Senior GPU Inference Performance Architect

Advanced Micro Devices; Santa Clara (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Advanced Micro Devices is seeking a Senior GPU Inference Performance Engineer to own end-to-end profiling of GPU-accelerated AI inference workloads. You will analyze workloads across AMD Instinct and NVIDIA GPUs, profile AI serving frameworks, and explain performance gaps with hardware and software evidence.

You will collaborate across teams on multi-server inference networking and Kubernetes-related performance concerns.

Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field preferred.

Responsibilities

• Full-stack GPU profiling across AMD Instinct and NVIDIA GPUs.
• Profile and optimize AI serving frameworks (e.g., vLLM, SGLang) and quantify throughput/latency implications.
• Design and run head-to-head benchmarks to explain performance differences with hardware/software evidence.
• Profile multi-server inference networking and distributed topologies across clusters.
• Identify and mitigate latency/jitter due to GPU operator, container, and Kubernetes stack.

Skills

GPU performance engineering

HPC performance analysis

Python

C/C++

Kubernetes GPU scheduling

LLM/ML serving profiling

Technical communication

Education

Bachelor's degree in Computer Science/Engineering or related field

Tools

ROCm / rocProfiler / Omniperf

CUDA / Nsight Systems / Nsight Compute

RCCL / NCCL

RDMA/RoCE tracing tools

Memory Fabric profiling tools

Job description

Advanced Micro Devices is seeking a Senior GPU Inference Performance Engineer to own end-to-end profiling of GPU-accelerated AI inference workloads. You will analyze workloads across AMD Instinct and NVIDIA GPUs, profile AI serving frameworks, and explain performance gaps with hardware and software evidence.

You will collaborate across teams on multi-server inference networking and Kubernetes-related performance concerns.