PinnedHao Zhuang, PhD - AI and Compute @ Tesla, GooglePlay OpenAI Triton MatMul IR LoweringNov 23, 20231Nov 23, 20231
PinnedHao Zhuang, PhD - AI and Compute @ Tesla, GoogleEnhancing Deep Learning Performance with Triton and PyTorch — vector add example with IRsGenerated by my old NVidia GTX 1060 GPUNov 22, 2023Nov 22, 2023
PinnedHao Zhuang, PhD - AI and Compute @ Tesla, GoogleExploring torch.compile with nanoGPT with Peffeto tracewhateverNov 20, 2023Nov 20, 2023
PinnedHao Zhuang, PhD - AI and Compute @ Tesla, Google2022 open source contribution — Sad…After leaving Google for Tesla, I spent way less time in the open-source community. I just made one PR in GitHub in the second half of…Jan 20, 2023Jan 20, 2023
Hao Zhuang, PhD - AI and Compute @ Tesla, GoogleReading: AI Hardware w/ Jim Kellerending: “this is the funniest thing you know you start training with complicated things you end training with complicated things and it’s…Jun 20Jun 20
Hao Zhuang, PhD - AI and Compute @ Tesla, GoogleRead: Google datacenter Optical Circuit Switching and TPU stuffsSome readings and collections were taken during my parental leave at home.Jun 7Jun 7
Hao Zhuang, PhD - AI and Compute @ Tesla, GoogleRead: Efficiently Scaling Transformer Inferencehttps://arxiv.org/pdf/2211.05102Jun 6Jun 6
Hao Zhuang, PhD - AI and Compute @ Tesla, GoogleRead: HC2023-K1: Exciting Directions for ML Models and the Implications for Computing Hardware @…“Keynote 1, Hot Chips 2023, Monday, August 28, 2023 Jeff Dean and Amin Vahdat, Google Jeff & Amin describe how Google is using ML models in…May 31May 31