Subscribe
Share
Share
Embed
For long-form interviews, news, and commentary about the WordPress ecosystem. This is the companion show to The WP Minute, your favorite 5-minutes of WordPress news every week.
Eric Karkovack (00:00)
Hi everyone, and welcome to the WP Minute. I'm Eric Karkovack. So today's episode is all about the pros and cons of Vibe Coding, specifically the long-term consequences of all these little apps we're generating. I'm lucky because I've got two knowledgeable guests here to help me out. First, we have Kim Coleman, the co-founder of Stranger Studios and a known Vibe Coder. And then we have my colleague, Matt Medeiros whom I like to call Mr. Vibes because his penchant for building all manner of AI powered apps is really kind of out of control right now. ⁓ Folks, welcome to the WP Minute.
Matt Medeiros (00:39)
Thanks for having us, Eric.
Eric Karkovack (00:42)
I'm hosting your show, Matt, it just seems very weird to me, but I'm okay with that.
Matt Medeiros (00:44)
Yes, it is quite
inception.
Eric Karkovack (00:50)
So my initial idea for this show was basically called the junk drawer for AI vibe coding. Because we are using AI right now to build all sorts of things. Matt just built a thing that kind of saved his city during a blizzard. And we'll get to that in a minute. And Kim, I know you've been using AI for paid memberships pro and some of the things you and your husband Jason are doing.
How have you been using this technology to have you been creating new features, fixing bugs?
Kim Coleman (01:29)
Yeah, hi, Kim here. didn't speak yet. But yeah, thanks, Eric, for having me. We're using it for all of the above. So I think the evolution was at the start just let's chat with AI in a chat-based tool. And then we moved into agent-based development, both for internal and customer-facing products. Now we're actually plugging an agent fully into our whole team. So all of the above.
Eric Karkovack (01:57)
So how does that work with the agent? What are your expectations for the agent being plugged into your team?
Kim Coleman (02:04)
I'm right now the expectations are really low. So the agent's name is Flint. Flint is in Slack. Flint is able to do pull requests, create issues. So it's kind of this training place right now where we're figuring that out. ⁓ And it's confusing for the team and it's interesting to watch the interactions happen. ⁓ But I think the expectations are that they grow over time.
Eric Karkovack (02:32)
Yeah, that's fascinating to think about a bot as part of your team all of sudden. So you can just tell Flint, please create a pull request with such and such in it. it's not autonomous yet, right? It's not actually like suggesting code.
Kim Coleman (02:50)
It is autonomous in a way. Flint has a Trello board. So Flint is able to process things from an autonomous queue when there's bandwidth for Flint to do that. So prioritizing direct conversations, direct asks, and then processing the autonomous items that are put on the Trello board. I mean, what does the future hold? It probably holds ⁓ us in the next step being issues tagged on the public GitHub repos. ⁓ Flint look for all issues tagged with
blocking next release or all issues tagged with good first bug are some things that we've used before and just saying go at it.
Eric Karkovack (03:29)
That's really cool. So, so Matt, I know you've, ⁓ you've been doing so many things here at the WP minute. You've, you've built custom blocks for us. You've built a plugin that, that goes into the YouTube API without needing an API key. And I know you've also created ⁓ a, app that helped your town ⁓ report which roads were plowed and which weren't during a recent snowstorm. And you've got. ⁓
cool ceremony from your town for that. Tell us about a few of the favorite things that you've done beyond that. mean, because it seems like you're really into this.
Matt Medeiros (04:09)
Yeah, and I think it stems back, I think one of real interesting things about AI is the same thing that really excited me about WordPress, you when I got into it 16, 17 years ago, was the ability to do something like cool that I couldn't do with my own abilities of writing lines of code, right? So when Front Page came out back in the day, and I was able to build websites graphically,
That was really interesting to me, right? When I first started learning websites, like a lot of folks, it was just writing HTML, making a bunch of tables, ⁓ and then putting ⁓ spinning animated skulls in horizontal rule bars, you know, on web pages. And that was fun. But then as technology accelerated and matured, like all, think, well, I think like for all software for folks like us and those of us watching to create things,
Eric Karkovack (04:53)
Yes.
Matt Medeiros (05:07)
It iteratively gets better ⁓ over time. And AI is just a massive accelerant to that, right? When I first got into building sites, it was with Drupal, and Drupal has CCK and Fuse, what we kind of know as ACF in the WordPress world. ⁓ And then the ability to display this content, like a loop block. ⁓ It was amazing. Like, I was like, look what I can do without having to write PHP and MySQL statements.
And this AI stuff has really accelerated that. So I started playing around with it maybe like two and a half-ish years ago, ⁓ specifically like chat GPT. And I know the overarching ⁓ thing here today is like the junk drawer. I have launched a lot of junk that I have not done anything with. ⁓ as we were leading up to this conversation, I was like, God, I...
The first things I was doing was making podcast downloaders and podcast players, and I'm like, I don't even remember the website I launched that on. What did I do with that API key? I don't even remember when I was launching this stuff in the beginning. But really, for me anyway, comes down to the excitement of being able to do things I'm not able to do, learning along the way. So it's not just a blind eye to just throw this stuff out there and see what happens.
It's accelerated my learning of other technology stacks, spending a lot of time in CloudFlare, understanding JavaScript more, understanding how other developers might look at WordPress and go, what is this thing? Look what I can do with these libraries here with, you know, whatever, like React app, so on and so forth. ⁓ So it's both, you exciting. It's a great education process for me, but I have launched a bunch of stuff that
I don't even know what to do with anymore. Like now I'm just thinking about having to go and take it all down. I'm like, God, do even want to do that? And what does that mean for the world of Vibe Coders? I have no idea.
Kim Coleman (07:09)
the interfaces of that earlier stuff, how much better is the stuff you're making now? It's like reps at the gym. You've iterated on things you might call junk that are improving what you've created later, even if they're unrelated, even if they're quote unquote unsuccessful. I think that is the value of having an AI junk drawer, is making yourself
face that empty IDE and say, can you make this? Was my prompt good? ⁓ Could I have given you clearer instructions from the beginning and reduce the cycles back and forth? So more junk.
Matt Medeiros (07:46)
Yeah, there's, and maybe we'll talk about this, ⁓ but there's also this level of ⁓ overwhelm and excitement at the same time. It's like, this new agent thing came out, or now Claude Co-Work can do all kinds of other things. Go buy yourself a Mac Mini to launch all this stuff. And there's this thing of, damn, there's so much stuff happening that I feel like I'm missing out not being able to keep up with the latest thing.
But I'm still leagues ahead of so many other people because I've created this junk drawer of stuff over the last few years. I'm like, at least I'm in it and I'm practicing this thing. Like at least I have that advantage to those who might not be, you know, even jumping in yet. So ⁓ yeah, there's that whole aspect of that junk drawer too, of like the weight of it all.
Eric Karkovack (08:34)
Now I wouldn't call what I've created a junk drawer so much as it is just kind of filling, fulfilling client requests and trying to get things done faster. And I'm finding like, I know we had a discussion in the WP minutes Slack not that long ago. I was asking about the advantages of Claude compared to what I had been using, which was mainly a co-pilot.
inside VS code, which was just a buggy mess, giving me wrong answers constantly. I have recently gone to Claude and I'm act, I'm really happy with, with the results so far, but I find myself like clients have had requests that have felt over my head over the years as a developer, things that I would spend hours on stack overflow, trying to feverish feverishly copy and paste code snippets to see if they would work. then when they didn't work, I found another code snippet and I, it
It kind of trained me to do things the wrong way. And it's funny, with AI, I find myself actually thinking more about architecture. And that's like the one thing I've really learned through all this is that I am thinking more like how this thing is going to work, not just that it works. I think that's been kind of valuable for me.
Matt Medeiros (09:53)
Yeah, yeah, I agree. Kim, I'm really just one.
Eric Karkovack (09:56)
So.
Kim Coleman (09:56)
Yeah, I mean, when you say the
word architecture, I think not only how it's going to work, but also how the code is structured. So something I think people who are using AI to vibe code more with are realizing that AI is starting to use these kind of standard folder structures and ⁓ building itself in a way that becomes repeatable. So I think ⁓ as you're vibe coding, consider vibe coding like a template for these things, like a framework for what you're going to vibe code.
it'll make everything more seamless when it needs to jump between projects. It's not like a whole relearn process, but I think, yeah, the architecture thing of how it works, but also how it's built and structured and how the code is structured. It's really cool to think about.
Eric Karkovack (10:42)
So one of the things I did want to bring up too is that, okay, so we're doing all this stuff really rapidly, right? Like we're pushing out, mean, Matt, you pushed out your ⁓ snowplow map within what, a day, pretty much. ⁓ So we're doing all these things really quickly. ⁓ recently the repository noted that the WordPress plugin ⁓ directory is now receiving over 500 new submissions a week.
And I've got to think most of that is AI, right? ⁓ I think for developers doing that, like we're going into the, now we're not just coding for ourselves, we're coding for other people. They're going to be using this code. It starts to become a certain responsibility with that, right? And how do we maintain this type of thing? we're, just say, you might've published five plugins this year.
You know, all of a sudden you've got all this maintenance on your hands. Did you think about that before you put the stuff on the directory?
Matt Medeiros (11:48)
Kim, I'll let you go first.
Kim Coleman (11:49)
I'll take that one
or Matt. OK, yeah. So we actually published three plugins this week, not this month, not this year, three in a week. And ⁓ I think when you're vibe coding, when you're launching things that are largely AI based, you have to use a human AI human system. So ⁓ Eric, talked about the architecture thing, the human designed and created a brief, even with the help of AI, let's say. You went back and forth.
Eric Karkovack (11:56)
Wow.
Kim Coleman (12:18)
in chat with AI to design the framework structure, functionality, no-goes, rabbit holes about a project. You handed it to AI to develop it and you got it back and you either tested it, tested it thoroughly, had the code reviewed, handed it to another AI to code review. All those things are great. At the end of the day, there has to be a human saying, how important and how critical and how bad would it be if this code is insecure or fails or breakdowns? ⁓
What am I opening my customer, my user up to? What risks am I opening them up to? And if it's relatively low and you're just like, you know what, this is a landing page about snow and has a map, there's really no risk. People hit the website. I mean, the worst thing that could happen, I have some kind of injection point that is in the server and something goes haywire and I pay some crazy Google Maps fee. I'm trying to think what the worst thing that could happen is.
So I think having a human sense of the security and the risk of it being incorrect is kind of an important thing. But also as quickly as that code was created, that's how quickly bugs can be fixed or faster because it's a day later and everything got way faster already. So we're kind of in this mindset of if it's low risk, push it, let users be the testers, let users be the stress testers, let them try it and say, I wish it did this, let them try it and say it's broken, it doesn't work this way. ⁓
It's way easier to get that kind of feedback when there's something for people to play with than it is to say, what do you want? I don't know what I want. Give me something to try. The same is true with localization for paid memberships pro. We have a localization server. We are putting AI translations out to our customer sites. The people that were contributing translations on their own, zero. The people that are fixing translations that AI created, 100%.
they're like, ⁓ it's easier to just fix this translation string than it is to start from scratch. They will take a 99 % complete translation with fuzzy strings and get it to 100%. They won't do it from zero to 100. So I think giving them a little bit of slop is useful.
Matt Medeiros (14:35)
useful slop. You know, there's something I want to. There's something I want to add, because I like. I mean, obviously, we're all trying to root for the human side of this stuff right now to save our jobs and make sure that we still have a purpose in like the careers that we have now or not. Things could change. don't know. But, you know, I think what I'm trying to get to now is do I need.
Eric Karkovack (14:35)
That's a really good point.
Kim Coleman (14:36)
Useful Slop. Use my Slop and I'll make it better.
Matt Medeiros (15:04)
Like, because I can build the thing now, do I really need to build this thing? Like, does it need to exist is like the first question that I have to ask myself because I will literally just keep opening up Cloud Code sessions and just creating stuff. And it's not a healthy space to be in because I can't even remember like what I started sometimes. Like, where did I leave off on this particular project? ⁓
But I think for the business side of things, especially for like Paid Memberships Pro, you probably have customers that come to you and say, I made this add-on with Cloud Code, it's not working, can you help support this? I think our abilities as business owners now might be to quickly launch these features that otherwise might never even reach the product. It certainly might take months because you're just like, it's not a high priority.
But you can now kind of like get out in front with the caveat of like, this is like, these are add-ons from the beta program. And of course, if you're gonna use the beta program, it's not gonna come with support, but we built a bunch of these things for you, go use it. And hopefully that's a stopgap for that end user to then look at that and go, all right, I was gonna build it, but I see that they have it, right? And if, you know, I don't know what the perfect balance is, but I think there's a way for product companies to
to get this stuff out faster. know, again, it might not be supported in the core product, but at least you have it. And maybe the customer wants to go and fork it and do something else, that's fine. But like, you're leading the pack because what'll happen is you'll have the people who start to complain about whatever, features lacking or this thing should be in paid memberships pro, I can do it faster, blah, blah, blah. I actually saw this on a Gravity Forums webinar yesterday where somebody was like, well, I could do this with AI and it's like.
Good luck, like you can do it. But we're not going to support it. So you have to use our official channels. I don't know how the dust settles, but I think like shipping a bunch of stuff is still smart. ⁓ And it's only going to hopefully benefit because you're the brand holder, right? You own the brand and people are going to get that code from you and trust you, you know, to to back it up. But ⁓ like even the snowplow app, just one last thing about like that.
Eric Karkovack (17:01)
You
Kim Coleman (17:26)
I have been, ⁓ yeah,
no, go ahead.
Matt Medeiros (17:28)
the
human side of it, there's things that you know, like there's things that you have decades of membership software experience. So if somebody comes along and goes, well, I can just replicate paid memberships pro, can just put the repo into the Claude code and I can just clone it and just move on. And they could, but they don't know the business. And even all of the chats that you might have with Claude code, it's not going to give you the 20 years.
of customer relationships and experience that you've built. Same thing when I built the Snowplow app. I I'm not a developer, but I had people messaging me saying, I was building one too. And you you're like, take a look at it, see what it is. And it's like user registration, user profiles, like a social media site built in. I'm like, this is somebody who's just having fun with vibe coding. I built it knowing what the end user is going to want in an emergency situation. That is zero logins.
zero stupid social media like chat rooms ⁓ in the fastest way possible to report an emergency. Everyone else that threw their app at me during the storm saying that they did it better, it was like they were replicating Facebook for snow plowing. It's just like, that ain't gonna work because you don't know, unfortunately, you don't know what customers need or what users need in a situation like this. So still having that human in the loop is very important.
Kim Coleman (18:55)
Yeah, I'm not gonna give away too much of what's coming for our team, but our support tickets are at an all time low because people are helping themselves through the issues that they face. And I think WordPress products that are open source and coded to be customizable used to get a really bad rap. So Paid Merchants Pro is probably known as the free option. It's probably known as the one you need to use code to customize. And that used to be a point of friction.
Eric Karkovack (18:56)
Kim, what were you gonna say?
Kim Coleman (19:24)
And for us, it was always the value, right? ⁓ For years it was. But we don't ever tell you no. There is no no. There is just yes with this code, yes with this hook. You can get it done. And it's interesting to be here now where people are no longer afraid of that and they're valuing it more. If you're on a hosted membership platform, for example, there's no way to drop custom code in, at least most of them and at least currently, to tweak the way something works. ⁓
So yeah, think that positions WordPress and other well-coded open source pluggable plugins in a really good place for where we're at. Yeah, Matt, I have a solution for that. I don't even know where they're all located thing. I don't know if you're doing this, but every time I work on AI, I use the same folder on my computer. All the projects exist in it. Maybe you do this too. And before I close, I say update the status file with all the things we worked on today. this, I don't know, are you doing that?
So then that's my list later.
Matt Medeiros (20:23)
I'm not doing the status, I thought, yeah.
Yeah, I thought of the status update stuff. ⁓ I do work out of one folder, thankfully. ⁓ It's the other side, it's the hosted side. I have some projects early on that I launched on Netlify, but then I, like six months ago, really pushed over everything to CloudFlare, doing everything over there. So that's where I have to wrangle it. It's at that domain level of like, where did I?
Where did I launch that? Where's my API key somewhere on the web? It's not funny.
Eric Karkovack (21:00)
Well, Kim, to your point, I've always loved when plugins provide me with code snippets to get me started on certain things. know, WooCommerce has always had that library of snippets that you could use, and the events calendar is another one of them. Just for those little niche tweaks that your clients may want.
I think I would have to think that with Claude or whatever model you're using, you have like this unbelievable opportunity to just fill your clients in and get them started on something. Whereas before you probably would have to send it to a human and go through a lot of back and forth just to say, I'm sorry, we can't custom code this for you, but it just seems like such a time saver there.
Kim Coleman (21:47)
Yeah, and I won't give too much away, but I think I hinted at the support being extra low. ⁓ Matt, you mentioned people just bringing you their own custom code or no, that might have been Eric, bringing you the thing like, I got this from TrackGPD, how's this work? I think all plugin companies are gonna be in a position where they become the business consultant. The, why do you need this? Why are you asking for this? And also probably the developer. So I think, unfortunately, everyone out there that's doing WordPress managed
services, Eric, maybe your role. I'm not sure that those last one, plugin companies are maybe better suited to do those custom code pieces for people. I don't know. We'll see where it all lands.
Eric Karkovack (22:30)
Well, I think for one thing, would make save me a lot of time dealing with customer relations and all that stuff. mean, AI has already saved me a ton of time just because I can rapidly put something out there that may have taken days or weeks before I can probably wrap it up in an afternoon. But I have another question for you both because I think this is, this is something that I've been thinking about a lot.
Think about when you first started using AI. You might have been using Chat GPT, or maybe you were using Google's whatever it was called before Gemini. I forget what they called it before Gemini. ⁓ But now you're probably using something else. when we're talking about maintaining this, at least on a freelancer and agency level, ⁓
What about hopping from these different platforms? Because that could be a problem, I imagine, with it picking up the context of your code and the type of features you want, right? mean, how would you suggest handling that?
Kim Coleman (23:35)
So I don't know if you know that the WP Fusion developer Jack Arturo, his company is very good plugins, but he has something he's open sourced and is working on called AutoMem. So for people that are using agents with AutoMem, they are not linked to a specific API, a specific AI tool. They are choosing between the ones they are connected to what to use for what outcome.
Matt Medeiros (23:36)
I'll let Kim go first.
Kim Coleman (24:04)
So that's more like a human saying how they want to work on something, where they want to go, making decisions based on what's going to work best for them, and doing it. So ⁓ as quickly as possible, move to something like that where you aren't ⁓ using one tool specifically, that you have an agent that makes those decisions for you and has the memory of what you've worked on already stored. So I don't know. That's how I'm approaching it.
why I'm not worried about it. do still use Claude code in VS code for a lot of things and I'm trying to give it a memory of things, but ⁓ yeah, I think moving to that agent with memory route and bootstrapping on all the LLMs is the right way to bulletproof yourself.
Matt Medeiros (24:52)
Yeah, I agree. This is an interesting space. ⁓ This is like part of that overwhelm that I was mentioning earlier on. ⁓ I never jumped like when Claude bot came and then turned into open claw. Now I know there's like a bunch of other ⁓ agent ⁓ based systems like that. That's like the one area I haven't gone down yet because I see like Jason talk about it all the time and like I'm really interested to like learn like
what he's using, how he set it up and stuff. But I feel like that is moving so fast from, like I've seen a lot of people do like the whole page builder thing of past, right? Where like, we launched Claudebot. And it's like, everyone's so excited, look what I'm doing with Claudebot. Two weeks of like all this awesome like posts that they're putting on Twitter. And then two weeks goes by and it's like, well I've been spending thousands of dollars on this thing. And it's really not doing what I thought it would do. And now they're onto like to the next thing.
So there's like that gut feeling of me is like, I want that, but I need something like that to be in a stable position. Because to me, it's almost like hiring a human, where I need somebody. ⁓ Like we recently hired ⁓ somebody at the WP Minute, Jacqueline Meredith, and she's amazing. And she's been working for me. And it's just so, you know, whatever, comforting to be like, ⁓
here's the things I need to get done, you already understand what I need, just please go execute on these things and there's nothing I have to worry about that. And I need that from the agent side of things. Otherwise, it's the same spaghetti brain of now I have to watch this thing to make sure it's actually getting it done. And that's not an advantage to me. I don't wanna tinker to build an agent.
And constantly, like I see people were complaining about Claude about like forgetting stuff in the memory and like not executing and spending too many tokens. It's like now I'm just now I'm just managing a person now. I've just hired the wrong person and it's not a person. It's just code. Right. So I want that to be stable. And I'm willing to like, you know, I know Claude is like pushing into that territory. I don't see it like what Claude Baud is doing or this other agent Hermes that people talk about and some other things. I don't see it being totally hands off yet with that. But if they do.
And if they charged another 200 bucks a month to have that, I'm paying it. Like as long as it's stable and it's actually doing what it's supposed to do. I'm paying it. Like if I can give it the six or seven repos of plugins that I've built and I just tell it, hey, go nuts with this stuff. Watch for security. Watch for WordPress compatibility. And every now and again, hit me with a feature that you think is cool. And if I can just talk to it and be like, yeah, cool feature, go ahead and add it.
and I can totally trust it, I'm in. And that's the part that ⁓ would be a huge benefit. I've just held off because I still feel like we're kinda duct taping these things together right now. ⁓ Unless you've really dialed it into this one thing you wanna do.
Kim Coleman (28:00)
I was going say not all of our work is automatable, but a lot of it is. So if you think of the handoff to a human and you said, she just knew what I needed, was that because you had ⁓ some kind of SOP or is it because you have developed a relationship? Was it a chat you had? ⁓ The weird thing is to constantly be in a head space of, can AI do this for me? Is this a task I'm going to have to do multiple times? Should I ask AI to do it as a one-off?
Should I build the skill as we do it together? And then it's something I can invoke later. So just like something you do repetitively in your work, you can write it down as an SOP. That's just a document on Google somewhere. The same thing can be true when you're working with AI. You can say, ⁓ we're releasing an add-on. These are all the things I think about when I do it. Let's release one together. And then when we're done, let's write a repeatable skill.
for you to use that I can invoke in the future to do this. So it's like training a human. It makes work maybe a little slower at first, but it's iteratively going to get faster because the SOP is the skill and it's a memory it can pull from. So I don't know, it's crazy.
Eric Karkovack (29:16)
We're really just scratching the surface of this right now. It's only been a couple of years of AI and it's like, I think about where we've come from, just the basic, look what I can do. I can recall ⁓ rewriting my bio in the style of Cookie Monster. my gosh, look what we can do. And now I'm using it to, you know, like increase revenue and actually do useful things, unless you consider Cookie Monster very useful, which, you know, it's fine. But yeah, the...
The agent stuff I have not gotten into yet either. I'm waiting for some stability as well. It just seems like...
I'm very distrustful of these things. I'm a control freak. I want to know exactly what's happening. And so for me, it's like, I'm just not ready to take that leap yet. But I've already come way further than I was, you know, in the last year. So I, it'd be interesting to see in like the next two years or something that we have as far as these agents that can go off and autonomously carry out our mission, you know, and report back to us. But
I guess I want to wrap up on this. For both of you, what are some best practices you would recommend to other people just for avoiding being overrun with all the stuff we've written and being able to maintain it in a way that also maintains your sanity? What would you recommend? Kim, I'll start with you.
Kim Coleman (30:46)
Let's say some things are just done for fun and just be comfortable with that, be okay with that. Put in the reps, learn the systems, see what AI can do for you. And it's okay if you don't ship those things. That's okay. What did you learn from it? Go that route. ⁓ Ship it to GitHub and do nothing with it. It's a resume item. definitely, when I'm hiring people, I'm looking at their usage of AI. ⁓
sensitivity toward it if they're skeptical of it, if they're trusting or untrusting, if they have concerns about it, that's a no-go for me because I know we're going to all need to be using it. So just their attitude toward it, that's important. I would say like loop AI into the, not just like the shipping of it, but the marketing and the nurturing. Like Matt was saying, ⁓ maybe you're not there yet to let it autonomously run, but at least ask, you know, does this have legs? Run an analysis of.
if this is something I could charge for. think we're not, we're looking at it as coding products for us, but we're not looking at the AI as like a mentor on top of that or a marketer on top of that or a salesperson on top of that. ⁓ I think unlock your brain to what's five coded products are and all the pieces that AI could ⁓ be a part of for you.
Matt Medeiros (32:07)
Yeah, for me, some of the more hands-on stuff that I try to do in every single project, ⁓ and this is like if you're just getting into this stuff and trying to figure things out, I'm constantly asking it to document everything as much as possible, create a systems breakdown of, especially if you're in the territory of building stuff that you've not done before, especially if it's outside of WordPress.
making like an architecture document just so you completely understand like where the requests are going. Again, I do a lot with Cloudflare, so it's like understanding where are the functions calling, you know, the database and at what layer and where is it sending this data? If we're doing edge caching, like help me understand that. That way, one, it's great to bring it to another, you know, agent in the future for it to understand what's happening, but also for your own understanding of every time you launch a new feature, constantly asking it to document. ⁓
The thing that, and another thing that I'll do is, because what I've found anyway, at least from my workflow, is design and UI is still not solved 100%. There's a bunch of apps I've been testing this last week, which are doing an okay job, but still not great. I'll often ask the agent to, like if I'm launching a new feature, it's something like, quickly mock this up in HTML. Give me like three variations in an HTML file or files so I can just view this.
Locally on the browser. I'm gonna tell you which one I'm gonna pick and we're gonna like build off of that basically, that's just like insight into Again, this is just my workflow. This is just insight into I'm not just burning tokens constantly being like can you move the button to the left? Can move the button to the right? You know like change the label of the button and it's like rewriting all the JavaScript code or all the front-end code I'm trying to take it in steps, especially with design because that's usually the biggest burn for me the last thing they'll say from
conceptual point of view I guess or a business point of view because I know everyone is trying to figure out how to You know make money with this stuff or improve their their margins as you still have to be a great human communicator right now I mean unless Jason has found an ⁓ agent system that is like doing cold outreach emails and landing customers I still don't know how you're going to make the sale of your thing if you can't be out front talking about it
You know, I saw a lot of people talking about, like, I actually made a video just to test things out. Like, I made a video with Claude Code of the Snowplow app when I was launching it. And it, like, created the whole MP4, like, you know, showing interactions in the browser and doing all this stuff. And it was, really cool. But that would never sell anybody to buy it. ⁓ I'm having deep conversations with local governments, and I'm a human.
and I am like experiencing the sales process. There's no agent in the world who's going to do that for you right now because the person on the other side is gonna not even know what to say to some kind of like automated system. So you still have to be, and this is where you can spend your time, know, on your marketing, on your messaging, on your connection with the customer, ⁓ and hopefully the right customer. it's, you know, all of this stuff that I've been doing with AI is great.
on the productivity side, and the coding side, and the brainstorming side, but to still land the deal, it's still happening what I'll call face to face, but you're still doing a podcast, doing a YouTube video, doing webinars, emailing people and responding to them in a human fashion. ⁓ Still gonna take some effort to run the business for you on that sales side, for now anyway, for now.
Eric Karkovack (35:56)
Yeah, I mean, the technology is there to help us, not necessarily replace us. At least that's what I keep telling myself time and time again, as more dire news comes out about what these things are doing. But I thank you both because you gave some great insight into ⁓ what we're doing with our Vibe coding. Kim, where can we connect with you online?
Kim Coleman (36:18)
The best place is probably our business website, paidmembershipspro.com. I'm also trying to ramp up my personal Instagram, the real Kim Coleman, and my Astro blog where I share, what I share on LinkedIn in more depth is the real kimcoleman.com.
Eric Karkovack (36:35)
Awesome. Matt, I bet people can find you at the WP Minute, but where else?
Matt Medeiros (36:39)
the WP minute and of course the day job at Gravity.com
Eric Karkovack (36:44)
All well thank you both for being a part of this and thanks to our audience for checking us out. Please visit the WPMinute.com slash subscribe to get our newsletter, support the work we do, and hey, join our Slack channel. Come talk with us about your vibe coding adventures. ⁓ That's it for now. I'm Eric Karkevac and may all your vibe coding ventures be productive.