deepsquirrelnet

Flock is only one company. Someone in my town smashed one from a different company and was treated like a hero in Facebook comments. It’s not mentioned in the article.

Not that I’d want to work there given what they do, but every time I’ve been contacted by a recruiter there, it seems like it’s within a month of a mass layoff they’ve had… which is maybe just because they seem to have mass layoffs every quarter now.

They also seem to have adopted a no-remote hire policy and are in an extreme high CoL location. It’s a truly awful mix for trying to attract outside talent. I don’t know why they even bother.

Professionally, he spells his name thusly: FBI Director Ka$h Patel, so you know he’s serious.

This is really cool. Any plans to release the dataset?

Any discussion related to this topic always seems to assume everyone uses code the same way and for the same function, and then forces the rest of the world through that lens.

So here we walk around the circle one more time again, voicing our anxieties, talking past each other, waiting for the next opportunity for commentary to come in half an hour.

I think it means which site is numberwang.

> At the time of the publication, Meta admitted subcontracted workers might sometimes review content filmed on its smart glasses when people shared it with Meta AI.

They just got fired for "piercing the veil". They committed the sin of bringing attention to the invasion of privacy.

And now debt is 1 GDP year. I don’t see the problem.

I would love to be able to run frontier locally, but I think the larger importance of open weight models is price accountability.

In the US with our broken system of capitalism, it’s the only way we can tether these companies to reality. Left to their own devices, I’m not convinced they would actually compete with each other on price.

Buy nobody like to talk about how “moat” building is fundamentally anti-competitive, even in name.

Funny that self proclaimed capitalists hate the system in practice. Commodity pricing is what truly terrifies them.

This is my own take, directly related to this that I posted a little while back. The one thing that I think the article missed is the geopolitical angle they’re also working:

* We need to completely deregulate these US companies so China doesn't win and take us over

* We need to heavily regulate anybody who is not following the rules that make us the de-facto winner

* This is so powerful it will take all the jobs (and therefore if you lead a company that isn't using AI, you will soon be obsolete)

* If you don't use AI, you will not be able to function in a future job

* We need to lineup an excuse to call our friends in government and turn off the open source spigot when the time is right

They have chosen fear as a motivator, and it is clearly working very well. It's easier to use fear now, while it's new and then flip the narrative once people are more familiar with it than to go the other direction. Companies are not just telling a story to hype their product, but why they alone are the ones that should be entrusted to build it.

> In a provocative GitHub post, machine-learning engineer Han-Chung Lee argued that even rosy internal numbers that do show AI-assisted productivity gains are suspect, as they’re produced to hit adoption targets no one can effectively audit.

Isn't this fundamentally what MBAs do with their time? Keep going with this analysis, because it goes much deeper... In my experience, BI is often a house of cards. A lot of times it's just narrative crafting, just like we're all encouraged to do when we write our resumes.

Can you embellish a story? Can you invent a convincing political narrative? As far as I can tell, that's the fundamental unit of US corporation.

> Traditional call noise canceling relies on those small onboard neural networks and can have difficulty isolating your voice in very noisy environments, which results in ambient noise leaking through or voices getting highly compressed, making it difficult to hear. Anker says the larger neural network available on the Thus chip, plus eight MEMS (micro-electromechanical systems) microphones and two bone conduction sensors to focus in on your voice, in its yet-to-be-announced earbuds will have significantly cleaner call audio, regardless of the environment.

Anyone who likes good noise cancellation, which is a lot of people.

Back in the day we just called it ML. But now you have to stop for a minute to read and determine what they’re talking about, because “AI” is primarily a marketing term.

I tried it on openrouter and set max tokens to 8192, and every response is truncated, even in non-thinking mode. Maybe there's an issue with the deployment, but in your link also shows it generates tons of output tokens.

My tinfoil hat theory, which may not be that crazy, is that providers are sandbagging their models in the days leading up to a new release, so that the next model "feels" like a bigger improvement than it is.

An important aspect of AI is that it needs to be seen as moving forward all the time. Plateaus are the death of the hype cycle, and would tether people's expectations closer to reality.

> Cuyahoga Valley: There is nothing wrong with Cuyahoga Valley. Statistically, you’re from Ohio, so why not?

In college, I took an interim elective course on geology of the national parks. On the first day of class, the professor asked an icebreaker for students to say which national park they lived closest to. I said Ohio - Cuyahoga Valley.

Well some snot nosed boy scout confidently piped up that there were mostly certainly no national parks in Ohio, and the professor agreed. This is a deep personal grudge that I still hold to this day.

CnakeCharmer - https://github.com/dleemiller/CnakeCharmer

https://huggingface.co/datasets/CnakeCharmer/CnakeCharmer

This project started from a belief that llms should be better at doing python to cython code translations than they are. So we started setting a large set of parallel implementations.

Then I realized that Claude code was much better at working on the data using tools (mcp) to check and iterate. The scope transformed into an platform for creating the SFT agentic trace dataset using sandboxed tools for compilation, testing, linting, address sanitizing and benchmarking.

We still need to bulk up the GRPO dataset with a large number of good unmatched python examples. But early results using SFT only on gpt-oss 20b are quite good.

There are so many reason if you look at how it's being sold.

* We need to completely deregulate these US companies so China doesn't win and take us over

* We need to heavily regulate anybody who is not following the rules that make us the de-facto winner

* This is so powerful it will take all the jobs (and therefore if you lead a company that isn't using AI, you will soon be obsolete)

* If you don't use AI, you will not be able to function in a future job

* We need to lineup an excuse to call our friends in government and turn off the open source spigot when the time is right

They have chosen fear as a motivator, and it is clearly working very well. It's easier to use fear now, while it's new and then flip the narrative once people are more familiar with it than to go the other direction. Companies are not just telling a story to hype their product, but why they alone are the ones that should be entrusted to build it.

I am working on a large scale dataset for producing agent traces for Python <> cython conversion with tooling, and it is second only to gemini pro 3.1 in acceptance rates (16% vs 26%).

Mid-sized models like gpt-oss minimax and qwen3.5 122b are around 6%, and gemma4 31b around 7% (but much slower).

I haven’t tried Opus or ChatGPT due to high costs on openrouter for this application.

Good article, and I think the "evolution of every AI system" is spot on.

In my opinion, the reason people don't use DSPy is because DSPy aims to be a machine learning platform. And like the article says -- this feels different or hard to people who are not used to engineering with probabilistic outputs. But these days, many more people are programming with probability machines than ever before.

The absolute biggest time sink and 'here be dragons' of using LLMs is poke and hope prompt "engineering" without proper evaluation metrics.

> You don’t have to use DSPy. But you should build like someone who understands why it exists.

And this is the salient point, and I think it's very well stated. It's not about the framework per se, but about the methodology.

This is even after the Hindenburg research report that found numerous screaming red flags a few years ago.

https://hindenburgresearch.com/smci/

I worked at Micron in the SSD division when Optane (originally called crosspoint “Xpoint”) was being made. In my mind, there was never a real serious push to productize it. But it’s not clear to me whether that was due to unattractive terms of the joint venture or lack of clear product fit.

There was certainly a time when it seemed they were shopping for engineers opinions of what to do with it, but I think they quickly determined it would be a much smaller market anyway from ssds and didn’t end up pushing on it too hard. I could be wrong though, it’s a big company and my corner was manufacturing and not product development.

Interesting read! I love to see this spirit. I grew up with a different - but similar experience. Only, as an 80s and 90s kid, computers were nothing but limitations. Even when my dad built a machine with a 133MHz Cyrix chip, already a year later, it was outdated by computers with literally double the computing power.

That Cyrix machine was already miles ahead of the 386 that was handed down to me to play text based games on and learn dos through hard knocks. I remember leafing through old hard drives that had 10mb of capacity and realizing they had no value despite not being that old.

Later in college, I had the confidence to build my own first desktop with parts cobbled together from sketchy resellers. Athlon A1 single core 1ghz. Man that thing could fly.

> I tapped into Pangram. Pangram is a remarkably good, conservative model for detecting LLM-generated text. These detectors have a bad rep among techies, but the objections are often based on outdated assumptions

Turing test is really in the rearview, huh?

Humans need machines to detect if a machine wrote the text, because humans aren’t sure.

The title being misleading is important as well, because this has landed on the front page, and the only thing that would be the only notable part of this submission.

The "new" on huggingface banner has weights that were uploaded 11 months ago, and it's 2B params. Work on this in the repo is 2 years old.

The amount of publicity compared to the anemic delivery for BitNet is impressive.

Zero-shot encoder models are so cool. I'll definitely be checking this out.

If you're looking for a zero-shot classifier, tasksource is in a similar vein.

https://huggingface.co/tasksource/ModernBERT-large-nli

Does this use something like xnnpack under the hood?

4-bit quantization on newer nvidia hardware is being supported in training as well these days. I believe the gpt-oss models were trained natively in MXFP4, which is a 4-bit floating point / e2m1 (2-exponent, 1 bit mantissa, 1 bit sign).

It doesn't seem terribly common yet though. I think it is challenging to keep it stable.

[1] https://www.opencompute.org/blog/amd-arm-intel-meta-microsof...

[2] https://www.opencompute.org/documents/ocp-microscaling-forma...

That’s why I unsubbed today! Otherwise I might forget.

I love the work unsloth is doing. I only wish gguf format had better vllm support. It’s sometimes hard to find trustworthy quants that work well with vllm.

Isn’t there some kind of term for when the government controls the means of production. I’ll think about it. It’s one of those terms that’s been thrown around so loosely by this regime you knew they were going there.