Applied ML Engineer - Edge Devices

About the role

Adapt speech models to run on non-NVIDIA accelerators, edge servers, and embedded platforms by adjusting operators, architecture parameters, and quantization while preserving accuracy and latency. The role validates each port on real hardware, builds repeatable conversion and deployment pipelines, and works with embedded kernel engineers. Candidates need prior production experience deploying models to edge hardware and familiarity with quantization and vendor inference runtimes.