Trent Lo
Senior Software Engineer at NVIDIA
San Jose, California, United States
Summary
👤
Senior🎓
Top SchoolTrent Lo is a Senior Software Engineer in San Jose with 13 years of experience building high-performance deep learning compilers and heterogeneous compute stacks for GPUs, CPUs, and ML accelerators. At NVIDIA he focuses on deep learning compiler performance and programmability, and previously led teams at MediaTek that delivered industry-leading OpenCL runtimes and the world's first production-quality device fusion for dynamic CPU/GPU scheduling. His open-source contributions to XLA and DeepRec demonstrate a knack for preserving parallelism and reducing kernel launch overhead through horizontal and slice input fusion. With a Ph.D. and a background in system-level simulation, he combines academic rigor with practical engineering to squeeze orders-of-magnitude speedups from complex systems. An under-the-radar strength is his repeated success refactoring large compiler codebases to both fix build issues and unlock measurable runtime gains.
13 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at National Tsing Hua University
Mandarin, English
Github Skills (18)
c-language10
build-system10
tensorflow10
compiler-optimization10
xla10
compiler10
cprogramming-language10
linear-algebra10
algorithm9
code-optimization9
algorithms9
python9
machine-learning9
optimisation9
compiler-design9
Programming languages (4)
Github contributions (5)
DeepRec-AI/DeepRec
Aug 2019 - Oct 2020
DeepRec is a high-performance recommendation deep learning framework based on TensorFlow. It is hosted in incubation in LF AI & Data Foundation.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:56 commits in 1 year 2 months
Contributions summary:Trent's commits primarily involve refactoring and modifying code related to the XLA (Accelerated Linear Algebra) compiler within the TensorFlow project. Their work focuses on enhancing the cluster scoping pass by renaming attributes and refining comments. This includes the introduction of a new attribute, `_XlaAutoJitScope`, to guide clustering results and a significant rework to prevent the loss of critical parallelism in the graph.
aideep-learningtensorflowmachine-learningpython
openxla/xla
Aug 2019 - Jul 2021
A machine learning compiler for GPUs, CPUs, and ML accelerators
Role in this project:Back-end Developer
Back-end DeveloperContributions:85 commits in 1 year 11 months
Contributions summary:Trent primarily contributed to fixing build breaks and improving the XLA compiler. Their work involved adding missing dependencies, refactoring code to use standard library functions, and implementing horizontal fusion and slice input fusion, which improved kernel launch overhead and potentially increased launch dimensions. The user's contributions focused on improving the compiler's build process, and enhancing its performance by modifying the core code base.
compilermachine-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.