building Software with Ruby on Rails since 2009

Joined February 2012
465 Photos and videos
Equivalent retweeted
DHH: Basecamp 5, Vibe Coding, and the Future of Rails Main point: @rails is not “old tech”. Rails may become even more valuable in the AI era. Why? AI works better with: > conventions > clear structure > stable patterns > mature APIs > less framework churn Rails code from 10–20 years ago still feels familiar today. That means Rails knowledge compounds. You don’t need to relearn the framework every 2 years. thanks @robbyrussell #RubyOnRails #Rails #Ruby #AI #Basecamp @37signals @dhh
7
11
109
4,048
Equivalent retweeted
Reminder: most suggestions you’ll read about “killer prompts to use with Fable” are not good for Fable. Anthropic published a guide for how different prompting is with their new token torching model: platform.claude.com/docs/en/…
11
56
582
65,271
Equivalent retweeted
Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use. Its capabilities exceed those of any model we’ve ever made generally available.
5,004
14,521
104,714
55,852,706
loop ➰ loop ➰ here and a loop ➰ loop ➰ there..
36
Herdr is really good, testing it entire day. Goes soooo well with my "run agents on VPS" way of doing stuff. I just do `herdr --remote myssh` and thx to WezTerm I can even paste images. Recommending a lot ! herdr.dev/
2
113
herdr.dev/docs/cli-reference… As soon as ai meet the creator of Herder I will hug him and whisper to his ear "Thank you"
35
👀
Andrej Karpathy spent 2h showing how he actually uses AI day to day he's a co-founder of OpenAI and led AI at Tesla, so when he shows how he works, it’s worth watching and the whole session is just him telling the machine what he wants in simple terms, like he's briefing a coworker watch what's actually happening the entire time: > he describes the task in normal words > it goes off and does the work > he glances at the result and nudges it with one more sentence that's the whole skill, and you've had it since you learned to talk the only gap between that and a worker that runs on its own is handing that sentence a schedule and the tools to act check his work, then build the version that keeps working when you stop
108
Herder looks interesting youtu.be/PlN86TvzGy4?is=STJX… given I use 2 Linux laptops (along 1Mac) it sucks not to have CMUX on them, this could be an answer 🤔testing it on monday
53
feels like every 2 weeks I look back how I produced features 2weeks ago and I'm like "what a noob"
1
16
Equivalent retweeted
235
1,852
31,916
1,185,555
I have one agentic coding trick that nearly no-one uses yet it will guarantee improve what Claude/Codex/... produces, You want to hear it? 🪄Last sentence of your prompt you end up with "do you have any questions?" ... Talk to the Model !!! before it start producing slop
1
25
Equivalent retweeted
Haha! Good one
Using this to explain my day to day coding life
1
1
117
huh, I never thought of that 🤔This level grapplers think on whole different level about such details
GSP left Mighty Mouse speechless with his takedown technique.
13
I'm actually paying $20 plan on all 3: Claude, Codex (ChatGPT) and Cursor just to figure out which one works best in which flow
1
44
Pay attention to Cursor, They really stepped up their game. They are no longer just VS Code with an Agent Chat window. They introduced separate agenetic app similar to Cursor APP (Which actually works on both OSx and Linux) The Composer 2.5 model is much better as it used to be
1
28
Lexxy is now on prod Basecamp 5
Lexxy is now at a production-Basecamp quality level. Getting there has required a huge amount of work. Amazing effort led by Samuel Péchèr and @zoltanhosszu. github.com/basecamp/lexxy/re…
1
119