Design and engineering consultancy for devtools, AI, and cybersecurity startups. Lago, Tines, Teleport, and Whop grow with Evil Martians. We host @sfrubyconf.
Rails + AI models, which to use?
Claude Opus 4.5 is slightly more accurate and less expensive than Fable, although a tad slower. However, you can get essentially the same results with Kimi K3 for almost half the price.
We know this because we worked with @Rails to benchmark the performance of eight AI models against 21 atomic tasks and evals.
This project gave us the amazing opportunity to work hand in hand with @AmandaBPerino and the Rails Foundation on the creative task of building Rails-centric LLM benchmarking.
But also, this helps move the ecosystem forward and hopefully allows the community to keep ideating with Rails.
Congrats to @skryukov_dev and Artur Petrov for their work on this!
Win a ticket to @Rails World!
This week, if you:
- Buy a ticket for the SF Ruby Conference by Aug 30
- Post about it (e.g., "I'm going to SF Ruby, who's coming? + Pic of your confirmation email")
- Tag us
You're automatically participating for a free ticket to Rails World. Flight and accommodation not included.
Send this to your Ruby friends.
Every scaled Rails app has a part no one fully understands or dares to touch until something breaks.
Madison Sites, staff software engineer at @OrderCo_, went through this. While the map said everything was performing as expected, the data presented on dashboards, approved by tenured engineers, and shared by AI was wrong.
In her talk, she'll share her learnings and how to proceed in the future by focusing on the structure, not the map.
Get your tickets to the conference before Sep 30: https://t.co/IgG6YAxij2
We're going to Austin, TX with @appsignal to host a RailsWorld after hours event.
Join us on Sep 23rd at a secret location to enjoy the sunset, eat tacos and donuts, and mingle.
Come for the tacos and the free swag. Leave with new Ruby friends.
Sign up: https://t.co/syumOzROiW
Distribution is everyone's kryptonite... well, except for William Imoh. He's the founder of @hackmamba, an AI-native developer marketing agency that helps teams run the entire content and distribution pipeline.
We had him as a guest on @dpl_pod and he shared all the secrets to getting his clients up to 50k organic impressions each month.
We recommend you take notes as you watch this conversation on developer marketing, AEO/SEO, distribution channels, and what metrics to track.
Full conversation in the link below.
Distribution is everyone's kryptonite... well, except for William Imoh. He's the founder of @hackmamba, an AI-native developer marketing agency that helps teams run the entire content and distribution pipeline.
We had him as a guest on @dpl_pod and he shared all the secrets to getting his clients up to 50k organic impressions each month.
We recommend you take notes as you watch this conversation on developer marketing, AEO/SEO, distribution channels, and what metrics to track.
Full conversation in the link below.
I had an opportunity to catch up with @inazarova from @evilmartians this week for the Technology for Humans podcast.
We discussed the SF Ruby Startup conference, and importantly, Irina's suggestions on how to make Ruby more visible and more likely to be used by AI tooling.
Spotify: https://t.co/6GJlAvfXS1
YouTube: https://t.co/00LuyuN7zB
The full discussion includes:
• Ruby on Rails and AI-assisted development
• How LLMs choose programming languages
• AI training data and coding benchmarks
• Open source and AI discoverability
• Why Rails conventions matter for AI agents
• Building startups with Rails
• Inertia Rails, React and Hotwire
• AnyCable and real-time applications
• The San Francisco Ruby community
• SF Ruby Startup Conference
• The future of Ruby and Rails
Irina (@inazarova) had an interview with Errol (@eSquareDesign) for the Technology for Humans podcast.
It was a very interesting conversation ranging from Irina's background and how she became our CEO, to why Rails and AI go well together, and the behind the scenes of @sfrubyconf.
Unfortunately (or fortunately), Murka stole the show in the cutest way and we have proof. Here's a promo clip, except it's just Murka being funny. 🐈
Links to the complete conversation below.
Our conference is called SF Ruby Startup Conference (@sfrubyconf), but don't let the name fool you.
The program is built for engineers who can read and write Ruby and founders who are looking for them.
In this article, we explain why this conference is specifically for engineers and why you can't miss it: https://t.co/n980BFU6RD
Persona (@withpersona) is actively hiring engineers and coming to SF Ruby.
We spoke with Melanie Alcantara, technical recruiter at Persona, who told us what it's like to work there.
If you're looking to work at a growing company, in a hybrid model, and with a team that ships fast but also cares about the people doing the work, this is the place for you.
Come meet the Persona team at SF Ruby: https://t.co/IgG6YAxQ8A
An AI assistant with read-only SQL is a superhero. Until it fails privacy responsibilities and leaks sensitive data to everyone. We noted this possibility in an internal tool, moved the data boundary from prompt to Action Policy, and kept open-ended SQL: https://t.co/nEhxf3oDWx
we replaced almost 20.000 LoC that we wrote for our workflows.
creating/managing/deleting worktrees and herdr panes, setting up postgres per worktree and tearing down after auto PR, ...
Turns out: T3code by @theo combined with Lefthook by @evilmartians do it all but better.
Agents on Rails: lemans is now open source, and today four new models hit the benchmark: @openai’s Terra, @Alibaba_Qwen 3.8-27B, @AnthropicAI's Sonnet 5, and ox-alpha, a new model just released in stealth mode.
As of August 24, 2026:
- Fastest: @openAI Terra is now the fastest model, at a median of 3 minutes 2 seconds per task.
- Most accurate: Still @claudeai Opus 5 by @AnthropicAI. Solved 92% of runs (58 of 63), and has the best API recall of all the models (35%).
- Cheapest: @OpenAI GPT-5.6 Luna is still the cheapest from the models we can accurately price - just .012 cents per run.
Read the latest benchmark report from @evilmartians to learn about lemans and see how the new models stacked up: https://t.co/rEmBqMIUAL
Great opportunity to attend two Ruby conferences this second half of the year.
Buy a ticket to SF Ruby this week and you're in the running for a free ticket to Rails World.
Divided between going to SF Ruby or Rails World?
Why choose?
Buy your SF Ruby ticket this week and participate for a free ticket to Rails World. Travel and accommodation not included.
Guidelines:
1. You must buy a ticket for our conference by August 30th
2. Tweet that you bought a ticket to the conference and tag us (@sfrubyconf)
The winner will be randomly selected. We'll announce it next Monday (August 31st).
Divided between going to SF Ruby or Rails World?
Why choose?
Buy your SF Ruby ticket this week and participate for a free ticket to Rails World. Travel and accommodation not included.
Guidelines:
1. You must buy a ticket for our conference by August 30th
2. Tweet that you bought a ticket to the conference and tag us (@sfrubyconf)
The winner will be randomly selected. We'll announce it next Monday (August 31st).