Dennis Dashkevich
@dskecse
Ruby, Rails, JavaScript, Go, Python, Postgres, Redis, Kafka, Elasticsearch, GCP
Several people asked for the slides after my GoLab talk, so here they are speakerdeck.com/konradreiche/w…
This is excellent - long, information dense, full of actionable advice on getting the most out of GPT-5 and Codex CLI
📢 Time for an update on my workflow. This one's a 23 min read, so buckle up. 100% organic and hand-written, like an animal. steipete.me/posts/just-tal…
Excited to release new repo: nanochat! (it's among the most unhinged I've written). Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single,…
With gpt-5-codex, some will feel that it's the best thing since sliced bread, some will see its quirks. Personally it is my favorite collaborator for coding, fast for small things, working hard when it matters and a solid jump in produced code quality. $ codex -m gpt-5-codex
Both DSPy and (especially) GEPA are currently severely under hyped in the AI context engineering world
Introducing Tidewave Web for Rails and Phoenix: a coding agent that runs in the browser alongside your web application, with full page and code context. Tidewave deeply integrates with your stack, from the database to the UI, making AI development more seamless and efficient.
Here's the concluding section of my write-up of the new OpenAI open weights models - they're really impressive, and I think they hold their own (or even beat) the models we've seen come out of the various Chinese AI labs over the past few months
Regarding agent-context, I wrote a specification github.com/ioquatix/agent… I think `CLAUD.md` and `.cursor/rules` are anti-patterns - we should have an open specification for sharing context between packages and projects.
I like this take by @KentBeck on how AI-assisted programming changes the balance of which skills are most important
Great reminder of how engineers should provide estimates to their managers: no hard dates, only a range, a probability distribution, with optimistic and pessimistic ETAs. And you can only promise to update your ETAs daily
Take advantage of `Array#inquiry` for more expressive and idiomatic code! 😊 For example, here we use it to simplify feature toggling by avoiding constantly writing `include?('feature_name')`. #rubyonrails #activesupport
I just opened a new PR to ohmyzsh to fix the "chruby" command (and its "rubies" alias) output and hopefully speed up the whole loading process a bit for the chruby plugin: github.com/ohmyzsh/ohmyzs…
I'll say it again, if you're not using jemalloc on your Ruby web application, you're missing out on saving ~50% of your memory usage. Here's yet another graph of us doing that for a client this week.
A bit outdated but still great article on how to work with dates in Rails and not only nandovieira.com/working-with-d…
One of the best productivity tips I have ever learned at Framer (from @eelco) is: Whenever you want to ask a teammate for help, don’t DM them. Instead, always post your question in a channel (and cc them).
Rails 8.0: #NOBUILD, #NOPAAS, all-in on SQLite as a production database option with jobs, cache, and cable, new authentication generator, and so much more! Final release is out 🎉 rubyonrails.org/2024/11/7/rail…
Kamal Handbook, 2nd edition for Kamal 2 is now available as a beta release🎉 Kamal 2 is even better than Kamal 1: - automatic SSL/TLS for small apps - multiple apps on one server - faster deploys for more servers Thanks to everyone for the support 🙏 kamalmanual.com/handbook/
This is why I love Cloudflare. Owning your own hardware doesn't mean you can't contract with a bodyguard. We do that and it saved our bacon a few months ago when we were attacked.
Not all records you’re happy about breaking: @Cloudflare recently mitigated the largest ever reported hyper-volumetric #DDoS attack. 3.8 terabits per second (Tbps) and 2.14 billion packets per second (Bpps). Handled automatically any without any customer impact. Details to come.
Because @fxn doesn't stop at simple explanations, I went to dig into why this benchmark give the opposite results of what one would expect 🧵
United States เทรนด์
- 1. Mike Evans 7,239 posts
- 2. Gibbs 13.2K posts
- 3. #WWERaw 17.4K posts
- 4. Lions 62.4K posts
- 5. Bucs 14.3K posts
- 6. #OnePride 4,842 posts
- 7. Josh Naylor 2,846 posts
- 8. Goff 6,810 posts
- 9. White House 215K posts
- 10. $BIGMI 1,651 posts
- 11. #TBvsDET 3,038 posts
- 12. Dragon Lee 1,094 posts
- 13. Baker Mayfield 4,697 posts
- 14. Ben Solo 9,520 posts
- 15. FanDuel 23K posts
- 16. East Wing 39.9K posts
- 17. Bron 18.5K posts
- 18. #RawOnNetflix N/A
- 19. #MondayNightFootball 1,074 posts
- 20. Game 7 54.6K posts
คุณอาจชื่นชอบ
Something went wrong.
Something went wrong.