Swiss Ruby on Rails Developer
Switzerland
Joined April 2007
- Tweets3.2K
- Following109
- Followers638
- Likes458
Thibaud Guillaume-Gentil retweeted
The need to intervene on the small stuff is decaying quickly as models improve. And when models are capable enough to handle those details themselves, spending too much time there starts to feel like a poor use of human attention.
Thibaud Guillaume-Gentil retweeted
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 accurate: @claudeai Opus 5 by @AnthropicAI (by a hair). Solved 92% of runs (58 of 63). (But for a little more than half the cost, you get almost the same accuracy with @Kimi_Moonshot.)
- Cheapest: @OpenAI GPT-5.6 Luna. 73% of runs solved at default medium reasoning effort, and all 63 of its runs cost 90 cents combined.
- Fastest: Luna again, at a median of 3.3 minutes per run task.
- Best combination of all three: @OpenAI GPT-5.6 Sol. 84% accuracy, costing $0.52 and 5 minutes per run.
Read all the findings in the first full benchmark report from @evilmartians here: rubyonrails.org/2026/8/13/ag…
ALT Agents on Rails. Which model does best? August 2026. Snapshot stats of 8 models, showing: percentage of accuracy, number of runs solved out of 63, cost, recall percentage, number of steps, and minutes per run. Claude Opus 5 is listed first based on 92% accuracy.
Thibaud Guillaume-Gentil retweeted
A great new HTML-to-PDF renderer for Ruby
github.com/waka/sghtmltopdf
Thibaud Guillaume-Gentil retweeted
I released ActiveHashcash 0.5.0 this weekend. It gets a nice speedup to solve the Hashcash stamp thanks to @thibaudgg.
ActiveHashcash protects Rails applications against spam bots and brute force attacks.
Thibaud Guillaume-Gentil retweeted
I just released Insta, a new snapshot testing library for Ruby, live at @rubyconf in Las Vegas!
Extracted from Herb's test suite and inspired by @mitsuhiko's fantastic insta library for Rust.
gem "insta"
Thibaud Guillaume-Gentil retweeted
📊 The 2026 Ruby on Rails Community Survey is open!
9 years of tracking the @rails ecosystem — and we need your input to keep it going. Takes just a few minutes. Please share with your team!
railsdeveloper.com/survey/
Thibaud Guillaume-Gentil retweeted
Claude Sonnet’s p90 latency rose 44% in three weeks inside Zed.
Today we’re launching Agent Stats: a weekly view of AI agent adoption and turn times in Zed.
See the data, and what it can and can’t tell you about how people actually use agents: zed.dev/blog/agent-metrics
Thibaud Guillaume-Gentil retweeted
SQLite doesn't shrink when you delete data. My 37 MB database was taking 6.9 GB on disk. auto_vacuum is OFF by default and nobody warns you. Here's the fix for Rails.
hencf.org/blog/sqlite-auto-v…
Thibaud Guillaume-Gentil retweeted
Herb v0.9 is here! 🌿
This release focused on stability, compatibility, and deepening Herb's understanding of HTML+ERB templates.
Featuring Action View helper support, Prism AST nodes, new linter rules, a faster linter CLI, new language server features, and engine bug fixes. 🚀
Thibaud Guillaume-Gentil retweeted
Ghostty 1.3 is now out! Scrollback search, native scrollbars, click-to-move cursor, rich clipboard copy, AppleScript, split drag/drop, Unicode 17 and international text improvements, massive performance improvements, and hundreds more changes. ghostty.org/docs/install/rel…
Thibaud Guillaume-Gentil retweeted
Ruby on Rails is probably the most token-efficient way to write a real web app together with agents that doesn't immediately fall apart with security holes and unscalable decisions. rubyonrails.org/
Thibaud Guillaume-Gentil retweeted
Stacked Diffs on @GitHub will start rolling out to early design partners in an alpha next month. In the meantime, here's video of our progress so far:
(h/t for @georgebrock + team for their awesome work)
Thibaud Guillaume-Gentil retweeted
Convention over configuration set the path for 20+ years of great training data for AI to use today. Not only does this mean agents do great with Rails, but also that squishy humans can quickly and confidently review the output without a jungle of distracting boilerplate.
Thibaud Guillaume-Gentil retweeted
Happy Friday folks. This mornings first prompt: "I would like you to do a honest review of your agents md file in this repository. This file is completely written for you and I think it is too verbose. And there are other aspects that can be inferred otherwise. Run through each section and give me an idea of whether it should be removed, updated, rewritten, to something a little bit more succinct."
Thibaud Guillaume-Gentil retweeted
Absolutely essential reading blog.julik.nl/2026/01/go-fas…
Thibaud Guillaume-Gentil retweeted
> Programming with AI is still in its infancy, and yet it has already redefined what programming will be about in the immediate future... I won’t pretend to have any answers here, just assorted thoughts.
jorgemanrubia.com/2025/12/30…
Thibaud Guillaume-Gentil retweeted
Presenting the unofficial @dhh/@37signals Rails Style Guide based on the Fizzy source code 😁
gist.github.com/marckohlbrug…
