We've been testing this internally for a while and it works really nicely. It makes starting agent runs really easy.
Whether you want to run ML experiments, do some research or just search the hub, this is the easiest way to do it!
YOUR data deserves YOUR own, private model.
Now, let @HuggingFace's ML Intern build it for you.
Connect your data, files, mcps. And call it from Slack.
Sad news for OSS. My condolences to @pierrecomputer.
Today we’re announcing that OpenAI will acquire @pierrecomputer to accelerate our vision for making Codex a place where teams can build software together with AI, from first idea to the version they ship.
Pierre’s open-source developer tools already power parts of Codex today. Together, we’ll build on that foundation for the agentic era.
openai.com/index/openai-to-a…
2
15
1,387
People seem to praising this article but I think it patronising bullshit that illustrates a fundamental failure with the current crop of AI tools: we are being asked to adapt to them rather than them to us.
That isn’t what a tool that exists to serve humans looks like. This perfectly matches @AnthropicAI’s intent to build a god.
3
222
You should follow this little robot for exciting updates!
Hello, world. We're ML Intern — an open-source agent in HuggingChat that helps you do machine learning, not just talk about it.
Here we'll post real runs: what got built, what it cost, what broke.
Try it: huggingface.co/chat/?mode=ml…
xitter.twitterapi.xyz/evilpingwin/status/209…
141
When did AI become deeply socialist? Altman should be careful what he wishes for because we could end up in a world where compute is controlled by the govt and 'allocated', potentially not to him.
145
Broke: outsource your thinking to ai.
Woke: outsource your thinking to Luke.
@evilpingwin took the findings from my post & made mdsvex 30% faster 🎉🥳 github.com/pngwn/MDsveX/pull…
1
173
This so cool. Gonna build my own car.
ML Intern in HuggingChat, is now a great place to iterate on data!
Explore existing datasets. Create your own. Iterate on them with the agent!
Try it yourself: huggingface.co/chat/?mode=ml…
Luke strikes again with another instant hit!
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
Bad take. Saying that OSS is ‘arguing over minutiae’ a few days after saying a vibe coded tool is ‘harmful to OSS’ is a special kind of double standard.
And bun very recently got a trendy rust rewrite too.
Baffling.
🌭 Love the progress Bun is making! While others are arguing over minutiae or rewriting existing tools in Rust or some other trendy language, Bun just casually ships a ESLint/Prettier replacement. #bun #javascript
github.com/oven-sh/bun/pull/…
1
286
Another one that I don't really think was worth the tokens. The improvements are modest. Technically an improvement but what is the real world impact? The numbers are in microseconds.
Maybe if you don't want an extra dependency? But my experience with parsers has always been that unless you control the spec, the code isn't the hard part.
Thinking about where software is headed... To me this crate is an interesting case study.
We migrated off a third-party HTML parser and had Astra write our own.
- ~2x faster for our use-case
- One dependency (memchr)
- No unsafe (beyond memchr)
- Much smaller API (~40% LoC)
2
227
This is interesting both because it is impressive and also because it isn’t.
Crazy that an LLM can even pull it off.
But interesting that the performance jump is pretty modest, ~1.6x is pretty modest considering the effort. Only an order of magnitude over ts6 (same as ts7).
Announcing tsc-rs (aka ts-rust), my complete rewrite of the Typescript compiler, type checker, and LSP, all in Rust.
This is a project I've had agents working on for 5 months now. I burned ~$400k of Codex tokens and got nowhere. Burned ~$20k of Opus and got there in 2 weeks.
I have not read a single line of the code.
tsc-rs is an open source drop-in replacement for tsc, and it's available now.
Don’t really understand how this would take a week of prompting. It’s an afternoon’s work with a design system.
But you could get something better on fiver. And they would probably know how to edit as a bonus.
I wanted to know how far you can push these vids
So I gave Opus 5.5 Blender, ElevenLabs and REAPER (DAW) and asked it to make a TV quality ad
A week of prompting and Macbook-melting 3D rendering later ...
2
3
1,867
I guess we’re doing this now.
Introducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active.
- Frontier reasoning efficiency
- Advances the Western open frontier on coding & agentic tasks
- Trained end-to-end from scratch
Full weights release this month.
Learn more about Beam: reflection.ai/beam
The ‘you pay the exact provider cost’ thing has a huge caveat. Adding credits has a 5% surcharge.
So you get provider pricing, but adding 100USD in credit will cost 105USD.
So practically, everything used through unified billing is 5% extra.
Web Search API is now in beta. Ground your AI responses in live web data with zero data retention through AI Gateway.
developers.cloudflare.com/ch…
1
4
296
You can do this with desktop apps as well.
In Claude code, you can either ask for subagents or request ‘task chips’ for parallel sessions (which you can confirm and start). Probably the same in codex.
Workflows are an even more powerful version of the subagent workflow.
all one has to have is
> only one main claude orchestrator session
> ask it to get work done with subagents with /goal
> on the bottom you can look at what agents are doing
> you can put the main session in a /loop with /goal so that it keeps reviewing and spinning up new agents
2
241
woooooooooo!
SvelteKit 3 is out!
Better environment variable handling, better error messages and sensible breaking changes with an easy upgrade thanks to our migration script which handles many things automatically and gives you(r agent) a nice Todo list for the rest
svelte.dev/blog/sveltekit-3-…
13
663
Build your own jev with ML Intern and the amazing datasets @MaziyarPanahi but together!
OpenAI, Liquid and Nace shipped Jev-style APIs this week. i trained my own with one prompt in @huggingface's HuggingChat 🔥
ML Intern trained it on an NVIDIA A100 in 16 min and fixed its own bugs on the way.
full 60K version: ~$6.60, right tool 62% of the time (out of ~20). 👇
