Senior Datacenter GPU Performance Architect

AMD; Austin (TX); On-site; USD 120,000 - 180,000; Full time

Job summary

AMD is seeking a performance-focused engineer to lead post-silicon characterization and optimization for Datacenter products. You'll work on single-node and multi-node deployments, collaborating across sites and time zones to push performance and efficiency.

Ideal candidates bring strong Python/C++ skills, ML framework experience, and a deep background in HPC/ML workloads, plus a track record of driving complex optimization efforts in distributed environments.

Qualifications

• Bachelor's or Master's degree in Computer Engineering, Electrical Engineering, CS, or related field.
• Strong programming skills with Python and ML frameworks encouraged.
• Proficiency in C/C++, Shell scripting and performance tooling.
• Deep knowledge of HPC/ML workloads and scaling behavior.
• Experience with scale-up/scale-out performance analysis at rack- and cluster-level deployments.

Responsibilities

• Develop and maintain automation frameworks for workload execution and performance data collection.
• Become a key stakeholder in product power and performance definition processes.
• Develop, execute, and evolve performance test plans across HPC/ML workloads.
• Drive performance attainment for scale-up and scale-out configurations.
• Analyze power-management interactions to optimize perf and perf-per-watt tradeoffs.
• Identify bottlenecks and drive root-cause analysis to resolution and performance closure.

Skills

Python

ML frameworks

C/C++

Shell scripting

Performance tooling

Collaboration

Education

Bachelor's/Master's in Computer Engineering/Electrical Engineering/CS

Tools

Python tooling

TensorFlow/PyTorch

Job description

AMD is seeking a performance-focused engineer to lead post-silicon characterization and optimization for Datacenter products. You'll work on single-node and multi-node deployments, collaborating across sites and time zones to push performance and efficiency.

Ideal candidates bring strong Python/C++ skills, ML framework experience, and a deep background in HPC/ML workloads, plus a track record of driving complex optimization efforts in distributed environments.