AI Infrastructure Engineer · Inference Optimization · GPU Kernels & Runtimes

Nanjing University alumna working on high-performance AI inference, from low-level GPU kernels to runtime and serving systems.

• ApxInf — Core contributor across the runtime and kernel stack, building a high-performance operator library through runtime–kernel co-design and megakernel-style execution. Established an agent-driven development workflow with standardized benchmarks and implementation guides, enabling rapid operator optimization and integration for new models.