NVIDIA CUTLASS Core Maintainer Steps Down, Driving GPU Inference Innovation
bingxu_ · x · 2026-08-11
Haicheng (@hwu36), a core developer of NVIDIA's CUTLASS project, announced he is stepping down as admin and will no longer work on it full-time, handing over responsibilities to other team members.
Former Meta engineer @bingxu reflected on CUTLASS's pivotal role in GPU inference optimization. After returning to Meta in 2021, he collaborated with Haicheng to adapt CUTLASS for inference needs, driving crucial innovations like Epilogue Fusion, Scatter-Gather Fusion, and Grouped GEMM. What started as a two-person project now powers the inference stack across the entire industry.
More from Companies & People
- Opinion: Every Company Needs a 'Cassandra' Background Agent — threepointone · 2026-08-11
- Legal Debate on AI Distillation Heats Up Amid Accusations Against Chinese Labs — herbiebradley · 2026-08-11
- OpenAI Believed to Cut Datadog Usage, Impacting Cloud Provider's Guidance — SumitGup · 2026-08-11
- Silicon Valley's Vicious Cycle: Hyperproductivity, 996, and Mega Rounds — jonchu · 2026-08-11
- Amazon Gutting Nova AI Models, Pivoting Hard to Robotics — max_paperclips · 2026-08-11
- Okta Hiring Principal AI/ML Scientist to Secure Production AI Agents — yenkel · 2026-08-11