Triton Conference 2024: Afternoon Session

1:25 Welcome: Ajit Mathews, META
6:05 Keynote: Triton Strategy at Meta, Aparna Ramani, META bit.ly/aparna-...
17:05 Keynote: Triton State of the Union and The Way Forward, Phillipe Tillet, OpenAI bit.ly/phil-ke...
Triton Backends
33:25 Enable Triton on Intel GPUs, Ettore Tiotto, Intel bit.ly/intel-b...
47:55 Triton on AMD GPUs, Lei Zhang and Lixun Zhang, AMD bit.ly/amd-bac...
1:01:30 Triton CPU, Minjang Kim, Meta and Ilya Enkovich, Intel bit.ly/cpu-bac...
1:18:20 Pipelining Persistent Kernels, Pawel Szczerbuk, OpenAI bit.ly/pipelin...
Triton Lightning Talks
2:02:20 Triton for Azure MAIA, Ian Baird, Microsoft bit.ly/maia-ba...
2:06:00 Rapid Innovation on AWS Trainium with Triton-Inspired Programming, Jonathan Henson, Amazon bit.ly/tranium...
2:12:00 Qualcomm Hexagon NPU Backend for Triton, Muthu Baskaran, Qualcomm bit.ly/qualcom...
2:15:35 Accelerating the Future: Triton on Blackwell Architecture, Pradeep Ramani, NVIDIA and Thomas Raoux, OpenAI bit.ly/blackwe...
Complements to Triton
2:28:00 Exo: Let's Build Power Tools for Performance Engineers, Gilbert Bernstein, University of Washington bit.ly/exo-com...
2:46:00 Mosaic GPU: A DSL for Fast Hopper Kernels in Python, Adam Paszke, Google bit.ly/mosaic-...
Panel:
3:03:37 Hardware Heterogeneity at Meta, Moderator: Ajit Mathews, Meta; Speakers: Jiaqi Zhai, Roman Levenstein, Jongsoo Park, Xiaodong Wang, representing ML Engineering, ML Software, ML Hardware, and AI Infra, respectively.