Senior Cloud & AI Workload Performance Engineer

Socket.dev; Town of Texas (WI); Hybrid; USD 140,000 - 210,000; Full time

Job summary

AMD in Austin, Texas seeks a systems-minded performance engineer to own workloads exposed to hyperscaler CPUs and AI systems. You will measure cloud-native performance, analyze CPU-to-accelerator data movement, and explain business impact with concise, evidence-based results.

You will work across architectures, coordinate with architecture experts, and develop scalable methodologies for forecast inputs and partner reviews. Hybrid work option available.

Qualifications

• Experience benchmarking cloud, distributed or accelerated-system workloads.
• Strong Linux systems skills and automation for deployment and analysis.
• Experience with containers and cloud-native workloads at scale.
• Ability to design fair experiments across CPU architectures.

Responsibilities

• Define and maintain cloud-native and AI workload taxonomy for analysis.
• Design reproducible methods for web services, containers and AI data paths.
• Compare AMD, Intel, Arm and cloud environments with aligned software versions.
• Analyze host-CPU impact on accelerator utilization and system throughput.
• Measure performance-per-watt, density and price-performance defensibly.
• Document cross-ISA methodology for scrutiny by partners and customers.
• Support validation with cloud/OEM/ISV partners and produce reports.

Skills

Linux systems

Benchmarking

Automation scripting

Containers

Cloud-native workloads

Performance profiling

Education

Bachelors or Masters in electrical or computer engineering

Tools

Perf tooling

Docker

Kubernetes

Arm Neoverse

Job description

AMD in Austin, Texas seeks a systems-minded performance engineer to own workloads exposed to hyperscaler CPUs and AI systems. You will measure cloud-native performance, analyze CPU-to-accelerator data movement, and explain business impact with concise, evidence-based results.

You will work across architectures, coordinate with architecture experts, and develop scalable methodologies for forecast inputs and partner reviews. Hybrid work option available.