Benjamin Chetioui - only @[email protected]
@_SIben_
Followers
1K
Following
5K
Statuses
2K
Moving to Bluesky @[email protected] Working on the XLA GPU compiler @Google 🇨🇭 🇫🇷 PhD in PL from UiB 🇧🇻 Go player Previous @FlatNetworkOrg
Joined September 2010
"Revisiting Language Support for Generic Programming" is the paper I enjoyed working on the most during my PhD :) Much of the literature on supporting GP focuses on retrofitting GP into existing languages. We ask: what happens if one designs a programming language *for* GP? 1/n
Announcing the second issue of Volume 7 of the Programming Journal: #research #papers #programming #prog23 #openaccess
1
2
11
... And now the thesis is online (! And the defense is scheduled next Wednesday ( 😱 It'll in principle be streamed—feel free to DM if you're interested in attending :)
1
1
10
... And now the thesis is online (! And the defense is scheduled next Wednesday ( 😱 It'll in principle be streamed—feel free to DM if you're interested in attending :)
And almost yet another year, but as of today, the dissertation is submitted! 🥳
0
2
7
@_xjdr @IlyasHairline @finbarrtimbers FWIW, we have the ability to use Triton to lower a lot of our code in XLA:GPU (and in fact, do it)! I'm curious exactly what are the performance cliffs that you're facing, and whether we can help with that!
0
0
3
Genuinely excited about this one, can't wait to find out where Mosaic GPU can take H100 performance!
Many of you are excited about H100 attention, so it’s a good time to show you Mosaic GPU: a Python DSL for H100s. The attention example matches FA3 performance, while being only ~200 lines of Python: It's easy to install too! Latest JAX packages have it.
0
0
1
RT @sharadvikram: Finally got around to writing a guide for matrix multiplication on TPUs using Pallas. Check it out!
0
25
0
RT @G33KatWork: @gamozolabs @gf_256 hi, program expert here! This is not funny. When a program is under distress and feels threatened, it s…
0
1
0
@PatrickKidger Likewise just moved to Zürich! I followed some of your work while at Google, would love to meet up :)
1
0
2
RT @dcuthbert: As much as I'm personally enjoying the AI ride at the moment, the attacks we are seeing are worryingly familiar web app atta…
0
7
0
RT @shncldwll: Tired of “LLM hacking” hype with no code? Here’s a breath of fresh air. 1. Challenges: open source…
0
63
0
@soumithchintala @AlexLaterre @sarahookr Note that JAX/XLA has also gotten *a lot better* on NVIDIA GPUs over the past year (and we're still going) :)
1
0
3
0
0
1
Lots of hard work from the team this year paying off 😇
JAX is great for GPUs too! With a strong team, and excellent execution, you can train a world-class LLM in ~two months!
0
0
6
First dissertation draft ready! 🥳 Only took me 5 years and 363 days out of a 4-year contract🤡
1
0
11
RT @arjunrajlab: Writing is an absolutely critical skill… and one that can be improved dramatically. Doing so requires not just repetitions…
0
196
0
RT @kamatsu8: Not announcing anything publicly yet, but please get in touch with me if you're interested in funded PhD opportunities in app…
0
9
0