codemog

Don’t work for evil companies.

South Park made a parody about that where the cure is injecting literal money into his veins. I don’t know if you’re aware, but most Americans don’t have access to these kind of things. If anything, it’s simply demoralizing to know your loved ones get to die and the billionaires and their families get to live.

Makes sense. Thanks for the explanation.

Was this AI modified or something? The writing is very odd. I wish it could go for more than 3 sentences without talking about how great our genius billionaires are.

Also this isn’t that interesting: billionaires have access to the best teams and clinical trials available. Of course?

Hey you’re welcome to your view. In my view that’s an insanely disgusting devaluation of human life that’s been put into peoples brains by propaganda. Capitalism and materialism sure is great huh.

Curious why the model architecture wasn’t talked about at all? Did I miss that part?

Nobody should have these types of problems in the age of AI agents. This kind of clean up and grunt work is perfect for AI agents. We don’t need new abstractions.

Meta doesn’t even need to use this, they’re just going to be constantly recording all video and audio from those rayban glasses ;)

Smart move, Zuck.

I haven’t seen tinygrad used for any mainstream production project or thing of value, yet.

Besides a lot of self congratulatory pats on the back for how elegant it is. Honestly, when I read it, it looked confusing as all the other ML libraries. Not actually simple like Karpathy’s stuff.

All that to say, I do really want it to succeed. They should probably hire some practical engineers and not just guys and gals congratulating themselves how elegant and awesome they are.

> Compliance isn't that hard once you stop looking for shortcuts and start spending time doing it correctly.

Trying to understand how someone can have this perspective when it’s usually someone’s full time salaried job in a lot of companies.

> If you aren't actually reinvesting capital, pay your damn taxes. Don't be an asshole.

Why? So my government has more missiles to blow up children? No thanks.

This reads like a scattered mind with a few good gems, a few assumptions that are incorrect but baked into the author’s world view, and loose coherence tying it all together. I see a lot of myself in it.

I’ll cover one of them: layers of management or bureaucracy does not reduce risk. It creates in-action, which gives the appearance of reducing risk, until some startup comes and gobbles up your lunch. Upper management knows it’s all bullshit and the game theoretic play is to say no to things, because you’re not held accountable if you say no, so they say no and milk the money printer until the company stagnates and dies. Then they repeat at another company (usually with a new title and promotion).

My friend, there’s tons of evidence of all that stuff you talked about in hundreds of papers on arxiv. But you dismiss it entirely in your second bullet point, so I’m not entirely sure what you expect.

Let me guess, it’s ok if they do it, but if you handed their crappy ticket to Claude and shipped whatever crud came out, you’d be held accountable? ;)

Funny how that works out.

As soon as MCP came out I thought it was over engineered crud and didn’t invest any time in it. I have yet to regret this decision. Same thing with LangChain.

This is one key difference between experienced and inexperienced devs; if something looks like crud, it probably is crud. Don’t follow or do something because it’s popular at the time.

Greatest hype man of all time and shows how whacked out reality and economics are.

Anti-tech contrarian sentiment happens with every new technology. Someone older than you probably said the same thing about the internet.

Personally, I don’t know how you can say the timing was never right and will never be right at any point in the future. That frankly seems impossible, unless it was something like a B2B SaaS that gouges out your eyeballs, but I guess we’ll agree to disagree.

It’s not arrogant to be firm in your beliefs. You’re not arrogant for believing the timing is never right. You may even be 100% right, but you don’t have to belittle or put down the other side. In this case, they already lost, what more do you want?

Also curious if any experts can weigh in on this. I would guess in the 1 trillion to 2 trillion range.

Great use of tax dollars while the American people face all time cost of living highs among a plethora of many other problems. It’s sickening.

Interesting. I see papers where researchers will finetune models in the 7 to 12b range and even beat or be competitive with frontier models. I wish I knew how this was possible, or had more intuition on such things. If anyone has paper recommendations, I’d appreciate it.

Can someone explain how a 27B model (quantized no less) ever be comparable to a model like Sonnet 4.0 which is likely in the mid to high hundreds of billions of parameters?

Is it really just more training data? I doubt it’s architecture improvements, or at the very least, I imagine any architecture improvements are marginal.

It’s kind of funny seeing all the AI hype guys talking about their 10 OpenClaw instances all running doing work and when you ask what it is, you can never get a straight answer..

For the record though, I love agentic coding. It deals with the accumulated cruft of software for me.