Qiuyang Mang

Hi, I’m Qiuyang Mang, a first-year CS PhD student in the Sky Computing Lab at UC Berkeley, advised by Prof. Alvin Cheung. I lead FrontierCS, a benchmark and framework for LLM-driven algorithm evolution on open-ended coding tasks. My research interests center on two themes:

• LLM Long-Horizon Optimization: Leveraging LLMs for complex, multi-step optimization, powering long-horizon code-agent from data curation to post-training. FrontierCS, Argus

• Machine Learning Systems: Efficient algorithms for ML workloads, from data-processing to inference scheduling. PLOP, SVG-EAR, Continuum

Prior to joining Berkeley, I received my B.E. from The Chinese University of Hong Kong, Shenzhen, where I was advised by Prof. Pinjia He. I also spent an unforgettable year as a research assistant at the National University of Singapore with Prof. Manuel Rigger. During my undergraduate years, I worked on robustifying NLP systems and Database Management Systems.