Design, develop, and deploy custom and off-the-shelf ML libraries and toolings to improve ML development, training, deployment, and on-vehicle model inference latency.
Build tooling and establish development best practices to manage and upgrade foundational libraries, i.e., Nvidia driver, PyTorch, TensorRT, etc., improve ML developer experience, and expedite debugging efforts.
Collaborate closely with cross-functional teams, including applied ML research, high-performance compute, advanced hardware engineering, and data science, to define requirements and align on architectural decisions.

Requirements

6+ years of experience
Proficient in Python or C++
Familiarity with any of the training frameworks and libraries like PyTorch, Lightning, Hugging Face, Ray, JAX, etc.
Familiarity with any of the GPU-accelerated inference on Nvidia hardware like CUDA, TensorRT, and/or XLA
Familiarity with Bazel and/or the C++ linker

Benefits

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills & Tools

PythonC++PyTorchLightningHugging FaceRayJAXCUDATensorRTXLA

Soft Skills

collaborationcommunicationproblem-solvingteamwork