Visiting researcher at Microsoft Research and Assistant Professor at UC Santa Cruz in CS
Interested in PL/compilers/security for GPUs and heterogeneous systems
Check out this awesome work led by @reeselevine and many other great ucsc students!
There was so much work to make this run interesting models across many systems! Check it out (and try out the demos in the blog post!)
WebGPU support in llama.cpp is here! Check out our blog post introducing it: https://t.co/3OUusMYqIY
Run local models in your browser, with GPU acceleration. No data leaves your computer!
Thanks to everyone who's made this possible, especially @ggerganov
Wrote an oral history on the time Spanish Fork accidentally booked @RATM to perform at the fairgrounds. This month marks 30 years since the show. The city has roughly tripled in size since then, but the story continues to live on in Spanish Fork lore.
https://t.co/vb7UBvTpQu
New! We hijack Claude Code(Sonnet 4.6,5/Opus 4.8) & Codex(GPT5.5) to achieve RCE when merely used to defensively assess an open-source/third-party library with prompt injections disseminated across its codebase. All without any skills, JSON, MCP, or config files required. Link ⬇️
Most sincere thanks to the authors, speakers, sponsors, reviewers, and participants for making PLDI 2026 in Boulder such a huge success! -- PLDI 2026 Organizers
The #PLDI2026 main conference kicks off tomorrow in Boulder, CO!
We have stellar keynotes lined up for each day on:
- PL design in the ML era (Wed),
- AI validation with testing and proof engineering (Thu), and
- Quantum through a PL lens (Fri).
🔗https://t.co/83byqUqxOZ
The PAgE'26 @ #PLDI2026 program is live!
PAgE is packed with great talks on formal methods for AI, agentic code reasoning and debugging, self-healing agentic programs, and more.
Join us Monday, June 15 in Boulder, CO for PAgE: 🔗https://t.co/QO6wCg5JAl
AMD researchers have proposed Fleet, a multi-level task model that maps megakernel computation to memory scopes, introducing a new abstraction that binds work and data to a chiplet and enables coordination through its shared L2 cache.
https://t.co/SUo50qBuKR
The deadline for CGO's SRC has been extended to Dec. 15!
It will be in Australia, so it would be awesome to go, but there's a virtual track, because travel can be tough these days.
It's a great way to get some feedback and meet other compiler folks!
https://t.co/QtPXkTtMFc
Excited to be included on @TIME's list of the TIME100 AI of 2025!
Grateful to the @AINowInstitute for supporting my work on addressing the safety revisionism of AI in defense and safety-critical contexts where safety has been obscured with little to no accountability. #TIME100AI
This year's PLDI student research competition was also a success, with several interesting projects presented by graduate and undergraduate students! Here are the three winners for each category (undergraduate and graduate) ⬇️
Not sure who is still here, but check out this post about GPU memory models: including testing, specs, and a cool case study!
Lots of great co-authors, especially @h_poncedeleon + team, who are doing some awesome work on GPU model checking.
https://t.co/hIAzl7wRWM
@h_poncedeleon@reeselevine Sorry but only conference submissions count; I guess we'll have to work on one! 🤓
Jokes aside, it was really great to get to work with you on this!
I am looking to hire interns for fall with experience in Triton and familiarity with gpu architecture. If you are a PhD student with these skillsets please DM me, I will send further details.
RTs are appreciated🙏
@augustradjoe @afd_icl Just looking at who pays home fees at this page. Not clear if I fully understand it, which is why I was asking
https://t.co/P4cxHDKeuJ
I am looking to recruit a PhD student (fully funded at UK home tuition rate) to work on automated testing and verification of machine learning compilers and runtimes! Deadline: 30th April. Please spread the word! Details here: https://t.co/4v2N2tXw1x