lelanthran

3 ms·

1.
by lelanthran 5d ago
> I get the feeling that mathematicians are needlessly panicking because they don't really understand how AI works. I sunno if mathematicians would be having problems understanding how matrix multiplication, backprop, sigmoid functi
2.
by lelanthran 5d ago
> AI does NOT “brute-force”, else we could have had those math breakthroughs decades ago. Brute-force as an approach, or partial approach, might only become feasible with more computing power. Decades ago, we had less computing power.
3.
by lelanthran 5d ago
>> The model was clearly trained or encouraged to work as it did, as evidenced by the fact it keeps using this particular escape hatch. > How does your logic explain that OpenAI seems to want to hide the extent of the HuggingFace i
4.
by lelanthran 5d ago
> So you're building a custom harness here that provides tools, and you're wiring up your custom harness to effectively do a subprocess execution of a CLI script for every tool call the model request? Well, yeah. Subprocess exe
5.
by lelanthran 5d ago
> If a full coding agent can access a CLI tool. that agent can almost certainly access the API keys being used by that tool. So, don't do that then? Why do you need to use a full coding agent as the interface between the model and
6.
by lelanthran 5d ago
> Exactly. There are so many replies here of the form “MCP sucks. If you just do <all the things mcp does a different way> you don’t need mcp at all.” See my previous reply to simonw elsethread. It's not "If you just do
7.
by lelanthran 5d ago
> So you've built your own agent harness that allows the model to call only your CLI but doesn't allow the model to run "env" and view the environment variables itself? Yes, with the difference being it can call oth
8.
by lelanthran 5d ago
> This article entirely misses the value that MCP brings today. Well, me too. I see the only advantage over a CLI app being an agreed-upon convention for syntax (not semantics). I have a CLI interface to my webapp, not an MCP. > If yo
9.
by lelanthran 5d ago
> Also, terminal access is extremely powerful on its own. Why should an agent even get access to that? I see it as a potential violation of least privileges. When even the official software from the token providers have never had human e
10.
by lelanthran 5d ago
> I think his “voice” has been filed LLM smooth by years of agent based interactions. I've wondered this for a long time now: how does one differentiate between your voice getting "filed smooth to match LLMs" or your brain
11.
by lelanthran 5d ago
> Just shut up dude. Show me your numbers or shut up. While Claude Code is a bunch of bugs held together by below-cost prices, and many have strong opinions on the seniority level of the lead dev(s), I think this is unnecessarily rude.
12.
by lelanthran 6d ago
Depends where the stress is. "They are some vendors, not all vendors" is natural. "They are vendors, not all vendors"is equally natural. Written without stress indicators, most dialog can sound unnatural.
13.
by lelanthran 7d ago
> I can't help but think that some of the arguments here apply equally well to writing _code_ as they to to other substantive writings. All of the arguments here apply to writing code; Why would the quality of writing differe when w
14.
by lelanthran 7d ago
> When I seek input (whether passively or actively; it doesn't matter) from another person, then I generally want that person's input. I want their words. There's a (probably subconscious) reason for you wanting that (and
15.
by lelanthran 7d ago
> this is where I perhaps differ from other commenters, I do not have higher tolerance, regardless of whether I asked it or am reading something someone else posted, fully or partially written by LLM. Perhaps English is not your first la
16.
by lelanthran 7d ago
I just explained why this is a bad idea: https://news.ycombinator.com/item?id=49767867 TLDR: Communication from a human is not only the material being communicated, it's also a signal that the human understands the
17.
by lelanthran 7d ago
> “You need to get the author to do an oral defense to verify that they wrote the paper.” While I agree with this point, it can’t scale. Are we prepared to sacrifice everything on the altar of scalability? At any rate, why can't thi
18.
by lelanthran 7d ago
> I feel like the only difference at this point is between people who do that and say that they do, and people who do that and don't say that they do. No. Did you "clean this up" with an LLM before posting it? I hate to
19.
by lelanthran 7d ago
> Not in a direct interview style, but most us conferences can request additional information or feedback. If they conditionally accept or reject a paper, that conditional relies on feedback from the author(s). Requesting feedback is us
20.
by lelanthran 7d ago
> Is it standard practice for authors to have to defend their submissions via interview like this? If not, why not? It isn't, but maybe it should be. For post-grad qualifications oral defense is standard, and I didn't mind defe
21.
by lelanthran 7d ago
> that the compiler optimized away some 'empty' loops a little bit too aggressively without realizing that these loops were setting output pins. Doesn't sounds like a bug in the compiler, sounds more like `volatile` was no
22.
by lelanthran 8d ago
> but none of the Office competitors had good enough compat to work. Don't worry about that - all the agents/harnesses/tools that produce .doc or .docx files do so with LibreOffice now. In a short time people would rather
23.
by lelanthran 9d ago
Like Anthropic and OpenAI are? After all, didn't they distill all the information in the world into their model(s)? I mean, if they get to distill other's IP, why can't others distill their IP?
24.
by lelanthran 9d ago
I have very little sympathy for thieves who get robbed of the goods they have stolen.
25.
by lelanthran 9d ago
> Why is that a problem? Just don't read her stuff if you're not interested. I agree, but... If you work hard to become a public figure, you must have known in advance that there'd be public opinions of you, right? I find
26.
by lelanthran 9d ago
> a risky memory fab in your home country is ten years ago. What makes it risky? Do you think that Apple will pivot into something that doesn't require RAM? Maybe like an IBM, they pivot to services?
27.
by lelanthran 9d ago
> Multiple places The people you are talking to have apparently never seen this research. Maybe you can help them out and provide links? It's faster to do that than to engage in some form of audience-persuasion, as well as more effe
28.
by lelanthran 9d ago
> Sorry friend, [...] Have a nice day. Not a problem; this is the playbook all the all-in AI boosters run to when they are called on their claims. Everyone reading this knows exactly what your claims meant, because the majority of HN h
29.
by lelanthran 9d ago
Who is a CEO going to believe? Amazon, Microsoft and the entire IT infra division or that lone SRE who disagrees with them? I mean, every time cloud comes up on HN we see legions of techies posting strowman arguments about why you should of
30.
by lelanthran 9d ago
It's not an insult. "People are deterministic" is news to me, so I'd really rather like to know which papers claimed that.

More ›