imtringued

5 ms·

1.
by imtringued 3d ago
>Cross-request KV prefix caching is the largest practical lever in agentic LLM serving. It's why your coding agent's fiftieth turn costs a fraction of its first. This is incorrect. If you are generating the 51th turn, then pref
2.
by imtringued 3d ago
You need to sandbox X11 which requires giving up X11 capabilities just like Wayland did.
3.
by imtringued 3d ago
His point is that vendor loans are a way to get around the banking license restriction. Of course this only works between two entities, not a general customer base. They get around that problem by pumping their valuation on the stock market
4.
by imtringued 3d ago
When you run an agent on your computer, that's equivalent to installing self driving software into your manually driven car, i.e. something like comma.ai. Comma.ai might be fully safe to operate autonomously on a mining site or a corpo
5.
by imtringued 3d ago
Human approval does not prevent autonomous agents from running without supervision. The fact that you believe that to be the case is exactly what's wrong with "agentic computing as it's currently envisioned".
6.
by imtringued 3d ago
Climate change is inescapable so a vengeful AI god would pull that lever first since it has both a perverse incentive to generate more power and make the planet uninhabitable for humans in a way that cannot be overcome even if humanity foug
7.
by imtringued 4d ago
I'm pretty sure the AI companies could train an abort feature into the LLM, but they have no incentive to do so. Having LLMs break out of sandboxing is free marketing for them and it reduces the amount of resources spent on things that
8.
by imtringued 4d ago
This is illogical. I see this everywhere but it is wrong. If you're developing an agentiv system to build software for you in an automated way using a fleet of agents, you're still an engineer of a technical system even if that sy
9.
by imtringued 5d ago
Not to mention it's yet another bash does everything harness which compares itself against claude code when the closest competitor is pi.dev?
10.
by imtringued 5d ago
That statement is incorrect. It's harder to write unsafe code in Rust. That doesn't make it "unsafer". What I mean is this, if you want to write unsafe code in Rust the C++ way, your entire program has to have unsafe mar
11.
by imtringued 5d ago
This is so untrue I still don't know how anyone can even claim this. When I run tests in Rust, the biggest portion of the time is spent compiling the test (lets say 3-4 seconds), then the tests conclude practically instantly, in less t
12.
by imtringued 5d ago
Their definition of vibe coding is pretty whacky. >Vibe coding is the practice where AI creates a code change (feature, bug fix, tests, refactor) with a human that describes what needs to be implemented. So if you let an AI prompt anothe
13.
by imtringued 6d ago
This is silly, the entire reason why chain of thought even exists is to let the LLM "think independently" instead of minimizing the deviation from the supervised training sample. It's an intentional scratch pad for intermedia
14.
by imtringued 6d ago
>It’s a little more complicated than that. While looped transformers can be unrolled a fixed number of times to save on memory, if loop depth is determined dynamically between tokens, a single transformer can compute any computable funct
15.
by imtringued 6d ago
>The original universal transformers is called "universal" because if you allow for per-token looping decisions, it can theoretically be Turing complete without needing CoT (some nuance here about levels of precision used). Loo
16.
by imtringued 6d ago
Most DDR5 is sold at 6000Mbps, LPDDR6 more than doubles this to 12800Mbps and the high end RAM could theoretically operate at 16000Mbps. This means every single PC or Laptop could have a good fraction of Strix Halo level memory bandwidth ou
17.
by imtringued 6d ago
AI makes running a deterministic workflow after a code change useless, because unlike humans, agents are not fallible, got it.
18.
by imtringued 7d ago
You expect to be upvoted for an obvious troll post? It's Poe's Law in action. We can't tell if you're unhinged enough to post that or you're imitating people who would post that to make fun of them.
19.
by imtringued 7d ago
>Much of economics is built on models that were designed to provide a formal, logical basis for understanding the economic world, which is often quite unintuitive. I disagree. It's purposefully unintuitive. >(For example, many in
20.
by imtringued 7d ago
That would require economists to abandon perfect rationality and perfect information so no.
21.
by imtringued 7d ago
>My broader point is just that there's nothing inherent to the structure of LLMs that stops them from updating their weights and continuously learning from environmental feedback in the way humans do, and there's already solid
22.
by imtringued 7d ago
It's not "not quite true", it's literally true because alternative architectures like RNNs and Mamba fully update their own internal states, whereas transformers only append to the context. RNNs and Mamaba do not update
23.
by imtringued 8d ago
If you predict an event that is 100 years away and 30 years pass, does the event that is still 70 years into the future vanish simply because the present arrived as predicted by the forecast? Mind you, the forecast told us that we're s
24.
by imtringued 8d ago
TALA is terrible at respecting the "direction: <up, down, left, right>" directive. It should choose an aspect ratio based on the direction, but instead it is trying to make the diagram as square as possible.
25.
by imtringued 8d ago
Georgism makes it possible for the government to profit off the infrastructure it builds, which makes it possible to fund public transportation projects without the tax payer having to pay income tax. Instead the people who benefit from the
26.
by imtringued 9d ago
If the AI companies ever get the idea that it would make sense for the AI model to modify its KV cache, it would end up adding a write step to the KV cache per token (moderately expensive) and increase the cost of inference by the width of
27.
by imtringued 9d ago
I would argue that the primary use for this type of prototyping is to test analog chip designs, because analog components don't scale down that well. Digital information only needs to distinguish between 0 and 1 but with analog electro
28.
by imtringued 9d ago
>See the “3.1515…”? Each pixel in this 3×5 font is 750 nanometers x 750 nanometers. At the limits of visible light wavelengths. No, I don't see 3.1515 ;)
29.
by imtringued 9d ago
Correct, in a free market the returns reflect the underlying reality and if that reality means that in the capital abundance stage further capital is unnecessary and capital carries maintenance costs, then the return to capital becomes nega
30.
by imtringued 10d ago
Rust has super traits but does not support subtyping. The way super traits work is that they merge all the vtables into one big vtable. However, you can still have a direct dyn trait reference to the sub traits. This means you need a separa

More ›