Jobs

This full-time Inference role in Engineering & Research in the Bay Area focuses on building and optimizing high-performance, low-latency inference pipelines for both on-device robotics and distributed GPU clusters. The ideal candidate will have deep experience in distributed systems, ML infrastructure, and high-performance serving, with a strong background in Python and systems languages like C++/Rust/Go. Responsibilities include implementing efficient low-level code (CUDA, Triton) and optimizing workloads for throughput and latency. This role requires a system-level mindset to tune hardware-software interactions for maximum efficiency and responsiveness.