Research

Home  |  Research  |  Projects  |  Blog  |  Courses  |  Ask Me

My research focuses on the intersection of machine learning and systems, with particular emphasis on:

• Machine Learning Systems: Optimizing the training and inference of large-scale ML models
• Distributed Computing: Efficient parallelization strategies for deep learning workloads
• Compiler Optimization: Automatic optimization of ML computations for various hardware backends
• Memory Management: Novel techniques for training larger models with limited memory
• Hardware-Software Co-design: Designing systems that leverage modern accelerators effectively

My primary focus is LinuxGuard, a pipeline that learns from Linux kernel bug fixes to generate custom clang-tidy checkers. By mining commit history, the system builds AST matchers that flag unchecked error paths across kernels v3.0 through v6.0, turning each fixed vulnerability into a proactive safeguard.

• End-to-end automation: bug mining, checker synthesis, LLVM build, and multi-version scans.
• Flagship checker: linuxkernel-must-check-errs catches missing error handling at scale.
• 200+ unchecked error flows surfaced per kernel release, revealing long-lived anti-patterns.

Representative projects are highlighted. See also my Google Scholar profile.

Feel free to clone my template Xuming Huang