1. Will AI Agents Replace Bubble Developers? A Father-Son Debate

It's a new week in the No Good Wizardry podcast slash show slash conversations with a dad and his son on tech, AI, building with Bubble. Welcome. Hey. Yeah, you've got the most cringe ways of saying hello, but hi, how are you? I'm doing good. Anyways, today we wanted to talk about agents and whether agents are here to replace us all. I mean, we're all using it, or at least if you're building with Bubble, you've probably come across an agent here and there. You've probably used Claude or you've used Codex or you've used ChatGPT in some way or another to automate some of your workflows. And I think it ties in really well with the conversation that we were having last week, Dennis, on, you know, what does the future of web look like now that agents are here? Are they just going to replace us all? What do you think?

Well, I hope they're not going to do it in the next week or two. Uh, it is a really interesting question because they are surprising us all. I mean, you know, I first hear— I mean, you can do some really amazing stuff, and they're getting better, and they're getting better really quickly. So, uh, I don't think that they're going to replace us all or even replace any of us really, but I think they are going to augment us all.

Mm-hmm.

Elaborate on that. And from a business owner point of view, I think that we are clearly now in the realm of hybrid organizations where humans and AI agents collaborate and work together to solve problems for their customers. And it speaks really well for, for the underlying mission that we have had ever since, you know, before you, when you were a little kid and it was just me, uh, the David versus Goliath mentality of small row teams competing against big organizations and winning. It's never been easier. So I think that, uh, this is the beginning of the age of small teams being super powerful. And part of that is because you can have virtual agents that collaborate with the humans that bring the humanity and the ingenuity and the creativity to the table, and machines that do the grunt work, that make sure that the grunt work is done correctly, which has immense value. I mean, there's no question about it. You know, nobody, nobody really wants to talk about the sexy part of bookkeeping, but no business survives without good bookkeeping. Well, that's just one example. So, and there's so much that goes on. There's so much grunt work in marketing. There's so much grunt work in software development. There's so much grunt work in almost any aspect of any business that really subtracts from what we're capable of delivering as human beings because we're putting all that effort in. We'll let the machines do that.

Um, so can you come up with an example of how we've been able to call— I mean, we at Raysha, we're a small team of couple of developers, some designers, some assistants here and there, but can you give an example of how agents are really helping us compete with, for example, bigger bubble agencies out there?

Sure. Recently a customer came to us, hired us. They were, as happens every once in a while, they needed rescuing. They had an, let me call it, a big-ass complicated mess on their hands. it was divided into a Bubble application and an InnateIn backend workflow and a bunch of other pieces that were kind of coupled together. And it was— there was all sorts of spaghetti going on there. And, uh, you know, one of the first big tasks that we took on was to audit and, uh, clean up that big InnateIn server, backend server, that they had created with that had gone through several iterations with different development teams, and it was just a mess. What could have been easily weeks and weeks of manual labor, we were able to accomplish in, in, in days because we were able to employ AI agents that helped us go through the nitty-gritty of, you know, which of these webhooks are not using the version, are using, you know, production and not test, or are, have, you know, syntax errors in, in JavaScript code.

Cool. Uh, I'm just going to be super outright forward with you. Is Buildprint going to put us out of business? No. Is Buildprint going to allow us to deliver better work?

Yes. And faster and more economically feasible for our customers. Uh, and I think even the biggest advantage is, is that it's going to ruin things for agents that— and agencies that don't know how to use it correctly. It's going to create massive problems, and then we're going to be the saviors, which has already happened. Yeah.

Um, let me ask you, let's take it outside of the Bubble ecosystem. Can agents really hurt Bubble in a way? Can people access, uh, you know, programming methods to build applications that don't even involve no-code?

Uh, for test prototypes, probably yes. But if you want a long-term solution that solves a real problem and is sustainable and can can be maintained. There's nothing better than having the entire Bubble ecosystem behind that because it handles everything from UI/UX to database to backend servers, uh, to API connectors. It is a legitimate, solid, full-stack platform, uh, that is able to be used in a very fast way. So I think that even now, uh, first of all, the actual human knowledge of how to do it correctly just gets more and more valuable. Okay. And the ability to deliver on time, in budget, and actual scalable, maintainable solutions— I haven't seen anything better yet. Mm-hmm. But yeah, if you just want to build a mini tool that helps you, you know, do some tiny little thing, you probably get Claude to do it and it'll work okay. But, you know, beware, you're going to maintain the infrastructure, you're going to maintain, uh, you're going to be debugging when things go wrong. Things go wrong and they will go wrong. Yeah. Or when the database version changes or the, you know, the Docker, uh, version changes or the Linux version has to be updated or whatever that all of that mess is. It's yours, baby.

And this is just out of curiosity. Um, of course we're always speaking inside the tech sector and we're always bringing it and tying it back to the idea of building applications, building websites, building online infrastructure. But are you on a personal level sad or concerned about other industries that have nothing to do with ours in terms of like, okay, are agents actually going to be killing and destroying jobs in are families at risk out there? For example, I keep hearing politicians mentioning truck drivers.

Yeah, I mean, I think the short and the long answer of that is yes. I think there are risks involved. I think that there are changes coming that nobody can just, you know, you can put your head in the sand and ignore it, but You know, if it's your brother-in-law, your sister-in-law who ends up not having a job, it's going to get real, okay? And some of that's going to happen because, you know, there are jobs that are probably doable by machines, and if they're not doable now, they will be soon. But I think that that just forces us to become more clever, you know, let's face it, uh, 150 years ago you could make a decent living on a small farm in, in the middle of Nebraska. And, you know, that probably isn't the case anymore. You have to have a huge farm with lots of machinery and lots of automation and lots of systems in place to be— you know, farming is now a big business, uh, and I think things are changing.

It's just concern. I mean, I, I keep thinking about the idea of potentially raising kids in these days. And it's like, okay, what do I tell my kid? He should study. What should I, what should he be preparing for? It's kind of really hard. You don't know what, I mean, I don't know what the future, what the world is going to look like in, in 10 years.

Well, I didn't know 10 years ago either when I was hoping to help you along. Right. And, and I think I, you know, you done a good job. No, I'm not here, you know, uh, we'll, we'll see. But I, you know, I think that problem solving, learning to think, learning to communicate. I don't think those skills are going away. Right. I don't think that that is in— those are the kinds of abilities that are not going to be replaced by agents. But for example, actually taking a good story and turning it into a movie, that may be something that happens before too long. That you just take a really great novel and you go to some, you know, AI platform and you say, can you make this into a movie for me? And half an hour later, it's going to spit out a 90-minute movie that is all about that book. And that's just the reality of where we're going. But somebody's got to write the book, right?

Someone has to have the idea, right? Unless you have anything else that you want to mention, I think—

Yeah, I do think though that there is one thing that I think is really important that we talk about, and that's about being a— playing nicely, being a good neighbor. I'm talking from a software point of view. I think this is really, really important for all of our customers, all the people out there that are thinking about building software. As a revenue stream in your business, you have to play nicely. And that means you need to start thinking about how do I make my application, which does something really well for a specific problem in a specific industry, how do I make it usable by agents? How do I make it communicate well with agents and make it so that an agent can come in and log into my platform and do stuff?

So what you're essentially saying is you gotta have, your application needs to have an MCP.

It needs to have an MCP server, which by the way, have a great API architecture underneath that.

Small call to action here. We actually did post a video on how you can create a, a, uh, an MCP server for your own Bubble application. Dennis did a really wonderful video. If you wanna go watch that, it's on our YouTube channel.

Uh, and I think that that is a really important message for us to get out in this podcast episode, that it's no longer good enough to build an application that solves a problem for people. It needs to solve the problem for people, and it needs to play nicely in the wider ecosystem of software solutions and agents and AI that's already out there. And that means that you need to get your ducks in a row. And if you have questions on how to best do that in your specific Bubble application, I mean, that's kind of what we're here for. We're good at it. So, uh, I think that that's, uh, you know, it's— and it's very doable.

So cool. Well, I think that's, uh, that's a good place for us to wrap it up. If you guys have any questions, any comments, any concerns, if you want to rant about anything, you've got the comment sections to be able to do that. Pleasure as usual to share some, some talk time with you, and I guess I'll—

we'll see them next week. See you next week.