![[Harrison Zhang]](assets/profile/myself.webp)
Harrison Zhang
Kernels @ NVIDIA
Email: harrisonzhang [at] alumni.harvard.edu
Harrison Zhang (张鸿宇) works on GPU kernels and optimizations at NVIDIA TensorRT-LLM.
Previously at Meta, he contributed to Prometheus 1GW
distributed training fabric bring-up,
and 2:4 sparse LLaMA-3.x pre-training with MSL Infra (PyTorch).
Before that, he was at AMD Research
working on accelerator inference optimizations.
He graduated with A.B. and S.M. in Computer Science from Harvard University.
[...]
Interests
Currently working on DiT inference kernel optimizations for TensorRT-LLM VisualGen and FlashInfer.
General interest in accelerator performance engineering, efficient inference and post-training paradigm exploration, hardware-software co-design.
![[Act-Sparse]](assets/act-sparse/teaser.webp)
![[Sparse-ANN]](assets/sparse-ann/teaser.webp)
![[H200-Matmul]](assets/h200-matmul/teaser.webp)
![[SC-DLM]](assets/sc-dlm/teaser.webp)
![[SplatGrasp]](assets/splatgrasp/teaser.webp)
![[ASPP-HWSW]](assets/aspp-hwsw/teaser.webp)
![[Sc-Perturb]](assets/sc-perturb/teaser.webp)
![[Chk-VM]](assets/chickadee/teaser.webp)
Education
S.M. in Computer Science
Sep 2025 - May 2026
A.B. in Computer Science
Aug 2022 - May 2026
Cambridge, MA
Experience
Software Engineer | Meta Superintelligence Labs (GPU Communications)
July 2026 - Aug 2026
New York, NY
Software Engineer Intern | FBOSS, Meta Superintelligence Labs (PyTorch)
May 2025 - Aug 2025
Menlo Park, CA
Researcher in Computation Structures Group | With Prof. Arvind and Dr. Xuhao Chen
Jan 2024 - Jan 2025
Cambridge, MA
Researcher | With Prof. Vijay Janapa Reddi
May 2023 - Dec 2023
Cambridge, MA
Selected Coursework
(^ indicates graduate level)
MIT 6.S894^ Accelerated Computing (GPU Performance Engineering; f25),
MIT 6.172 Software Performance Engineering (f23),
MIT 6.8700^ Advanced Computational Biology (f25),
MIT 6.8370^ Advanced Computational Photography (f24),
MIT 6.4212^ Robotic Manipulation (f24),
MIT 6.5930^ Hardware Architecture for Deep Learning (s24),
CS 242^ Computing at Scale (f25),
CS 222^ Algorithms at the Ends of the Wire (s25),
CS 262^ Distributed Systems (s25),
CS 161 Operating Systems (s24),
STAT 111 Statistical Inference (f25),
CS 175 Computer Graphics (s23),
CS 124 Data Structures and Algorithms (s23),
CS 181 Machine Learning (s24),
MATH 112 Real Analysis (f23)