Rails World 2026 Opening Keynote [video] (youtube.com)

I sat front row for David's talk yesterday morning. Having talked with him the day before, I’ll admit… I wasn’t particularly surprised (nor unprepared).

A bit of a field report from #RailsWorld: the vibe here is far from doom and gloom. Quite the contrary. Wherever our industry is headed, most of us are still employed as menders… tending to systems that customers rely on and businesses are quite happy to keep paying for.

Most of us aren’t waking up to a blank canvas and designing the architecture of the future. We’re inheriting decisions made years ago, updating old patterns, working around constraints, and keeping this shit running reliably.

I think these newer tools give us an opportunity to wonder a little more about the systems we’ve inherited. To unpack why things work the way they do. To tinker with assumptions we haven’t had the time, confidence, or permission to revisit… and share what we learn so the next person, or agent, has an easier time.

That deployment model we picked eight years ago? Worth another look. Some of our web apps probably wish they were native apps. And there are plenty of architectural decisions we’ve been living with mostly because… well, we’ve been busy living with them.

There’s plenty of understandable anxiety about what these tools mean for our work. I’m increasingly curious about what they give us permission to revisit.

Once my keynote is published, I’ll share more about a little programming-language-adjacent framework I’ve been working on for approaching exactly this kind of curiosity.

In the meantime… keep showing up. Keep wondering.

Long live Ruby. Long live Rails.

p(bloom)

Regarding the native vs web app problem, this was a completely made up human-made problem. I think I would have preferred a deterministic solution like a game engine but having agents code up native apps based on your web spec is not a horrible idea.

The bigger question to me, is the original implementation, are agents going to do that, or will humans hand code it. My two cents, I’ve been using agents heavily now, I review thousands of lines of code a day, I don’t think this works unless I have a clear idea of the spec I want to build in my mind already.

I get a better understanding of what the architecture should be and how I should build it by actually doing the thing, which means typing code at human speeds. I don’t reason my way into a good architecture, I don’t think I ever did that much in the past and I definitely don’t get it by asking the agent to check its plan and explain it to me the hundredth time. And then once I know the architecture, I don’t think the agent even speeds me up that much. It changes 7 days of coding to 1, but those 7 days weren’t worth much and I generate even more ideas and mental models in those 7 days that pay dividends for years to come.

Would you have told a tailor that using the industrial sewing machine improves his skill at making delicate, beautiful clothes? Of course not, actually even in 2026, large swathes of high quality delicate clothes are hand sewn. There will be an equivalent for coding, even DHH admits it, when he hedges with “most”.

I think, for me I’ll navigate my career to focus more on extremely high quality, delicate custom code and leave this agentic coding to the sweatshop workers. I wanted to do this anyway before agents, now I need to accelerate it. yes FAANG is sweatshop work.

Concerns with his politics aside, I do think there's a truth to what he's talking about here, and that he is just spelling out the reality that developers are, or shortly will be, facing. For many this will be deeply uncomfortable to hear.

It is notable that his perspective in this talk is very much from a developer-user side rather than someone who is responsible for the framework itself. That surprises me, and I suspect it is not a good omen for Rails.

For all of this embrace of agentic development, there is nothing here on how they are adapting the framework for this new agentic development reality. Agents do currently work well with Rails, but there's nothing here pushing things forward as best as I can see.

I know for my own projects I've largely moved to Elixir/Phoenix, for similar reasons to his use of Rust... I didn't want to have to learn it, but now I don't have to and I get to benefit from its strengths.

How a developer oriented conference can give one minute of screen time to a by now fully open, mask off, racist / fascist is beyond me. Does the Rails world don't know, or don't care?

Edit: it's his conference because the actual Rails conf banned him. Sad victory I guess.

He is a provocateur, many things to ponder. Rust fans are jumping up and down that the backend is going to be in Rust. That's not what he saying though, his rewrite snapshot had a bunch of languages with Shell being 26%, Python 13% and Rust 7%, in the previous version Ruby was 55%. His comment on Rust was "I love Rust! Rust is amazing ...if you never, ever, EVER have to look at it yourself." It's going to be interesting to see where this lands in one year when a bunch of projects have been rewritten by Agents in Rust and our Rust Hero shows up and has to maintain that code. Seems like a Zen kaon to me - if you don't read, write or have to understand Rust but the Agent does, is your program safe or performant. Your intuition of what safety is or performance no longer comes from looking at the source code but something else and in the next turn your entire Rust code base could get optimized to some IR or assembler that you no longer can even understand!

So the keynote mentions that in a perspective you are going to be a "maker of things" rather than a coder. But, nobody is asking, in a perspective, why would anybody use the stuff you "make" rather than using AI directly? All of the notion of apps disappears, in a perspective.

I really enjoyed DHH's keynote speech using his own family lineage story via the art history intertwined with technology evolution portrait paintings -> photography pictures as analogy of what's happening right now in the technology landscapes.

So far a Rust rewrite's success rate directly, probably super linearly, correlates with the presence of a good, well designed test suite. This is why Bun's rewrite and others are successful. I anticipate that this will be true for the 37Signals rewrites.

I want to learn a couple of things once they are done:

1. Once the test suite assistance is over, how do features get built in their backends and continue to maintain the same quality. LLM test writing is a bit trickier as it often favors what Randy Coulman called tautological tests back in the day. It takes a lot of work to get them not to do this, so I am guessing folks at 37Signals will read at least this part. I can't imagine zero Rust read at all.

2. Over the years they had a few innovations in their domain design, ones that Rails made easier to do, for instance their delegated type pattern. Will Rust replicate this, and how would one know without reading it, or perhaps it does not matter? Furthermore, how do newer patterns emerge? and how does our arsenal of better abstractions keep growing?

I don't have the answers but I am glad there is a chance we can learn said answers.

I was hoping there will be news on Rails 9.0. Feels like this is the end of Ruby Rails.

Hey is not even a Ruby Rails App anymore. And indirectly admitting Ruby is slow considering they manage to cut 99.9% of CPU with their rewrite in Rust.

I think the perf jump from Ruby to Rust is much greater than .NET Core to Rust. Not much will change in .NET land for the foreseeable future. I see this migration happening from Ruby and others so it is not surprising from that aspect.

Also - what he is talking about was true for well over a year. It is just that a lot of day-to-day gaps between prompt and output were closed in the meantime. In a sense, Ruby on Rails/37 Signals is a late adopter.

I don't understand what is supposed to come next. So its not Ruby on Rails anymore. What is it? Are agents going to magically convert all the RoR ecosystem to Rust or Whatever?

Serious question: Is there a framework that is better suited for agents? There must be something out there that makes agents fly through development with minimal token expense.

Next year they will rewrite rails in rust with AI

AI-psychotic and proud!

Talk is sponsored by Shopify, which also sponsored the Lex Fridman 5 hour promotion.

All DHH's AI opinions are sponcon.

But he has great flagging support and his perception management agency will suppress this one, too.

A lot has been said about DHH, but, I think the moment where you can see that the guy has _clearly_ lost it is this quote:

> It is impossibly fast. And you know what? The binary? Half a megabyte. Half a fucking megabyte. I— you could fit that on a floppy disk almost. With a little compression. This is the other payoff.

The guy now(?) thinks like a lossy abstraction that speaks in tropes.

He has become the LLM - I would say, if I didn't know that even a 9B one can tell me with absolutely certainty how much space there is on a fucking floppy.

There is no reason to get into all the politics stuff when this is where the guy is at. Basic fact retrieval.

(Pedantically, one could now argue that it was not clear which floppy format the guy was referring to. And yes, that would be true, but by thinking that thought, you have thought about the sentence more than its author has.)

tldr:

dhh says 37signal's apps for Hey (their email platform moving to native)

backend for their email platform is now Rust - written by agents

says Rust produces small binaries - cutting number of servers needed from 110 to 10. hell they could serve peak traffic on a Raspberry Pi

says now that agents write all the code - use faster / native languages.

rails or ruby hardly get a mention besides that they might be well placed to capture web apps written by agents.

That's cool, but like with most things, I think the manager types will misunderstand:

- A lot of this is looking at an existing app and/or code for reference.

- The people doing/guiding this are skilled.

- Skilled people have thought out what to make.

Otherwise they hear "20 minutes" and think AI can spit out a logical or good product from no reference or planning.

Haven't watched the video, but I've been spending a few hours the past week re-acquainting myself with Ruby before diving back into rails introductions.

Ruby is an interesting language to work with. I don't know it like Python, but it's a language that finds problems developers would run into and built solutions in to make it easier.

As for Rails, as a total novice: It is a lot of Convention over Configuration, and it's almost a second language on top of Ruby. Everything works together implicitly so you have to recognize what's happening when you look at the code. I don't know if any Ruby/Rails oriented IDEs exist that guide the developer with hints and shortcuts to how the pieces of a Rails app fit together, but it sure would be handy.

I can see how typed, high-performance languages with explicit references to all the integrations would be easier for AI to handle. But as long as humans sometimes have to look at the code, I would think Rails still has a place.

weird talk; is not reading any code _at all_ is the new standard now? I do happily ai-slop side projects here and there but seems like just going full steam ahead _at work_ is ok? sorry, I meant - is the future now? xd

DHH is "off the Rails" is not false now?

Adieu, Ruby on Rails.

He may be racist but what a great orator! (Who else does that apply to?) Can't really disagree with him about the AI stuff either. I mean maybe it's a bit premature but if it isn't true now it will be in 5-10 years.

I guess the death of Ruby is a silver lining to AI at least.

This link says ‘This video is private’ to me, here’s a working one https://youtu.be/vDjW_dRyKXY

it's funeral of Ruby on Rails