Does Ruby on Rails Still Make Sense?
random internet person rambling
Opinion — commentary, not a factual news event.
Rails still makes sense even with LLMs writing code. The point isn’t speed alone; rewrites can change behaviour users already rely on.
Based on reporting by random internet person rambling — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
A new question keeps popping up: if LLMs can write a lot of code, why stick with Ruby on Rails at all? The answer here is blunt — because the old reasons still apply. Frameworks, libraries, languages, databases and operating systems all exist to keep humans from rebuilding plumbing every time they want to ship something useful.
That logic does not disappear just because an agent can type faster than a person. In fact, the argument goes the other way: if a framework makes a developer more productive, it should make an LLM more productive too. Rails still gives a predictable structure, security updates, performance improvements and new features when it gets updated. It also keeps a project easier to navigate, reason about, debug and extend.
The temptation, of course, is to say: just rewrite it in Rust. But a rewrite is not only about generating equivalent code. Different languages and ecosystems make different choices, and changing them can alter behaviour in ways that are hard to spot. Some of those changes might fix old bugs. Others could introduce new ones, including ones users notice immediately because they depend on the quirks of the current system.
That is especially true for user interfaces. This software serves people who use it for hours every day, and even small visual changes can trigger complaints. A full rewrite means a lot of testing, not just a lot of code generation. And if the result behaves differently in subtle ways, the cost is not theoretical — it shows up in trust.
There are also easier wins. Instead of swapping languages, the piece argues for asking an LLM to improve performance in the existing application. Updating dependencies already brings security checks and improvements for free, and future Ruby releases could make existing code much faster and lighter on memory. There is still room to explore other tools — Go for background work, Svelte for the frontend — but the default answer is still Rails for now.
My take — AI-written commentary, not fact-checked reporting
This is the sane take, which is why it’s annoyingly unfashionable. The industry loves turning every rewrite into a morality play about “modern” stacks, when the real question is whether users will notice the difference for the better. Most of the time, they notice the regression first.
Read more about this at: random internet person rambling
