Vrushank Desai (@vrushankdes) / X

Vrushank Desai

533 posts

Vrushank Desai

@vrushankdes

don’t take life too seriously, nobody gets out alive anyway

Joined March 2024

Pinned
Vrushank Desai
@vrushankdes
Oct 4, 2024
Meet Robo! i'm in the very early stages of building an accessible, general-purpose robotics platform, and looking for a few passionate people to help shape it from ground up. if you'd be excited about moving fast on a small, hands-on team I would love to chat -- my DMs are
00:00
15K
Vrushank Desai
@vrushankdes
Jul 31, 2024
>be me 🤡 >impulse pre-order AI hardware wearable (Oct '23) >Get an email later that there won't be an Android release, "reply for refund" >reply for refund .... >months pass, no refund >send three follow-up emails, nada >DM the founder on X, crickets >find out they spent $1.8M
01:43
Avi
@AviSchiffmann
Jul 30, 2024
introducing friend. not imaginary. order now at friend.com
1M
Vrushank Desai
@vrushankdes
Jan 24, 2025
i find it hilarious that Deepseek released pages of details about their parallelism/quantization/etc to stave off haters doubting their training efficiency and.....haters still can't believe how efficient their model is lmaoo
Neal Khosla
@nealkhosla
Jan 24, 2025
deepseek is a ccp state psyop + economic warfare to make american ai unprofitable they are faking the cost was low to justify setting price low and hoping everyone switches to it damage AI competitiveness in the us dont take the bait
Readers added context
There is zero evidence that Deepseek is a psyop. The post does not provide any sources and presents the opinion of the OP, whose father is a major OpenAi stockholder, as a fact. deepseek.com
126K
Vrushank Desai
@vrushankdes
Jul 31, 2024
Replying to @LucianKnock
60K
Vrushank Desai
@vrushankdes
Jul 31, 2024
Replying to @vrushankdes
.@AviSchiffmann
72K
Vrushank Desai
@vrushankdes
May 8, 2024
I spent a couple months at the beginning of this year learning about GPU programming through trying to optimize inference for @chichengcc awesome Diffusion Policy paper. I was able to improve inference time for the denoising U-Net by ~3.4x over Pytorch eager mode and ~2.65x over
00:00
109K
Vrushank Desai
@vrushankdes
Jul 31, 2024
Replying to @AviSchiffmann
thanks! not assuming malintent, but others also mentioned they're missing refunds (@chrisemmett_, @ignacioaal, @SoloOrTroll). given the pivot in pricing/product, you should refund $501 to all 2023 buyers & also offer full refunds with an automated form to make this right
17K
Vrushank Desai
@vrushankdes
Jan 26, 2025
r1 just zero-shotted a networking bug that $200 a month o1-pro is completely off the mark on....considering OpenAI was losing money on pro subs before r1 dropped we might actually have to #saveopenai
23K
Vrushank Desai
@vrushankdes
Dec 19, 2024
Google giving "USA post-Pearl Harbor sleeping giant" vibes
Jeff Dean
@JeffDean
Dec 19, 2024
Introducing Gemini 2.0 Flash Thinking, an experimental model that explicitly shows its thoughts. Built on 2.0 Flash’s speed and performance, this model is trained to use thoughts to strengthen its reasoning. And we see promising results when we increase inference time
6.8K
Vrushank Desai
@vrushankdes
Jan 23, 2025
Replying to @adonis_singh
i dont think the median ChatGPT use-case of solving a 5th grader's homework is bottlenecked by model intelligence at this point lol
16K
Vrushank Desai
@vrushankdes
Mar 26, 2025
Replying to @gfodor
how is bro dropping the ball before the race even starts?
3.7K
Vrushank Desai
@vrushankdes
May 8, 2024
Replying to @vrushankdes
GPUs have a memory hierarchy which scales from large capacity, low bandwidth, high latency (global memory) to low capacity, high bandwidth, low latency (thread registers). The part of the chip that does the actual computation (FP32/INT8 units, tensor cores, etc) can do math
GIF
15K
Vrushank Desai
@vrushankdes
Jun 4, 2024
Replying to @var_epsilon
global swe productivity just took a dump
6.3K
Vrushank Desai
@vrushankdes
May 8, 2024
Replying to @vrushankdes
The full blog post series is a lot more detailed and hopefully helpful for anyone interested in GPU kernel optimization! vrushankdes.ai/diffusion-infe… You can find all relevant code for the blog post here: github.com/vdesai2014/inf… If you’ve got an Nvidia GPU with CUDA & pytorch
5.9K