Liam put something together, dude. Put together a proposal. Austin is the most important tech city in the in the world. Get it done, dude, come on.

[00:48:38] David: I have looked on like Meetup to see if there are any Shopify meetups in St. Louis and doesn't seem like there is or there was one, but it was like five years ago.

[00:48:47] Well, David,

[00:48:47] Kalen: something that sounds like you need to start one. It does. Sounds like I need to make one. Would you ever be like a meetup starter guy? Would you ever be that guy?

[00:48:59] David: I would probably get bored of it. Like really fast. Yeah. But it's one of those things that I would be like on a Saturday, just in the middle of the day, be like, oh, you know what?

[00:49:08] I should create a community and then by the next day I forget. Yeah. And I'm just like, yeah, back to playing video games or whatever. Yeah, yeah,

[00:49:16] Kalen: yeah. Like I'll tell myself I can do stuff like that and then I'll do it like one time and then I'm like, I don't wanna do this again for like another six years at least.

[00:49:27] You know, if I have a burst of enthusiasm, like I can do, I can pull that kind of like organize an event, but then mm-hmm. Like I don't have any consistency with it whatsoever. Like I wouldn't do it, you know, I would suck. It's like mm-hmm. Anyhoo. Okay, here's, here's

[00:49:48] David: one. Did we have

[00:49:49] Kalen: a,

[00:49:49] David: yeah. Oh, I did we have a list today.

[00:49:51] List. Are we just vibing today? Damn, we're just just vibing, man. My,

[00:49:54] Kalen: yeah, no, we could vibe, we could vibe it out too. So I was just gonna tell this stupid story, but I will skip it. Um, what? No. Bring it next time. Yeah. Listening. No, it was just, this dude like, was I, I met additions. He is like, oh, and we've been DMing.

[00:50:14] He is like, he's like moving to Texas. I'm like, oh, dude, that's so cool, man. He's like, yeah, moving to, you know, whatever. And oh, we should hang out. Like, oh, dude, totally. And then like, oh, hey, like, there's this surfing place in Waco. We could both drive to Waco. And I'm like, dude, that's so fucking, I've been wanting to go surfing for like years.

[00:50:29] I've ne I've been, haven't gone to that place. Ma, he's into surfing too. I'm like, dude, this is so cool. He's like, oh. He's like, okay. I, and then he moved and he's like, I moved here. I'm like, dude, that's so cool man. Let's like, and then it is like, let's go to the thing. I'm like, cool. I'm go Sunday. He's like, I can only go on Wednesday.

[00:50:45] I'm like, oh, like you can't ever go like, like on a weekend or something like that. And he is like, no. I'm like, oh, okay. Damn. That's

[00:50:55] David: sounds like you're taking off on a Wednesday.

[00:50:59] Kalen: Well, I can't, like I, yeah, I probably can't do that. I mean maybe, maybe once in a blue moon, but I got a word, man, it's rough. It's rough out on these streets.

[00:51:07] Gotta work. Got them clients that need me. Gotta pay these bills. Okay, so onboard B2B app. Shout out to onboard B2B app from, uh, the helium helium guys. They actually paid for a quick ad spot. No, I'm just kidding.

[00:51:26] But no, so they have a, so on onboard B2B is like, is like an app for doing like onboarding stuff for for B2B. So for example, like allowing people to apply to be companies and stuff and having that kind of a flow. You know what I'm talking about, that type of flow? No, I don't know what you're talking about.

[00:51:43] No. You gotta let me know if it's not making sense. What's

[00:51:47] David: a, what's a B two feel? What's a B2B on? Like a, it's like a, yeah, yeah. Tell me.

[00:51:52] Kalen: Yeah, basically like, it's like if somebody wanted to be a reseller of yours or something, and then they would apply, you know, so, so they can't just sign up as a customer and buy, like they have to apply and then you guys have to approve them and give them, like, I see pricing terms and stuff like

[00:52:07] David: that.

[00:52:07] That's like a typical B2B flow.

[00:52:09] Kalen: Yeah, yeah, yeah. Gotcha. So like in b2b, it's like you can just sign in with your email and now you're a customer, but you don't have a company associated. So like that whole like type of flow. Okay. There's some weird flows in there with B2B, but anyways, so I was building this kind of stuff, um, like la like uh, la uh,

[00:52:32] David: earlier this year.

[00:52:32] Yeah, I remember you were doing a bunch of like

[00:52:34] Kalen: Yeah, yeah, yeah. Uh, access control. Yeah, yeah, yeah. And then I started working on it, went back to it and I thought I was gonna do it all in new customer accounts. I was all excited. I was starting to rebuild it and they were like, no, we're not gonna rebuild it.

[00:52:43] I was like, okay, whatever. But in the process, the PM pointed me to, to the onboard B2B app, which I had actually seen before when they launched it. And he's like, check it out. I was like, I was like, oh yeah. I was like, no, this is cool. I know these guys. And then I was like, at some of the functionality, we had a lot of overlap.

[00:53:00] And I was like, why don't we just use the app? And then, um, so basically they kind of said like, well, we want them to use our app type of thing. You know, I was like trying to encourage like, we should, you know, use the app, we can, but on the other hand. The thing is with doing custom stuff, like as an agency is like, there's always some edge case or quite a few edge cases.

[00:53:23] Mm. And then the public standalone app just doesn't cover everything. You know? It's just Yeah. Always the case. So like there is something, I mean, it's like if I need to do something for a site build and there's gonna be a bunch of complicated stuff going on, it's like. I'd rather just do it myself. 'cause then at least know what I'm doing.

[00:53:41] Like I, I know that there's no additional limitations I have to worry about, um, versus like using someone else's app and then trying to augment. Yeah. Yeah. It's like you just know that there's gonna be things it doesn't do type of thing, but, you know, but maybe there is, maybe it has a really good API or maybe, you know, maybe they do a ton of customization or, I don't know.

[00:54:01] But it's, but the reason I brought it up is it's like, it's like an interesting trade off. I mean, I don't know if. You run into that? I dunno if you guys have, you probably probably don't have to do a lot of custom apps, but like when you're doing No, we don't. Yeah. But like, when you're doing custom apps, it's kinda like it, there's always public apps in, in that category and you know, but there's just always something because agencies get weird requirements too.

[00:54:24] Like clients like, we needed to do this, and it's like mm-hmm. The worst idea ever. But you get to the point where you're just like, we just have to do this. That's what they want. Yeah. And you do it because they're paying the bills. Yeah. And, and like it makes no sense to be part of an app. You know, you're just gonna run into stuff like that and then now you're figuring out how to work around.

[00:54:44] So I kind of get the trade offs, but, but like, I was thinking like, our agency is incentivized to like not use apps, you know, because they'd rather bill for the time. It's like mm-hmm. Yeah. Kind of. But also, unless you know an app really well, and you kind of know the, you have a mental model for how it's gonna fit into what you want to do, there's just a lot of unknowns and just even taking the time to talk to a vendor about it all, you know, they're gonna pitch you and, you know, whatever.

[00:55:13] I don't know. Yeah. It's just so complicated.

[00:55:15] David: Yeah. When I was in an agency that, that was definitely like, one of the considerations that we always made, even in Magento, was like, okay, we know there are some apps or, uh, modules that we're always gonna buy for people and like put them on their site.

[00:55:29] Kalen: Yeah.

[00:55:30] David: And like one of them was the, um, what was it?

[00:55:33] Karen Baker's apps. I forget what they're called. Web shop apps. Oh, web shop apps. Yeah. Yeah, shop apps. Uh, that was like, yeah, that was one that we always installed. And then there were a couple other ones. But then, yeah, anytime it went outside and like started doing custom stuff, we're like, we're gonna have to build this.

[00:55:49] But we always tried to like, save time and, and money based on like, apps we knew worked well. Yeah.

[00:55:55] Kalen: Yeah,

[00:55:56] David: so I bet that's the same thing that happens in Shopify. Like there's everyone probably installs Matrix. I, there's probably like this awesome agency plan for Matrix I or something like that. But then, yeah, there's always something custom.

[00:56:08] So I agree that trade off, especially when it's like something that's critical to part of your business that is not out of the box. Right. It's gonna have to fit like pretty perfectly and Right. Or you're just gonna have to be cool with it, not doing exactly what you want. Yeah.

[00:56:24] Kalen: But yeah, but having said all that, like I was impressed with the app because we'd spent a lot of time thinking through these different flows and where to put things and how to use new customer accounts for them, blah, blah, blah.

[00:56:34] And then when I just looked at their app, I was like, oh, they really, they figured it out, which makes sense. Like this is what they do. You know, like custom app, like builder type deals? Mm-hmm. Or like custom form builders. So this

[00:56:46] David: is one that you think they, they made like for a client and then made into an app or their, they're full-time app.

[00:56:52] Oh, I don't know. Developers.

[00:56:53] Kalen: No, I mean they're full-time public. The app developers, I don't, I don't even think they have clients, but I don't know. Gotcha. But yeah. But it, no, it just, they did a good job. They did a good job with the implementation of it. You know, it was, it felt like the type of app that I should use.

[00:57:07] You know what I mean? It's like it does a specific thing. Mm-hmm. These guys are in the community, they're solid. You know what I mean? I was like, well, I don't use it, you know? But then it's just gets tricky, gets compli and it is different when it's open source versus when it SaaS. I mean, you always have the API.

[00:57:21] Mm-hmm. To work around things, but there's something a little more comfy about open source from this perspective. Just everything else breaks.

[00:57:31] David: Totally. Everything

[00:57:32] Kalen: else is trash and you can modify

[00:57:32] David: it if you want to.

[00:57:34] Kalen: Yeah. Everything but Kaylyn's thing is trash. Yeah, exactly. The main character. Speaking of which I got my first SDA warning.

[00:57:44] I, I had a firm speaking to in my dms for, uh, there was like, you're not supposed to share anything from the slack. Like there was a tweet that it had no identifying information whatsoever. It just, it was like, oh, since I already shared it, I guess I can share it again. It was like in the venting channel and people vent all sorts of wild shit in there.

[00:58:06] But it was just a screenshot that said like, the trifecta of venting has happened today. And it was like customer support. It was like three categories of venting or whatever. I just thought it was funny. Mm-hmm. And then, so then I got, I got an official morning that I could be banned for such Oh, such activities?

[00:58:24] No meme sharing. Yeah. So I almost rage quit, but I calmed myself down and I was like, just, just relax. You don't need to rage. You don't need to rage quit. It's okay. Everything's fine, everything's cool. You know, you gotta, you want to stay on the inside of the tribe. It's cold. It's cold out there. Yeah. Out there on your own.

[00:58:45] Stay in, bro. What?

[00:58:49] David: That would be awkward.

[00:58:52] Kalen: So listen, moving. If you,

[00:58:54] David: if you want to be a part of the community with Kailyn, you gotta accept his faults, right? Yes. They set

[00:58:59] Kalen: my faults like I'm not a perfect guy, David. Right. I never said I was, I never said I was perfect. I'm far from it. Far from it, you know what I mean?

[00:59:12] Very, very far from it. Let me ask you a question. Did you ever feel like you were like mostly a good person and then start to feel like differently about that? Or have you always felt the same way on that question?

[00:59:26] David: Hmm. I think I, I think I changed, like, I, uh, I probably feel worse about myself now. This is like a therapy session.

[00:59:35] Kalen: Yeah. Quick therapy session. No big deal. Quick ses. You have the greatest life, by the

[00:59:40] David: way. Uh, it always fills me joy. It's to hide my severe depression.

[00:59:48] Kalen: I thought you were gonna say it's, it's to hide the pain, but then you upgraded it.

[00:59:58] David: But to answer your question, yeah, I'm probably like more critical of myself now. I'm not even like, yeah. But in, in terms of like being a good person, I feel like I try to be a good person. Yeah. Maybe I'm not. No, no, no. I like, maybe I'm cranky with my kids sometimes or something. I don't know. Yeah,

[01:00:13] Kalen: no. As I'm asking this question, I'm realizing it was kind of obvious 'cause like you are, you're like a, you're like a good dude.

[01:00:19] I used to feel like I was a good person and now I kind of don't, which is a little, I don't know. It's crazy to say that, but you know, how do you feel about that? Well, well, I don't know, like part of me feels like I'm kind of being more honest about what has always been the case versus being naive or something like that.

[01:00:39] And then part of me thinks I just have changed in certain ways. And then part of me feels like I feel differently about what good and bad even means. Yeah. So that's too many variables, dude. We gotta reduces variables. Lot variables.

[01:00:54] And also, I'm a lot more irritable probably than I used to be. Yeah.

[01:00:58] David: I feel like I'm starting to learn that what is dangerous is thinking too much and you gotta like get out of there sometimes. Yeah. 'cause you just put yourself in those positions and then like now it's part of your thoughts that Well,

[01:01:13] Kalen: yeah.

[01:01:13] And it's like, I do that too, and that that's something I have to work on really hard. But like, on this topic, it's not even like I'm sitting here stressing about it and, and overthinking it. I'm just kinda like, yeah, I'm just, I'm not like the greatest person. And then I just go, all right, hey, let's play pickleball.

[01:01:28] Like it, you know, it doesn't like, I feel like there's a difference between are you a good person versus are you working hard at your goals and like, if you're working hard on your goals, like that's the main thing I'm trying to focus on. But yeah, I don't know. It's weird.

[01:01:42] David: There are also things that like will objectively make you a bad person and you're probably not.

[01:01:48] Doing those things. Like you're not like committing crimes or something, right? Or I mean,

[01:01:54] Kalen: we all speed, you know? So what type of crimes are we talking about? Yeah,

[01:01:58] David: true. I mean, yeah. So who's to say what, what makes a bad person?

[01:02:02] Kalen: I saw this TikTok video and it was this guy who was saying like, I, he's like, I get told that I give serial killer vibes.

[01:02:09] And dude, every single comet was h just his whole look, his hair, his eye, his expression, his, the shape of his eyes, the car that he was in while he was recording the video, the flannel sweater that he was wearing. It was like, and it was just h it was crazy how much he looked like a serial killer. He was just getting roasted in the comments.

[01:02:34] Um, that'll happen. Yeah, I think that about covers it.

[01:02:41] David: Dude, I went to the mall last weekend. Nice.

[01:02:44] Kalen: Nice.

[01:02:45] David: They have these pop mart vending machines. Do you know about Pop Mart? Nope. Like Loop La Booboos and stuff. Oh, I've heard about La Booboos. I don't know what they are though. Doesn't matter. But Pop Mart makes them and we found a vending machine at the mall, which was like an incredible experience for my daughter.

[01:03:01] But I got this weird robot thing, uh, and it's actually kind of sweet. I gotta send a picture of it to you. Okay. But yeah, they just have like these little action figure things that are so stupid, but. For some reason you want them, your daughter wants them. Well now I want these robot ones and I gotta get all of them.

[01:03:21] 'cause there's like, there's 12 of 'em plus two secret ones. Nice. It's like a blind box thing. You don't know what what it is, dude, this is the new, this is the new craze. You gotta know about

[01:03:32] Kalen: this. Okay. Okay, okay. Okay. Sounds sort of familiar. Wait, look it up. What's okay? Blind box. I'll look it up. I'll look it up.

[01:03:39] Wait, what's a blind box? I don't wanna look it up. The labu boos.

[01:03:45] David: When you buy them, you don't know what

[01:03:47] Kalen: color or style you're gonna get. Got it, got it, got it. Okay. That makes sense that that would be, yeah. It's like when you bought baseball cards, dude.

[01:03:54] David: Yeah. You know, except this company is like making so much money right now.

[01:03:58] Pop Mart is like a $40 billion business or something, just selling these little figurines that's.

[01:04:05] Kalen: Insane. Where are they based? Korea. Fri I was about to say fricking China again by Korea. Yeah, they, they did it. Shout out to Korea. I think

[01:04:14] David: it's, I think it's Korea, South Korea. Shout out to South Korea dude.

[01:04:18] We'll see if their website works. 'cause it's always broken 'cause they just have insane amounts of traffic. That would be a cool team to talk to. The Pop mart site team. Wait, are they, they're on Shopify? No, if they were on Shopify then I'd be buying La boo boos all the time.

[01:04:35] Kalen: Oh, they're on a, some custom thing?

[01:04:37] David: Yeah, I think so. 'cause their site goes down every time they release something. Oh

[01:04:41] Kalen: it does?

[01:04:43] David: Yeah.

[01:04:43] Kalen: That's hilarious. So their size sucks.

[01:04:46] David: I think it sucks 'cause I haven't been able to buy anything on it.

[01:04:49] Kalen: That's so funny. Dude. For a second I got mad. I was like, I can't tell for a second. I got mad. I was like, oh, why aren't they using Shopify?

[01:04:56] And then. Like, I hope they're not using something better. It's like, yeah, they're using something trash.

[01:05:03] David: Yeah, no, it sucks. Whatever it is.

[01:05:08] Kalen: Yeah. I feel like, like some people talk about like, Medusa as like an alternative to Shopify or whatever. I guess it's like open source or something, and I guess it's like headless and I'm like, I've looked at it a little bit, you know, like I don't wanna ignore everything.

[01:05:23] That could be better. But then again, it's like, I don't know. I mean, who knows?

[01:05:30] David: You never know. Yeah. This, uh, this is a nice website, so it's like one of those js it's medusa js.com

[01:05:39] Kalen: looks like. Yeah, yeah, yeah, yeah. And I think they have e-commerce. Oh, they have a

[01:05:42] David: cloud version. I see.

[01:05:44] Kalen: I think they have an e-commerce component, or I don't know if that's the main component is e-commerce, but, um, B2B, B2C.

[01:05:53] Yeah. And like I have to say for like B2B, there's a lot of weird Shopify stuff. Like if you could just have a nice little front end that was open source and then have like some kind of a super flexible backend, like, I guess from what I've heard about commerce tools, although I, I don't know exactly what that means.

[01:06:12] I've never like really worked in it, but I could see that being better. Better for,

[01:06:16] David: are there things that are weird? Like is the, is the Shopify API not, not fully. Like have you found yourself not being able to do certain things? Well,

[01:06:24] Kalen: like pricing is like complicated, you know, like you can create separate catalogs.

[01:06:28] Yeah. But you know, some people have like lots of logic to their pricing in some ERP, you know? Totally. So I feel like if you could just somehow wrangle, prompt that,

[01:06:40] David: pull that out data structure

[01:06:42] Kalen: into some backend, and then boom, you're done.

[01:06:47] David: That's it. That's all we needed,

[01:06:49] Kalen: you know, versus having to build some crazy cart transform.

[01:06:53] That's like only limits limits you to 50 line items or whatever. Mm. It's cool that you can do a car transform. It's freaking amazing. You're running code deployed in directly on Shopify, but there are limitations. You know, it's like all of a sudden your B2B client has a fucking 90 line item, gigantic order, and you're like, oh yeah, sorry.

[01:07:15] Worth

[01:07:16] David: $0.2

[01:07:16] Kalen: million. Yeah. We told you guys eight months ago that we could only support X line items and nobody even replied to that email or whatever. And then now you're like, you know, and I get it. I get why it's a big deal because that big order is a fucking big order. It's like mm-hmm. That's exactly the order you don't want to, to lose.

[01:07:40] Um, they just needed to yell at someone about it.

[01:07:43] David: Yeah.

[01:07:44] Kalen: Wait. You needed to what? To yell at someone.

[01:07:46] David: They needed to yell at someone. Oh. Or like blame someone about it. Yeah. You know? Yeah, yeah, yeah. So you're not going surfing this weekend?

[01:07:52] Kalen: Not going surfing this weekend. Although I will be playing some pickleball, so.

[01:07:59] Nice. I should be good. Gonna try to take my daughter paddleboarding. What

[01:08:03] David: about you? What do you got cooking? I might go to this scout thing, boy scout thing, or Cub Scout thing. Sick over at Beaumont.