TPU Architect, Silicon

Google organizes the world's information and makes it universally accessible and useful, combining AI, Software, and Hardware to create helpful experiences.
New Taipei, Banqiao District, New Taipei City, Taiwan
Hardware
Mid-Level Software Engineer
In-Person
5,000+ Employees
3+ years of experience
AI · Hardware

Description For TPU Architect, Silicon

Google is seeking a TPU Architect to join their Silicon team, focusing on developing and optimizing Tensor Processing Units (TPUs) - custom silicon solutions that power Google's AI and machine learning capabilities. This role combines hardware architecture expertise with machine learning knowledge to drive innovation in AI accelerator design.

The position requires deep understanding of computer architecture, particularly in areas like microarchitecture, cache systems, and memory subsystems. You'll be responsible for analyzing and optimizing the performance and power efficiency of TPU architectures, working with machine learning workloads, and collaborating across teams to improve overall system performance.

This is an exciting opportunity to work at the intersection of hardware architecture and machine learning, directly impacting Google's AI infrastructure. You'll be part of a team that pushes the boundaries of what's possible in AI acceleration, working on technology that powers many of Google's core products and services.

The role offers the chance to work with cutting-edge technology in AI hardware acceleration, collaborate with world-class engineers and researchers, and contribute to the development of next-generation AI hardware. You'll be involved in the complete development cycle, from architecture design to performance analysis and optimization.

The ideal candidate should have strong technical skills in computer architecture, experience with machine learning concepts, and the ability to work effectively in a collaborative environment. This position offers the opportunity to make significant contributions to Google's AI hardware infrastructure while working on challenging and impactful projects.

Last updated 14 hours ago

Responsibilities For TPU Architect, Silicon

  • Build up tools, flows, and dashboards for Tensor Processing Unit (TPU) power/performance analysis
  • Analyze important Machine Learning workloads, evaluate power and performance and propose architecture or compiler improvements
  • Analyze micro-architecture of the TPU, engage with the implementation team and propose power or performance optimization opportunities
  • Collaborate with cross-functional teams to improve the end to end workload analysis flows, including debugging and tracing

Requirements For TPU Architect, Silicon

  • Bachelor's degree in Electrical Engineering, Computer Engineering, Computer Science, a related field, or equivalent practical experience
  • 3 years of experience with computer architecture concepts, with microarchitecture, cache hierarchy, pipelining, and memory subsystems

Interested in this job?

Jobs Related To Google TPU Architect, Silicon

Formal Verification Engineer, Silicon, Google Cloud

Formal Verification Engineer position at Google Cloud, focusing on ASIC design verification and implementation of formal verification methodologies for complex hardware systems.

TPU, Pre-Silicon Validation Engineer, Google Cloud

TPU Pre-Silicon Validation Engineer position at Google Cloud, focusing on hardware acceleration for AI/ML applications and TPU development.

Formal Verification Engineer, Silicon, Google Cloud

Formal Verification Engineer position at Google Cloud, focusing on ASIC design verification and implementation of formal verification methodologies for complex hardware systems.

GPU Design Verification Engineer

GPU Design Verification Engineer role at Apple, focusing on pre-silicon RTL verification of GPU sub-units, requiring 3+ years of experience in hardware verification and GPU architecture.

TPU, Pre-Silicon Validation Engineer, Google Cloud

TPU Pre-Silicon Validation Engineer position at Google Cloud, focusing on hardware acceleration for AI/ML applications and TPU development.