Minimum qualifications:- Bachelor's degree or equivalent practical experience.
- 8 years of experience in software development and system programming.
- Experience with Linux kernel internals, kernel virtualization, or hypervisor development.
- Experience in performance analysis, debugging, and system optimization on server architectures.
Preferred qualifications:- Master's degree or PhD in Computer Engineering, Computer Science, or Electrical Engineering.
- Experience with Kernel-Based Virtual Machine (KVM), two-level paging, VM-exit optimization, vCPU scheduling, and bare-metal microbenchmarks.
- Experience with hardware/software co-design and collaborating with cross-functional teams to influence SoC/CPU feature specifications.
- Expertise in virtualization performance characterization and root-causing virtualization overheads on enterprise server platforms.
About the jobAs a Staff Software Engineer on the Architecture, Analysis, Automation, and Accelerators unit, you will guide chip-system joint engineering for Google's upcoming proprietary chips.
Connecting physical layout with mass-scale data centers, you will investigate low-level operating system components and hypervisors to reduce virtualization latency across multiple compute environments. By constructing direct-hardware testing frameworks, you will pinpoint and unblock critical throughput constraints to steer downstream CPU development.
In this role, excel at untangling ambiguous runtime mysteries and partnering with cross-disciplinary engineering groups to shape the future of worldwide computing.
Individual pay is determined by factors including job-related skills, experience, and relevant education or training.
US: $207000 - $300000 (USD) 20% bonus target equity benefits
Learn more about benefits at Google .
Responsibilities - Lead hardware-system co-design for next-generation custom silicon and SoCs, translating low-level software performance insights into architectural requirements for future processor and server platforms.
- Optimize Linux kernel internals and hypervisors to systematically identify and eliminate virtualization overhead across both x86 and Arm-based hyperscale infrastructure.
- Architect bare-metal microbenchmarks, diagnostic tools, and profiling suites to isolate and resolve complex system bottlenecks spanning CPU cores, caches, and memory hierarchies.
- Drive cross-functional technical alignment across silicon architecture, platform firmware, operating system, and cloud infrastructure teams to set the performance roadmap for planet-scale workloads.
- Mentor and technically direct engineers, fostering rigorous systems performance methodologies, experimental benchmarking standards, and architectural best practices.
