Embedded AI Engineer, On-Device Models

About the role

The role writes custom kernels, operators, and runtime components that let speech models run efficiently on non-NVIDIA accelerators, embedded SoCs, DSPs, and NPUs under tight power, memory, and latency budgets. It requires strong C, C++, or Rust skills, model quantization and optimization experience, and hands-on work with vendor toolchains and edge inference runtimes.