I've been writing a lot of code recently, of course with lot of help from Claude and Codex. I mostly write in Ruby on Rails and I am falling in love with RoR all over again.
I've never really felt the urge to switch from Ruby on Rails to JavaScript. Not because JavaScript is bad. But because Ruby has always felt natural to me.
It's simple. It's expressive. It lets me think about the product instead of the ceremony around building it.
And now, in the age of AI-assisted coding, Rails has become even more valuable to me.
The reason is simple: AI doesn't just write code. It makes decisions.
Every time an AI assistant generates a controller, a model, a service object, a migration, a view, an API endpoint, or a background job, it's making dozens of small architectural choices.
Where should this logic live? How should the data flow? What should the folder structure look like? How should validations be handled? What should be abstracted, and what should stay simple?
In a framework with weak conventions, AI has too much freedom. And too much freedom breeds inconsistency.
One file follows one pattern. The next follows another. A third introduces an unnecessary abstraction. Before long, the app still "works," but the codebase feels like it was assembled by a dozen developers who never spoke to each other.
That's where Rails shines. Rails gives AI a path to follow.
The framework already has opinions. Models go here. Controllers go here. Views go here. Migrations follow this pattern. Routes follow this structure. Validations belong in the model. Background jobs, mailers, and tests each have their place. Security defaults aren't an afterthought.
Caching, Active Record, naming, structure, conventions, integrations — all part of the same ecosystem.
This matters because AI performs better when the environment has strong patterns.
Rails isn't just a framework. It's a set of decisions made in advance. And that's powerful.
As DHH put it:
"Rails is opinionated software. It makes the assumption that there is a 'best' way to do things, and it's designed to encourage that way."
That opinionation used to be a convenience. In the AI era, it's an advantage.
For human developers, convention over configuration has always meant speed. For AI-assisted development, it means something even more important: reduced architectural drift.
When I ask AI to help me build in Rails, I'm not starting from a blank canvas. I'm handing it a well-marked road.
It still needs guidance. It still needs review. It still needs a developer who understands the product, the domain, and the trade-offs. But Rails reduces the number of unnecessary choices the AI has to make.
That makes the output easier to review. Easier to reason about. Easier to maintain.
And this is also why I love Ruby. Ruby doesn't feel like I'm fighting the language. It reads closer to how I think.
The simplicity of Ruby isn't just aesthetic. It changes the way I build. When the language is expressive and the framework is opinionated, I get to spend my time on the customer problem, the workflow, the data model, the business logic.
That's where real value is created. Not in arguing about folder structures. Not in wiring together ten libraries before the first feature works. Not in reinventing authentication, background jobs, form handling, routing, or deployment for every small product.
Of course, Rails isn't perfect. No framework is.
Large Rails apps can still turn messy without discipline. AI can still over-engineer. It can still misunderstand the domain. It can still produce something that looks correct but quietly creates technical debt.
AI doesn't remove the need for engineering judgment. It raises the penalty for not having it.
But that's exactly why I prefer Rails. It gives me leverage without removing responsibility. It gives me speed without forcing chaos. It gives AI enough structure to be useful, while leaving room for a good engineer to make the decisions that matter.
For me, the AI coding era hasn't made Rails less relevant. It's made Rails more relevant.
Because the future of software may not belong to the language with the most hype. It may belong to the ecosystem that gives both humans and AI the clearest path from idea to working product.
And for many products — SaaS apps, internal tools, business workflows, dashboards, automations, MVPs — Ruby on Rails still feels like one of the best paths available.
Simple. Opinionated. Productive. Human-friendly.
And now, surprisingly, AI-friendly too.
I built Formserve.IO using Ruby on Rails, check it out
Top comments (0)