Join Google's pioneering Silicon Engineering team and shape the future of massive-scale computing. You will contribute to next-generation technologies that redefine how billions connect and interact with information. This role offers the chance to work on critical projects, with opportunities to explore diverse challenges across the full stack as our business rapidly evolves.

We are seeking innovative engineers with expertise spanning information retrieval, distributed systems, large-scale design, networking, data storage, security, AI, NLP, UI, and mobile development. Your fresh ideas will drive advancements in these dynamic fields.

Develop cutting-edge ML compilers for the Tensor TPU, specifically engineered to accelerate Generative AI and other complex machine learning models on custom hardware. You will be instrumental in mapping ML models efficiently to hardware ISAs, with a focus on always-on computing use cases.

Collaborate with fellow compiler engineers to identify and resolve correctness and performance issues. You will also work closely with model owners, influencing their architecture design to ensure optimal performance on the EdgeTPU. This includes triaging and performing root-cause analysis for issues encountered during ML model enablement.

A Bachelor’s degree or equivalent practical experience is essential. You should possess at least two years of experience in ML compilers or general compiler development, coupled with solid experience in system software.

Preferred qualifications include a Master's degree or PhD in Computer Science or a related technical field. We value candidates with two years of experience in data structures and algorithms, a proven track record in optimizing ML model inference on devices, and expertise in power and performance optimizations. A strong understanding of hardware, particularly systems with high parallelism, is highly beneficial.