I'm Zhuo Cui, a researcher in computer science with a primary interest in compiler optimization.
My work focuses on how compilers reason about programs and turn them into efficient code. I'm especially drawn to the interplay between program analysis, intermediate representations, and the performance characteristics of modern hardware — the design space where a carefully chosen transformation can change the shape of a program's runtime behavior.
Current directions:
Optimization passes & IR design: exploring transformation passes on modern compiler IRs, and how pass ordering and cost models affect final code quality.
Program analysis: static and dynamic analyses that feed optimization decisions — aliasing, dependence, and profile-guided techniques.
Code generation & performance: bridging the gap between high-level language abstractions and low-level hardware features, with an eye toward vectorization, memory-hierarchy behavior, and specialized accelerators.
Alongside compilers, I've also worked on problems in signal processing, machine learning, and quantum information — each has shaped how I think about "optimization" at different layers of the stack.