Agents on Rails: We ran 8 models against 21 atomic tasks to see which were best at writing Rails code. 3 runs each: a bug report, a security finding, a feature request.
The first benchmark report with findings is now live. So: what did we discover?
As of August 2026:
- Most
Ruby on Rails scales from PROMPT to IPO. Token-efficient code that's easy for agents to write and beautiful for humans to review.
Joined April 2007
- Today we shared the first results of Agents on Rails, a new, ongoing initiative to measure how well today's leading agentic coding tools (both frontier and open-weight) actually perform on Ruby on Rails codebases. Built by @evilmartians, commissioned by the Rails Foundation.
- Three endpoints handle millions of requests a day, and most Rubyists have never seen them. In her #RailsWorld talk, @jenshenny of @Shopify opens up the #RubyGems compact index, the hidden API behind every bundle install, and walks through where the format is headed next,
- MCP is suddenly everywhere in AI tooling, but do you actually know what it is? Enrique Mogollan (@mogox) of Handshake didn't, so he built a Ruby inspector gem from scratch in Rails 8 to figure it out. And then...he kept going. Join his #RailsWorld session and leave with a clearer
- This Week in Rails is out. Updates to the framework this week include: - `alias_attribute` now works in associations again. - Active Record types can define query predicate hooks to customize how `where` clauses are built. - `pluck` correctly returns in-memory records on unsaved

