Staff Software Engineer, ML Systems,… at Google · Sunnyvale

Define the architecture and long-term technical roadmap for Google’s accelerator software stacks, focusing on distributed systems, Linux OS, networking, power management, and hardware interfaces. Collaborate with HW engineering to design, test, deploy, and debug low-level software across the stack—including firmware, kernel drivers, simulators, and cluster configurations for next-gen TPUs, GPUs, and custom accelerators. Provide technical leadership, mentorship, and coaching to a distributed team while fostering cross-functional alignment on goals, timelines, and program outcomes. Manage priorities and end-to-end deliverables to power Google's AI hyper-computers and data centers, supporting internal workloads and Google Cloud offerings. Drive large-scale technical programs to maximize production health, reliability, and scalability of distributed accelerator systems in partnership with multiple teams.