Join a forward-thinking team dedicated to crafting custom silicon solutions that drive Google's consumer products. You will be instrumental in developing groundbreaking hardware, impacting products used by millions globally. Your technical acumen will guide the creation of next-generation hardware, delivering superior performance, efficiency, and seamless integration.

The Platforms and Devices organization focuses on Google's diverse computing software platforms across various environments. It also encompasses our proprietary devices and services, merging Google's AI, software, and hardware expertise. Teams here are engaged in researching, designing, and developing novel technologies to enhance user interaction with computing, making it faster and more fluid.

Take ownership of all phases of ASIC RTL to GDS implementation for the Graphics Processing Unit (GPU) within the Tensor SOC.

Oversee block-level and full-chip physical implementation, focusing on achieving optimal Quality of Results (QoR) concerning power, timing, and area.

Drive signoff convergence, including Static Timing Analysis (STA), Physical Design Verification (PDV), and Electro-Migration (EM) assessments.

Optimize power, performance, and area to meet demanding performance targets and cost efficiencies.

Accelerate time-to-market and enhance PPA by implementing Artificial Intelligence (AI) in physical design tools and workflows.

A Bachelor's degree in Electrical Engineering, Computer Engineering, Computer Science, or a closely related field is required. Equivalent practical experience will also be considered.

Candidates must possess at least 4 years of specialized experience in physical design, with a proven track record of completing two or more chip tape outs.

Preferred qualifications include a Master's degree or PhD in Electrical Engineering, Computer Engineering, or Computer Science, with a strong concentration in computer architecture. Experience in optimizing for power, performance, and area, particularly within high-performance design contexts (e.g., CPUs, GPUs, DSPs), is highly advantageous.