Staff ML Performance Engineer (Compiler)
This role focuses on optimizing machine learning inference performance for edge accelerators and GPUs, particularly for running large transformer models efficiently on low-power in-vehicle systems. The engineer will work across compilers, runtimes, and kernels to deliver measurable improvements in latency, memory, and power, while collaborating closely with model developers and contributing to early-stage, high-impact projects. The position involves deep technical work on multiple hardware platforms and shaping performance engineering standards across the team.