<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/what-about-rails--vht0t4vji" -->

---
title: What About Rails? | daily.dev
description: A critical breakdown of DHH's Rails World 2026 opening keynote, arguing it barely addressed Rails despite being the flagship conference talk. DHH announced...
canonical: https://daily.dev/posts/what-about-rails--vht0t4vji
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: What About Rails? | daily.dev
og:description: A critical breakdown of DHH's Rails World 2026 opening keynote, arguing it barely addressed Rails despite being the flagship conference talk. DHH announced...
og:url: https://daily.dev/posts/what-about-rails--vht0t4vji
og:image: https://api.daily.dev/og/posts/vHT0t4vji.png
og:image:alt: What About Rails?
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# What About Rails?

**[Lobsters](https://daily.dev/sources/lobsters)** · 11 min read · 0 upvotes · 0 comments

## Summary

A critical breakdown of DHH's Rails World 2026 opening keynote, arguing it barely addressed Rails despite being the flagship conference talk. DHH announced he's

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://jardo.dev/what-about-rails>

## Questions this post answers

### Why is 37signals rewriting Hey away from Ruby on Rails and into Rust and native apps?

37signals is rebuilding Hey as six native applications with a Rust backend instead of a Ruby on Rails web app, because DHH now believes LLM-assisted coding removes the productivity bottleneck that previously made Rails the practical choice for small teams. He claims the new backend requires 99% less CPU and 95% less memory, though critics note this comparison conflates gains from Rust with gains from dropping the web frontend entirely.

_Teams weighing a Rails to Rust rewrite can follow real-world tradeoffs like this one on daily.dev._

### What percentage of DHH's code output in 2026 was written in Ruby versus other languages?

Only about 3% of DHH's code output in 2026 was Ruby, down from roughly half of his work over the previous two decades. He reported writing around 150,000 lines of code in August alone, mostly Rust generated by LLMs, compared to a pre-LLM average of about 30,000 lines per year, though he did not personally review much of that generated code.

_Developers tracking where Rails' creator is placing his bets can follow the fallout on daily.dev._

### Is the claim that AI makes some programmers 10x more productive backed by real research?

No, the widely repeated 10x (or 100x, 1000x) programmer productivity claim traces back to a study that actually measured differences in developer tooling, not developer productivity, and has been heavily critiqued from multiple angles. The figure of an average 10x difference does not even appear in the original paper, making it closer to folklore than a validated finding.

_Anyone citing AI productivity multipliers when comparing tools can check the sourcing on daily.dev._

## Community take

How the wider developer community reacted, aggregated from 2 discussions and 427 comments across lobsters, hackernews (as of 2026-09-27).

**TL;DR:** The discussion mostly moved past Rails itself into a broader debate about whether frameworks like Rails still make sense in an agent/LLM-driven coding world, alongside a heated, largely off-topic argument about DHH's politics and a possible community fork.

**Sentiment:** 15% positive · 35% mixed · 50% skeptical

**The case for**

- Some argue Rails' strong conventions still help even in an agentic workflow, giving agents structure to work within.
- One commenter reports seeing no evidence that LLM review bots struggle with Ruby or Rails in practice.
- A few note that trusting agents heavily can still work well for shipping real production systems quickly.

**The pushback**

- Several argue Rails' implicit, monkey-patching-heavy design and code generation conventions map poorly to how coding agents operate, requiring more explicit code.
- Some suggest frameworks like Rails may need a 'massive shift' to remain relevant for agent-driven development, which risks alienating the existing community.
- Commenters note LLM-generated Rails code tends toward poor scalability (N+1 queries, duplicated code, bloated test suites) unless carefully reviewed.
- Distrust that DHH will act in the interest of the broader Rails community given his current stance.

**By community**

- lobsters (mixed): A civil back-and-forth debates whether Rails' abstractions and conventions are a liability or an asset for agent-driven coding, without reaching consensus.
- hackernews (heated): Splinters into multiple contentious threads — UI vs. chatbot interfaces, trust in unreviewed LLM code, dynamic vs. static languages for LLM-era apps, and a sharply divisive argument over DHH's politics and a potential fork.

**Hottest debate:** Whether DHH's personal politics and conduct discredit him as a steward of Rails, sparking a heated tangent about a values-based fork versus accusations of overreach and 'wrongthink' policing.

**Open questions**

- If Rails needs to change for an agent-first world, would that alienate its existing human-centric user base?
- Are agent coding benchmarks and language/framework tradeoffs actually measurable in a meaningful way, or just opinion dressed as data?
- Will proposed community forks of Rails actually ship meaningful work or fade out once attention moves on?

**Highlights**

> If a framework like rails is the right tradeoff or not has little to with if you read the code or not. You can evaluate its usefulness in an agentic world really on some very measurable dimensions. Does the agent default to safe code? Does the agent suffer from context rot? Does the agent write code that becomes understandable in code review? Does the agent write code that allows formal verification or static analysis? Rails tradeoffs map quite poorly to agents because Ruby what it stands on already maps badly. Ruby is all about monkey patching and requires global knowledge. What methods are available on an object is still very different from within a Rails world than a standard Ruby world. All those things make review for an agent harder, and also to some degree to a human.
> — [mitsuhiko on lobsters · 1 points, 1 comments](https://lobste.rs/s/rvvqos/what_about_rails#c_pyapff)

> > a framework like Rails which has been built around code generation and scaffolding  That's really a narrow view of what Rails is. Generators are far from its main value proposition, they're just a quick convenience for starting a bit faster and with more consistency, that's all. > Agents make writing code cheap Sure. But if you don't subscribe to the idea that it's fine to send code you don't even read nor understand in production, then in my mind readability, conventions, terseness and higher level building blocks are still valuable. Both Ruby and Rails main value proposition always has been to optimize for the human, not the machine. If you go full "vibecoding" then indeed I agree it no longer makes sense. But if you keep reading that code to understand it, I think it still does make sense to optimize for the human, not the LLM.
> — [byroot on lobsters · 1 points, 1 comments](https://lobste.rs/s/rvvqos/what_about_rails#c_lkgfeg)

> Not trying to be a jerk but at this point SRE is my main use case for heavy LLM stuff. It's pretty awesome at that. Deployments and CI pipelines have become a breeze. I used to have to ask DevOps for that and wait days. I have even let agents run deployments for side projects and they seem to do better work than most humans I've worked with. It's wild they will actually read all the logs and debug problems so fast. Not something we could ever keep up with. The actual code and architecture is where it still lacking IMHO. Especially in rails... Like it will just build the least scalable features if you let it do it's thing. Ten queries for what could be one. No separation of concerns. Huge files, lots of duplicated code and then tens of thousands of units tests that just grow like a fatberg. If your app does anything serious, if you have serious traffic... you are going to need to review each session finely (and your DB schema with each deploy). It could be that this is maybe an indictment of rails more than LLMs, I guess maybe time will tell.
> — [jnmandal on hackernews · 3 comments](https://news.ycombinator.com/item?id=49843023)

> Aside from the actual politics, DHH is clearly in "fuck you mode". I certainly don't trust that he's going to make decisions in the interest of the broader rails community.
> — [flomo on hackernews](https://news.ycombinator.com/item?id=49846856)

> From the talk it’s clear, Rails needs to be forked.
> — [potato-peeler on hackernews · 2 comments](https://news.ycombinator.com/item?id=49841597)

**Source threads**

- [lobsters](https://lobste.rs/s/rvvqos/what_about_rails) · 54 points · 25 comments
- [hackernews](https://news.ycombinator.com/item?id=49839664) · 280 points · 402 comments

## Similar posts on daily.dev

- [People have been begging David to loosen…](https://daily.dev/posts/people-have-been-begging-david-to-loosen--sljfkuxj2) · Justin Searls · 0 upvotes · 0 comments
- [Plex's classic app is better than its modern redesign, so I went back](https://daily.dev/posts/plex-s-classic-app-is-better-than-its-modern-redesign-so-i-went-back-q3ymvgynj) · XDA Developers · 0 upvotes · 0 comments

---

Tags: [#rust](https://daily.dev/tags/rust), [#ai-coding](https://daily.dev/tags/ai-coding), [#rails](https://daily.dev/tags/rails)

[View this post on daily.dev](https://daily.dev/posts/what-about-rails--vht0t4vji)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"What About Rails?","url":"https://daily.dev/posts/what-about-rails--vht0t4vji","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/what-about-rails--vht0t4vji"},"datePublished":"2026-09-25T21:07:14.100Z","dateModified":"2026-09-27T00:37:00.999Z","description":"A critical breakdown of DHH's Rails World 2026 opening keynote, arguing it barely addressed Rails despite being the flagship conference talk. DHH announced...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/94d6587f96fae3ffb40450b7c681a011?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/94d6587f96fae3ffb40450b7c681a011?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Lobsters","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Lobsters","logo":"https://media.daily.dev/image/upload/s--tl8v_Fku--/f_auto,t_logo/v1698841318/logos/lobste.jpg","url":"https://daily.dev/sources/lobsters"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/what-about-rails--vht0t4vji","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"rust,ai-coding,rails","timeRequired":"PT11M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Lobsters","item":"https://daily.dev/sources/lobsters"},{"@type":"ListItem","position":3,"name":"What About Rails?"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/what-about-rails--vht0t4vji#faq","mainEntity":[{"@type":"Question","name":"Why is 37signals rewriting Hey away from Ruby on Rails and into Rust and native apps?","acceptedAnswer":{"@type":"Answer","text":"37signals is rebuilding Hey as six native applications with a Rust backend instead of a Ruby on Rails web app, because DHH now believes LLM-assisted coding removes the productivity bottleneck that previously made Rails the practical choice for small teams. He claims the new backend requires 99% less CPU and 95% less memory, though critics note this comparison conflates gains from Rust with gains from dropping the web frontend entirely. Teams weighing a Rails to Rust rewrite can follow real-world tradeoffs like this one on daily.dev."}},{"@type":"Question","name":"What percentage of DHH's code output in 2026 was written in Ruby versus other languages?","acceptedAnswer":{"@type":"Answer","text":"Only about 3% of DHH's code output in 2026 was Ruby, down from roughly half of his work over the previous two decades. He reported writing around 150,000 lines of code in August alone, mostly Rust generated by LLMs, compared to a pre-LLM average of about 30,000 lines per year, though he did not personally review much of that generated code. Developers tracking where Rails' creator is placing his bets can follow the fallout on daily.dev."}},{"@type":"Question","name":"Is the claim that AI makes some programmers 10x more productive backed by real research?","acceptedAnswer":{"@type":"Answer","text":"No, the widely repeated 10x (or 100x, 1000x) programmer productivity claim traces back to a study that actually measured differences in developer tooling, not developer productivity, and has been heavily critiqued from multiple angles. The figure of an average 10x difference does not even appear in the original paper, making it closer to folklore than a validated finding. Anyone citing AI productivity multipliers when comparing tools can check the sourcing on daily.dev."}}]}
```

