Benjamin Muskalla

Some of the best engineering best practices were established under different economics. Agents are repricing the work. Some practices protect us from fundamental problems. Others protect us from costs that don't exist anymore. Those have a half-life. bmuskalla.dev/blog/2026-09...

Today we're releasing Laguna S 2.1, our most capable model to date. It's a 118B total parameter MoE model with 8B activated per token, a context window of up to 1M tokens, and thinking and no-thinking modes. Small enough to run on a single @NVIDIAAI DGX Spark poolside.ai/blog/introdu...

Today we’re publishing the technical report behind Laguna M.1 and Laguna XS.2. This report opens up more of what went into them: Model Factory, pre-training data, distributed training, post-training, agent RL, quantization, and evaluation. poolside.ai/assets/lagun...

As agents get more clever, so do their attempts at benchmark hacking. Last Monday, we found one of our RL runs jumped ~20% on SWE-Bench-Pro over a weekend, reaching ~64% which would make it #1 on the leaderboard. These were our findings: poolside.ai/blog/through...

Today we’re shipping Laguna M.1 and Laguna XS.2 – our first public models. We’re also shipping our agent harness and a preview product experience. Both models were trained from scratch on our own stack: data pipelines, training infrastructure, and agent RL. poolside.ai/blog/introdu...

you cannot compete with someone who is having fun

I clicked a link and ended up on sourceforge.net - AMA

We're still looking for talent across the board, be it in inference, training, backend or IDEs. Feel free to ping if you want to know more. jobs.ashbyhq.com/poolside?utm...

ikea captcha

Why don't we include this on every website? So true.

Need a dev advocate? Look no further, Mark is a rock star

branch names often make it obvious how stupid a bug was: bmuskalla/stopAtEnd

how you know you're still a startup? someone just mentioned "long term" and "2 weeks" in the same sentence

Excited to announce the official launch of our online book, WebAssembly from the Ground Up! 🎉 It's the book we wish we'd had 3 years ago. No messing with tools and frameworks. It's a hands-on guide to the core of Wasm: the instruction set and module format. Link below. 👇

Reminder: if you haven’t yet read “The Mythical Man Month,” buy two copies so you can read it faster.

my personal AGI benchmark is implementing atproto oauth from scratch. ultimate evidence of beyond-human-level intelligence

Saving papers or articles to read? Just mark them as stale every morning; last 3 things I pulled from my "to read list" were outdated already. Sigh Anything interesting you got that I can mark as stale by tomorrow?

New prompt technique: stop being smarty pants. nobody wants code comments for indecipherable code. write readable code instead. thank you.

E_TOKENIZER_GONE_ROGUE

It's funny how people think that AI tools out there can just implement a "model switcher" and be independent of the underlying model behaviour...brace yourselves. SOTA doesn't mean equal

this s1 paper is so refreshing! Love seeing ML breakthroughs that don't need massive compute or complex tricks 🙌 Clean, simple, and it actually works? More of this please! github.com/simplescalin...

I don’t care how hacky it feels but I really dig `uv run —with=“dep>1,<2” myscript.py docs.astral.sh/uv/guides/sc...

Across various programming languages I work with, LLMs became more reliable in fixing imports properly than language tooling itself 🤔

Tiny TIL: I just figured out how to run pytest with a different Python version against my pyproject.toml/setup.py projects using uv run uv run --python 3.12 --with '.[test]' pytest til.simonwillison.net/pytest/pytes...

Programming languages: "We are just a way to operate computers in a way that makes sense to humans." Programming languages [takes a big joint hit]: "What if there were 5 kinds of nothingness?"

When you just thought you‘ve seen everything in the hype cycle. Germany goes

We’re building a new static type checker for Python, from scratch, in Rust. From a technical perspective, it’s probably our most ambitious project yet. We’re about 800 PRs deep!

transposed[:,:,:,:max_idx] not sure if code or accidently spilled coffee on the keyboard

People really seemed to enjoy the exploding code minigame from my @intellijidea.com as a game engine talk, so I figured I should release it as a plugin. You can download it here 💥💣: plugins.jetbrains.com/plugin/26424... Have fun! And if you find any bugs or have feature requests, let me know!!!

Such exciting times here at poolside. Let me know if want to join the party to re-think software engineering with AI from the ground up poolside.ai/careers

Agentic tool: I'll solve this problem in a super elegant way by combining 2 separate things into a single call Agentic tool: That didn't work. Let's try again ... Agentic tool: (round 10) That didn't work. Let's try again Benny: Ok Smarty pants, stop it. Nobody cares about it being 2 calls.

Want to easily reduce your companies CO2 footprint? 🏭Finally disable those lint builds that retrigger a full build pipeline because of that missing space 🍃🌱

Null pointer: I'm a billion-dollar mistake. I am the worst AWS Regions: Hold my 🍺

Just typed `git aloas` to setup an alias - oh the irony

Been surprised that github.com is the only place that the hotel wifi doesn't allow connecting to. Hmm..ssh key broken? No..Maybe DNS? It gotta be DNS...No...oh, it is literally github.com being down and me blaming the poor hotel wifi

“Scientists believe all the gold on Earth formed in supernovae and neutron star collisions that occurred before the solar system formed. In these events, gold formed during the r-process. Gold sank to the Earth's core during the planet's formation” www.thoughtco.com/how-is-gold-...

Can't allocate CUDA memory...who is running inference on this GPU??? Yeah, it was the other shell I had open

"Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead." Erm...yeah...thanks, I guess

"Generating your answer..." reads quite similar to "Hold on while I'm making things up..."

Wow, just stumbled over some oss code I wrote 17 years ago. Crazy that some of this actually still runs in production github.com/agavi/agavi/...

💡 idea: ai agent to automatically archive emails after a period of time and send an apologetic reply: "I'm sorry, I'm an llm responsible for protecting Daniel's mental health and I've archived this email for his own good. (he intended to reply.) if it's important, do remind him!"

🧑‍🎄 Node.js now supports TypeScript by default 🤯 🎄

What’s your favorite use case/tool/approach for LLMs right now when it comes to maintaining software and related artifacts (so not producing new code)? 🛠️

The most important thing is the thing that’s nearly finished