Principal GPU Workload Architect (Remote)

Advanced Micro Devices, Inc.; San Jose (CA); Hybrid; USD 150,000 - 200,000; Full time

Job summary

Advanced Micro Devices, Inc. is seeking a Principal Modeling Architect in San Jose, CA, responsible for leading advanced workload modeling for AI/ML and HPC in data center environments. Candidates should have over 12 years of experience in workload modeling and performance engineering, demonstrating expertise in analyzing large-scale workloads on GPU platforms. The position offers the flexibility to work in the Bay Area or remotely. This role is crucial for optimizing AMD’s next-generation GPU platforms.

Qualifications

• 12+ years of experience in workload modeling or related technical domains.
• Expertise in analyzing AI/ML, HPC, or large‑scale data on GPU platforms.
• Deep understanding of performance modeling methodologies.

Responsibilities

• Develop workload modeling frameworks for AI/ML and HPC workloads.
• Analyze emerging model architectures and scaling methodologies.
• Collaborate with design teams to translate insights into requirements.
• Lead benchmarking and simulation efforts.
• Produce detailed workload characterization reports.

Skills

Workload modeling

Performance engineering

System architecture

AI/ML analysis

Data analytics

Team collaboration

Education

Advanced degree in Computer Science or Electrical Engineering

Tools

ROCm

CUDA

Job description

Advanced Micro Devices, Inc. is seeking a Principal Modeling Architect in San Jose, CA, responsible for leading advanced workload modeling for AI/ML and HPC in data center environments. Candidates should have over 12 years of experience in workload modeling and performance engineering, demonstrating expertise in analyzing large-scale workloads on GPU platforms. The position offers the flexibility to work in the Bay Area or remotely. This role is crucial for optimizing AMD’s next-generation GPU platforms.