Hardware Systems Engineer - AI Hardware
Core
Design, qualify, and maintain Cloudflare's worldwide fleet of GPU servers and AI infrastructure (compute, networking, storage, power, cooling) to serve 20% of global internet traffic.
Role type
Senior IC hardware systems engineer (AI infrastructure)
Builds
Rack-scale GPU systems, AI server platforms, and high-speed interconnects for hyperscale cloud infrastructure
Domain
Cloud computing / AI hardware infrastructure
Deliverable
production ML models
Required skills
AI accelerator architecture, rack-scale systems design, high-speed interconnects (NVLink, InfiniBand, RoCE), thermal management, vendor qualification, Python/Bash/Go scripting, performance benchmarking, Redfish/IPMI protocols
Preferred skills
ODM/OEM server customization, RDMA networking, LLM training/inference workload patterns, open compute hardware programs, kernel/driver development, observability tools (Prometheus, Grafana)
Technologies
GPUs, custom ASICs, SmartNICs, NVLink, InfiniBand, RoCE, HGX, OAC, OAM, UBB, Prometheus, Grafana, Git, TeamCity, Jira
Responsibilities
Evaluate and qualify AI hardware platforms based on workload requirements; design and validate rack-scale GPU systems; run hands-on qualification (benchmarking, stress testing, failure analysis); assess ODM/OEM server platforms; build tooling and automation for qualification workflows; communicate results via technical documentation and internal talks
Seniority
Senior, hands-on IC
