There is something slightly funny about the way we talk about no-code today. Sometimes it gets presented as if someone suddenly invented the idea of building technology without writing everything manually. In reality, we have been moving in that direction for a very long time.
WordPress started in 2003, while Joomla arrived in 2005. They were not application builders in the way we understand no-code platforms today, but they helped normalise something important: you did not necessarily need to write every line of code yourself to build and manage something useful on the web.
Over time we added visual editors, plugins, page builders, workflow tools and increasingly sophisticated abstractions. Nobody really questions whether using WordPress instead of building a CMS from scratch makes you less serious. We understand that the tool is simply helping you work at a different level.
For me, Bubble is a natural continuation of that idea. It was founded in 2012 around lowering the technical barrier to building software, but what I like about it is that it doesn’t feel like pressing a button and hoping something clever happens in the background. You still design the database, workflows, permissions, conditions, APIs and application logic yourself.
This is also where I see a big difference with some of the conversation around AI today. AI can generate an impressive amount of software very quickly, but there is always a risk that you end up owning something you don’t properly understand. It works, great. But do you know why? Do you know what happens when a permission is wrong, a workflow behaves differently in production or sensitive data is exposed?
With Bubble, the code is abstracted, but the responsibility isn’t. You can see your workflows. You define your data structure. You decide who can access what. You build the logic. That control is important to me because I don’t want to create software by simply trusting that something somewhere generated the right solution.
That absolutely does not mean anyone can suddenly build serious software just because the interface is visual. The tool might be accessible, but building a secure and reliable product still requires experience. You need to understand databases, authentication, privacy, permissions, application logic, performance and what can go wrong when real users start doing things you never expected.
I don’t see no-code as being against developers, and I don’t think Bubble removes engineering from software development. Quite the opposite. For me it removes a lot of the repetitive implementation work and lets you spend more time thinking about the actual system. A badly designed application is still badly designed, whether it was written in JavaScript, PHP or built visually.
That’s probably why I’ve never really liked the argument of code versus no-code. The serious question is whether the person building the product understands what they are doing. Bubble gives me speed without completely taking that understanding away. And for the kind of software I like building, that balance between abstraction and control is exactly why it works.
