marais retweeted
Astro: 14s 🐢
bongo: 2.2s 🥁
Same 10k Markdown pages, a sixth of the memory.
Edits rebuild in ~250ms.
Incremental builds.
Introducing bongo – static sites with a good beat~!
Hit it 👇
$ bun create bongo my-site
Docs: bongo.build/
GitHub: github.com/lukeed/bongo
marais retweeted
Introducing pocketty_
SSH for iPhone & iPad, made for herdr~! 🎉
🔸 Get notified when an agent on any host is blocked
🔹 Tap → you're in that exact pane
🔸 Supports Tailscale/Bonjour natively
🔹 Diffs for every agent turn
🔸 No account or subscription
pocketty.app/?utm_source=x&u…
marais retweeted
Sad but necessary:
nomeatproxy.com/
marais retweeted
Mind boggling to me that I can make a thing faster and there's always people that ask "but why?" What kind of mentality is that? The pursuit of excellence does not need justification. Also, I find in so many cases, we can't know the impact of an improvement until we do it.
For example, one I've talked about before: Ghostty's high IO throughput has enabled terminal program (emulator and TUI) fuzzing at a speed thats incomparably fast to prior solutions. This has resulted in upstream patches to resolve issues in popular projects like btop, tmux, and more.
Speed enabled that anecdotally example that lifted the tides of adjacent communities that don't rely on Ghostty technology at all. I didn't predict this.
Make things better because they can be better and let the results naturally play out.
marais retweeted
Reminder: a little while after Order of the Sinking Star ships, we will be giving out the game engine (full source) for free!
marais retweeted
🚨 FYI, fellow pkg maintainers:
(This was news to me, at least)
Bad actors are pushing "Verified" co-authored commits to further appear legitimate. They set me as the authoring user & tried to use their own GPG key to pass it off as mine.
Reported:
github.com/lukeed/resolve.ex…
marais retweeted
A few years ago, I released a tiny utility for object canon's in JavaScript.
I recently picked it back up and, well, got a little carried away. 😅
Reintroducing `object-identity` v0.2 🎉
→ 535B
→ zero dependencies
→ ~2× faster than v0.1
→ the fastest canon I've benchmarked
github.com/maraisr/object-id…
marais retweeted
In celebration of @rolldown_rs 1.0 🎉
Announcing `comptime` — a Zig-inspired build-time evaluation primitive, exposed as Vite and Rolldown plugins
This allows you run code at build time, replacing the call site(s) with the evaluated output value.
github.com/lukeed/comptime
marais retweeted
We're the CMS @leerob migrated away from.
He's got valid points. But we've got some context to add.
sanity.io/blog/you-should-ne…
I migrated cursor.com from a CMS to raw code and Markdown.
I had estimated it would take a few weeks, but was able to finish the migration in three days with $260 in tokens and hundreds of agents.
Here's how I did it + all my my usage stats.
leerob.com/agents
marais retweeted
Marais is one of the early supporters of tiny.supply. He is the 24th who made an order to be more exact.
We also worked together on a lovely brand and a landing page for one of his projects.
Just finished to tinify his entire family squad!🩶
@slightlycode
marais retweeted
Giveaway of 2 tiny avatars!
Contest sponsored by Armstrong Agency @EmmettArmstrong 🥳
To enter:
• simply give a like🩶
• not mandatory to RT or comment, only if you want it 😁
We will randomly pick the winners in 48 hours. Good luck!
marais retweeted
🚀 Just shipped: AI prompt/model management, built right into your @GitHub repo.
Version prompts, test 40+ models, and run evaluations all in the same place you ship code.
Here’s a (short!) demo from our launch at @Microsoft Build: youtube.com/watch?v=xJR5Dnz_…
TLDR:
– Version prompts like code with .prompt.yml files
– Compare outputs across 40+ LLMs
– Run lightweight evals (your own metrics or LLM-as-a-judge)
– Review and version prompt changes via standard PRs
Why does this matter?
Because in the world of AI, prompts ARE the product. If you're building with AI, the quality of your prompt/model combos defines the quality of your output.
GitHub is taking you into a future where AI isn’t duct-taped onto your stack, it’s baked (with love ♥️) right into your SDLC.
Try it out and let us know any and all feedback! We're early-stage and can't wait to iterate based on your feedback!
#MSBuild #msbuild2025 #ai
marais retweeted
You can now use Social Changelog with GitHub Models, meaning you don't need an Open AI account or key. Just use GITHUB_TOKEN in your workflow.
cc @slightlycode
Social Changelog 0.3.0 has been released!
✨ Added support for GitHub Models
Details:
github.com/humanwhocodes/soc…
marais retweeted
Ready to dive into GitHub Models?
In this GitHub Checkout episode, @alacolombiadev and @damovisa explore models such as GPT-4.1 & GPT-4o mini, compare outputs, demonstrate GitHub Actions integration, and more.
See how you can experiment with models for free 👇
gh.io/checkout-models
ALT Promotional graphic for 'What is GitHub Models?' featuring the logos for various AI models along with a screenshot of code on a dark computer screen.