HackerTrans
TopNewTrendsCommentsPastAskShowJobs

yu3zhou4

no profile record

Submissions

Exapunks (2018)

zachtronics.com
333 points·by yu3zhou4·9 gün önce·123 comments

William Thurston: On Proof and Progress in Mathematics (1994) [pdf]

ams.org
1 points·by yu3zhou4·9 gün önce·0 comments

AMD contributes their GPU support to tiny-vLLM

github.com
2 points·by yu3zhou4·12 gün önce·0 comments

The Cypherpunk Library

cypherpunkbooks.com
378 points·by yu3zhou4·geçen ay·97 comments

Show HN: Tiny-vLLM – high performance LLM inference engine in C++ and CUDA

github.com
205 points·by yu3zhou4·geçen ay·18 comments

TorchWebGPU: Running PyTorch Natively on WebGPU

github.com
2 points·by yu3zhou4·3 ay önce·0 comments

Ask HN: Did Claude lowered its usage limits?

7 points·by yu3zhou4·3 ay önce·2 comments

Benchmark LLM Inference on WebGPU

arxiv.org
1 points·by yu3zhou4·3 ay önce·0 comments

Show HN: WebGPU LLM inference comprehensive benchmark

arxiv.org
2 points·by yu3zhou4·3 ay önce·2 comments

Show HN: You can run real PyTorch in a browser

browser-torch.maczan.pl
3 points·by yu3zhou4·6 ay önce·0 comments

Show HN: I got PyTorch models running on WebGPU without ONNX export

github.com
1 points·by yu3zhou4·6 ay önce·1 comments

Show HN: Run and Compile LLMs in PyTorch on WebGPU

github.com
1 points·by yu3zhou4·6 ay önce·0 comments

Show HN: WebGPU back end for PyTorch sneak peek

github.com
4 points·by yu3zhou4·7 ay önce·0 comments

Show HN: Hirschberg Algorithm in PyTorch

jedrzej.maczan.pl
2 points·by yu3zhou4·7 ay önce·0 comments

20x less peak RAM in the new PyTorch memory budget solver

jedrzej.maczan.pl
8 points·by yu3zhou4·8 ay önce·0 comments

Ask HN: What's new in operating systems these days?

5 points·by yu3zhou4·10 ay önce·0 comments

Poland shoots down drones over its territory amid Russian attack on Ukraine

theguardian.com
13 points·by yu3zhou4·10 ay önce·0 comments

comments

yu3zhou4
·8 gün önce·discuss
Print ondemand can be ordered for less than $5 per zine in Lulu, since 2022 or so. I guess Zach doesn’t make a cut from it since this $5 probably barely covers the cost of print and Lulu fee :(
yu3zhou4
·23 gün önce·discuss
PyTorch?
yu3zhou4
·geçen ay·discuss
README is in my opinion (author here) the most interesting - I wrote it to help others build useful mental model to be able to recreate the project yourself, without need to even read my code
yu3zhou4
·2 ay önce·discuss
There was onivim that was a bit hyped a few years ago but unfortunately it died
yu3zhou4
·2 ay önce·discuss
Just learning math and trying myself in ml research
yu3zhou4
·2 ay önce·discuss
Laughed hard about Collegium Tumanum
yu3zhou4
·2 ay önce·discuss
Poland was sort of occupied until 1989
yu3zhou4
·2 ay önce·discuss
Good idea, too! Why do you see my explanation as cynical, though?
yu3zhou4
·2 ay önce·discuss
Maybe it’s in order to have an external provider to blame for failures and shift the blame/responsibility?
yu3zhou4
·3 ay önce·discuss
Adding a support for new hardware to PyTorch is actually quite convenient. I did that with WebGPU using the same PrivateUse1 mechanism TorchTPU used. Every hardware has its own slot and identifier, and when you want to add a support for a new one without merging it into PyTorch, PrivateUse1 works essentially like plug-in slot

https://github.com/jmaczan/torch-webgpu
yu3zhou4
·3 ay önce·discuss
They write that they use PrivateUse1, so it’s a custom out-of-tree backend
yu3zhou4
·3 ay önce·discuss
The system like iOS is still closed. Once Apple ends support for a device, being able to swap a battery won’t help much
yu3zhou4
·3 ay önce·discuss
Thx, great read about Yemen from Maciej Cegłowski https://idlewords.com/2014/07/sana_a.htm
yu3zhou4
·3 ay önce·discuss
An open course on building high performance LLM inference engine! Hope to finish by the end of April

https://github.com/jmaczan/tiny-vllm
yu3zhou4
·3 ay önce·discuss
I love this article, the edge cases are where the seeming “simplicity” of floating-point numbers breaks down

Recently wrote a chapter in tiny-vllm course about floats in context of LLM inference, much shorter and not that deep as this one, for anyone interested in topic you might like it too https://github.com/jmaczan/tiny-vllm?tab=readme-ov-file#how-...
yu3zhou4
·3 ay önce·discuss
Honestly there is a lot for room of improvement in torch-webgpu for performance. Needs involvement of community but the opportunities are definitely there
yu3zhou4
·3 ay önce·discuss
Overall I think it’s a good perspective and worth reading to learn your perspective and experience. Just for anyone who doesn’t come from a good place, please be kind to yourself. You probably really need to work exceptionally hard and need to have top 1% resume. And articles like this can be quite depressing. Don’t let it you down

The author tries to be mindful about it and I appreciate it, but from a perspective of someone from a shithole it still might feel bad

Referring to:

> I've never gotten a job by applying to it. It's always been referrals or someone reaching out to me. So honestly, my resume is shit compared to my peers. I'm terrible at interviewing, and I've never done LeetCode. This is not a brag; it's just not my style. Am I a nepo baby? I don't know. Was I a morale hire? I'm pretty disagreeable. Was it merit? Also not sure

> I recognize that not everyone has access to the same networking opportunities, and the traditional job application process can be a valid and necessary path for many. But social media, when used correctly, is a great way to get an opportunity
yu3zhou4
·3 ay önce·discuss
Good for the maintainer, hope they find peace and do things just for their fun, without needing to deal with comments like that anymore
yu3zhou4
·4 ay önce·discuss
So in some sense we start recreating an operating system, or at least the userspace, within the Claude code. There was some name for this pattern but I can’t recall
yu3zhou4
·4 ay önce·discuss
Can confirm that it also happens in other complex systems! Still a lot of good time and the novelty factor helps with pushing through