Description
The role focuses on shaping Google’s TPU hardware architecture and accelerator performance, including architectural innovation, power and thermal optimization, compiler and runtime integration, workload characterization, benchmarking, and hardware/software co-design. It requires a bachelor’s degree in a relevant engineering or computer science field and at least five years of experience in computer or chip architecture, hardware design, or related areas, along with C++ or Python software systems experience. Preferred qualifications include a master’s or PhD, processor or accelerator design experience, deep learning framework knowledge, and familiarity with hardware/software stacks for deep learning accelerators. The listed compensation is $163,000–$236,000 USD plus a 15% bonus target, equity, and benefits.
