almostgotcaught

I'm a researcher in FAANG - I'll keep in mind your condolences when I get my yearly RSU re-grant lololol.

> if you are an aspiring mathematician, and you do not have a sense of a good taste or mathematical beauty, you probably will probably not have a good time

Lol I have a PhD from a T10 and 15 published papers. I'm pretty sure I don't need your advice on "taste" or "beauty".

> What fraction of people who use differentials in their published work still cite Newton or Leibnitz was the point

Those papers were written in the 1600s. "The character of physical law", the essay you're ripping off, was written in 1964. 100% papers from the 1960s are cited every single time the techniques are used.

You are as tedious as the original refrain I was complaining about (which is not at all ironic). What's most tedious is you're not actually a mathematician but presume to speak for them.

.... Of course they do? You can still find citations of those papers to this day.

For all of your "forceful" comments on math, I think probably you don't actually know much about it.

I have no idea what you're trying to say - it is generally understood everywhere in the world (ie all forms of human culture) that it's pathetic to pass off someone else's insights as your own.

> Much like sex.

My guy you know lots of people in here have read Feynman right? You should cite him instead of pretending you were clever enough to come up with the analogy yourself.

> There are many mathematicians who find beauty

Lol did you think this was clever? You just literally reiterated exactly what I said. See, if you had said "there are many pianists that find beauty in math" - you know like how many mathematicians find beauty in piano concertos - then you'd have me.

the aesthetics of math and physics is by far the most boring discussion that can be had. i used to be utterly repulsed by such talk in undergrad - beauty this and that. it absolutely always felt affected and put on - as if you talk about it enough, you'll actually convince people outside of the major to give you the same plaudits as real artists.... yea right lol.

what compels me to be which way? to call out obnoxious things in the people i'm surrounded by? dunno probably my natural revulsion to noxious things?

Have you really never heard the expression "opinions are like ...." or did you not understand the point of the expression?

I would love for you to show me any novel theorems you've proven or publishable experiments you've run during your career as a swe.

Yes and you have to be invited to publish in a place. Meaning at least one other person has to believe your opinion is significant........

> Computer Science

do we really have to retread this? unless you are employed by a university to perform research (or another research organization), you are not a computer scientist or a mathematician or anything else of that sort. no more so than an accountant is an economist or a carpenter is an architect.

> The article you are complaining about starts from the presumption that software

reread my comment - at no point did i complain about the article. i'm complaining that SWEs have overinflated senses of self which compel them to write such articles.

what compels software people to write opinion pieces. like you don't see bakers, mechanics, dentists, accountants writing blog posts like this...

Edit: to everyone responding that there are trade mags - yes SWE has those too (they're called developer conferences). In both categories, someone has to invite you to speak. I'm asking what compels Joe Shmoe SWE to pontificate on things they haven't been asked by anyone to pontificate on.

> nonlinearly

Calling it nonlinear paints some horrible exponential picture. It's just squared (all to all communication). We deal with squared problems all the time (that's literally what distributed consensus is all about.....)

> no special vector instructions (e.g. SVE)

Wut - SVE and SME are literally Apple designs (AMX) which have been "back ported".

Anyone that has worked at FB knows that LeCun does nothing all day expect post archive links to WP.

hn has become literally just twitter level hottakes

> There are multiple TPU vendors

There are not - TPU is literally a Google trademark:

> Tensor Processing Unit (TPU) is an AI accelerator application-specific integrated circuit (ASIC) developed by Google.

https://en.wikipedia.org/wiki/Tensor_Processing_Unit

The rest of what you're talking about is irrelevant

I've said it before but no one takes it seriously: LLMs are only useful if you're building something that's already in the training set ie already commodity. In which case why are you building it???

Lol I love these clueless takes. I'm just curious who you thinks actually writes the stuff within LLVM? Chris lattner? Lololol

this is like 5 different questions all across the landscape - what exactly do you think answers will do for you?

> How often do hardware optimizations get created for lower level optimization of LLMs and Tensor physics?

LLMs? all the time? "tensor physics" (whatever that is) never

> How reconfigurable are TPUs?

very? as reconfigurable as any other programmable device?

> Are there any standardized feature flags for TPUs yet?

have no idea what a feature flag is in this context nor why they would be standardized (there's only one manufacturer/vendor/supplier of TPUs).

> Is TOPS/Whr a good efficiency metric for TPUs and for LLM model hosting operations?

i don't see why it wouldn't be? you're just asking is (stuff done)/(energy consumed) a good measure of efficiency to which the answer is yes?

> Her accomplishments are impressive and well deserved

what exactly are those even? that she went to MIT? from her linkedin she's at some blockchain startup (for only 4 months) doing "compiler work" - i put it in quotes because these jobs actually happen to be a dime-a-dozen and the only thing you have to do to get them is pass standard filters (LC, pedigree, etc.).

> and now CUDA is all but irrelevant.

Lol this is so wrong it's cringe.

> There's now so many different and opinionated takes on how you should write high performant accelerator cluster code. I love it.

There are literally only 2: SIMT (ie the same as it always was) and tiles (ie Triton). That's it. Helion is just Triton with more auto-tuning (Triton already has auto-tuning).

Bram always manages to build quite elegant stuff.

be sure to let us know when you can run eg nginx on a GPU in said enclave.

This question makes zero sense - PCC is a (proprietary) system not an interface. There is no spec just like there's no spec for how you have the furniture arranged in your own house.

I guess English isn't your first language so nbd but I have no idea what you're asking.