BENJAMIN NETANYAHU IF YOU CAN HEAR ME REMOVE THE INFINITE RECURSION IN MY NIXOS CONFIGURATION

The glory work of GPU scheduling is in the frontier data centers with hundreds of thousands of GPUs, but a lot of research work is done with single GPU jobs on modest clusters, and the scheduling leaves much to be desired. I wish there were a clean way to preempt GPU tasks, so

foid foid go away spike my cortisol another day

british ai researcher be like anfropic

cortisol spike is at an all time high

no uni break during ramadan

ok bro

I wrote a lil personal reflection blog called "Why I Always End Up Going Back to C" please note this is not a "why you should use C" use whatever you want Link in 1st reply P.s. Jai will invalidate most of these points because its just that good

Gonna start calling tattoos “goystamps”

The PyTorch Caching Allocator technically has zero justification for its existence. Cuda memory pools when properly configured can already serve hot-loop async memallocs essentially for free, and the cuda driver has the option to virtually defragment the memory pool before

It do be like that sometimes

Watching the labor market collapse while already being unemployed

the downfall of opm needs to be studied

THE REVOLUTION WILL NOT BE TOKENIZED

https://t.co/lB8OT2VdXI

I am very happy to report that I am joining @tenstorrent. It’s only been a year since I quit my job as a web developer. I promised to pivot to working on difficult and meaningful things, and I’m truly honored to have been given such an opportunity so early. There’s an

the problem with sf niggas is that they dont care whether they make a good useful product or not, they just wanna say we raised from mr x at y valuation at a "cool" sf party so they dont feel like they are left out

waking up at 4am is so chill bruh ashton hall was onto sum

Ill let yall know when my pager decides to explode