Apply on Employer Site

Google DeepMind · 3 days ago

Staff Hardware Systems Architect

Mountain View, CA

Full-time

Onsite

Lead/Staff

$183K/yr - $271K/yr

10+ years exp

Google DeepMind is a team of scientists and engineers working together to advance artificial intelligence. They are seeking a highly motivated Hardware Systems Architect to design and build next-generation datacenter infrastructure focused on machine learning acceleration, playing a pivotal role in shaping server architectures and optimizing system performance.

Artificial Intelligence (AI)Business DevelopmentMachine Learning

Growth Opportunities

Responsibilities

Lead the architectural design of large-scale, high-density server platforms, considering system topology, power distribution, advanced cooling methodologies, and mechanical constraints

Architect the physical, electrical, and thermal design of server racks, hosts, trays, and other datacenter hardware from concept to deployment

Drive system-level decisions for host and server architecture, including motherboard design, power delivery, memory subsystems, and high-speed interconnects

Oversee and steer the design and development of complex printed circuit boards, ensuring signal and power integrity for cutting-edge interconnects and components

Define and analyze system requirements for reliability, availability, and serviceability (RAS) at the server and rack level

Develop and analyze system-level performance models to guide architectural trade-offs between performance, power, and cost

Collaborate with software and firmware teams to ensure seamless integration and co-optimization across the entire system stack

Lead cross-functional teams to drive technical alignment and decision-making from concept to high-volume deployment

Qualification

Systems architectureDatacenter hardware designHigh-performance computingServer architectureThermal managementPower deliveryHigh-speed interfacesPerformance modelingAnalytical skillsProblem-solvingCommunication skillsLeadership skillsCollaboration skills

Required

Master's or Ph.D. in Electrical Engineering, Computer Engineering, or a related field

10+ years of experience in systems architecture, with a primary focus on datacenter and server hardware design

Proven track record of architecting and delivering complex, high-performance computing systems at scale

Deep expertise in server and host architecture, including motherboard design, power delivery, and thermal management for high-power components

In-depth understanding of datacenter power and cooling infrastructure, including both air and liquid cooling solutions at the rack and facility level

Extensive experience with high-speed system-level interfaces (e.g., ICI, PCIe Gen5/6, CXL, high-speed Ethernet)

Experience with system-level performance modeling and analysis

Preferred

Experience influencing the selection and design of SoC interfaces, with a focus on high-speed serial interconnects

Knowledge of chiplet-based design methodologies and advanced semiconductor packaging technologies (e.g., 2.5D/3D integration)

Expertise in datacenter networking, including leaf-spine topologies and RDMA

Working knowledge of transformer-based large language models is a plus

Knowledge of high-performance and low-power architectures for ML acceleration

Exceptional problem-solving and analytical skills

Excellent written and verbal communication skills, with the ability to present complex technical concepts to a variety of audiences

Strong leadership and collaboration skills, with the ability to influence and guide cross-functional teams

Benefits

Bonus

Equity

Benefits

Company

Google DeepMind

Glassdoor4.5

Google DeepMind aims to research and build safe artificial intelligence system to solve intelligence and advance science and humanity. It is a sub-organization of Google.

Founded in 2010

London, England, GBR

501-1000 employees