Summary
Cuiqing Li is a Senior AI Algorithm Engineer based in Shanghai with a decade of experience building high-performance ML systems and deploying LLMs in production, currently applying AI to quantitative trading workflows. With prior roles at Meta and ByteDance, she specializes in deep learning compilers, model optimization, and low-latency inference—contributing to projects like PyTorch, TVM and Colossal-AI and to runtimes such as PyTorch Executorch. She has hands-on expertise developing operator kernels, KV-cache managers, and tuning pipelines that bridge research and production at scale. Her background in applied and pure mathematics (Johns Hopkins) underpins a rigorous approach to profiling, quantization and compiler-level optimizations. Notably, she moved from improving edge-device runtimes to integrating LLMs across entire research cycles, reflecting both deep systems skill and product-focused impact.
10 years of coding experience
4 years of employment as a software developer
Bachelor of Arts - BA Mathematics and Computer Science(Minor), Bachelor of Arts - BA Mathematics and Computer Science(Minor) at University of Minnesota
Johns Hopkins University
English, Chinese