Shape the future of AI/ML hardware acceleration as a Silicon Architect/Design Engineer. Drive TPU (Tensor Processing Unit) technology powering Google's most demanding AI/ML applications. Collaborate with hardware and software architects to define and design next-generation TPUs. This dynamic role involves product definition, design, and implementation, optimizing for performance, power, features, schedule, and cost.

Join a team that builds the infrastructure behind Google's online services. We develop and maintain data centers and create next-generation platforms, making Google's product portfolio possible. We are engineers for engineers, dedicated to innovation and maintaining robust systems for optimal user experience.

Revolutionize ML workload characterization and benchmarking, proposing optimizations for future TPUs. Develop architecture specifications aligned with AI/ML roadmaps. Create power/performance models and RTL designs, evaluating quantitative and qualitative performance. Partner with hardware, software, compiler, and ML teams for effective hardware/software co-design.

Drive advanced AI/ML capabilities and efficient design verification. Utilize AI techniques for optimal physical design convergence, including timing and floor planning. Investigate and optimize DFT, post-silicon test, and debug strategies to advance silicon bring-up and qualification.

Requires a PhD in Electronics and Communication Engineering, Electrical Engineering, Computer Engineering, or a related technical field, or equivalent practical experience. Proficiency in programming languages such as C++, Python, and Verilog is essential, along with experience using Synopsys and Cadence tools. Familiarity with accelerator architectures and data center workloads is also necessary.

Preferred qualifications include 2 years of silicon domain experience post-PhD. Experience with performance modeling tools is advantageous. Knowledge of arithmetic units, bus architectures, accelerators, memory hierarchies, and high-performance/low-power design techniques is beneficial.