Olivia Hsu
PhD Candidate
McLean, Virginia, United States
Summary
🤩
Rockstar🎓
Top SchoolOlivia Hsu is a PhD candidate in Computer Science at Stanford and an incoming Assistant Professor at Carnegie Mellon University with a decade of engineering experience spanning compiler research, hardware, and space systems. She contributes to the tensor-compiler (taco) project, where she improved sparse tensor code generation by implementing workspace reuse, loop optimizations, and new IR nodes—work that bridges high-performance compilers and practical accelerator targets. Her background includes hardware engineering internships at Apple, a circuit design stint at Ayar Labs, and co-founding a multi-university student space program at Berkeley, reflecting a rare mix of systems-level research and hands-on hardware leadership. Olivia’s work uniquely blends compiler internals with real-world hardware constraints, positioning her to advance performance-focused ML infrastructure from algorithm to silicon.
10 years of coding experience
4 years of employment as a software developer
Bachelor of Engineering (B.Eng.), Electrical Engineering and Computer Science, Bachelor of Engineering (B.Eng.), Electrical Engineering and Computer Science at University of California, BerkeleyThomas Jefferson High School for Science and TechnologyDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Stanford University
English, Spanish, Chinese
Github Skills (12)
irr10
compiler-design10
c-language10
tensor10
cprogramming-language10
intermediate-code10
algebra10
linear-algebra9
code-generation9
sparse-data9
sparse-matrix9
code-optimization8
Programming languages (8)
Github contributions (5)
tensor-compiler/taco
Oct 2020 - Aug 2022
The Tensor Algebra Compiler (taco) computes sparse tensor expressions on CPUs and GPUs
Role in this project:
Back-end Developer
Back-end DeveloperContributions:27 reviews, 115 commits, 19 PRs in 1 year 9 months
Contributions summary:Olivia focused on improving the Tensor Algebra Compiler (taco) by implementing workspace reuse and optimizing for loop execution. Their contributions included refactoring code related to temporary workspace initialization, improving the handling of multi-dimensional workspaces and fixing related bugs, with an emphasis on handling precompute operations within the compiler. They also introduced a right shift operation and added a BinOp node to the compiler's IR, demonstrating a focus on enhancing the compiler's core functionality.
cudatacolinear-algebratensor-algebra-compilergpu
weiya711/weiya711.github.io
Apr 2021 - Dec 2024
Contributions:3 PRs, 33 pushes, 5 branches in 3 years 8 months
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.