Principal AI Systems Engineer - GPU Performance Lead

Advanced Micro Devices; San Jose (CA); On-site; USD 190,000 - 230,000; Full time

Job summary

Advanced Micro Devices is seeking a Principal Engineer to lead a hands‑on technical team focused on the performance and scalability of frontier AI workloads on AMD GPUs. You will guide long‑term technical vision, influence architecture decisions, and tackle the toughest challenges across the stack—from kernels to serving frameworks and distributed systems.

The ideal candidate will combine deep GPU architecture expertise with end‑to‑end AI systems knowledge and a proven track record of

Qualifications

• Deep software/hardware expertise in GPU computing, HPC, or AI systems.
• Broad understanding of GPU micro-architecture, memory hierarchy, and performance tradeoffs.
• End-to-end AI systems knowledge: model architectures, inference paradigms, and system design.
• Experience with multi-GPU topologies and high-performance distributed serving.
• Strong background in performance engineering: profiling and roofline analysis.

Responsibilities

• Lead a small technical team of engineers and set technical direction while remaining hands‑on.
• Define long‑term technical strategy for AI workload performance on AMD GPUs.
• Tackle cross‑stack performance challenges from kernel optimization to framework architecture decisions.
• Design and implement novel GPU kernels, compiler optimizations, and framework features.
• Establish performance methodology and contribute to upstream open‑source AI frameworks.

Skills

GPU computing

HPC

AI systems

C++

Python

Linux

Open-source contributions

Leadership

English communication

Education

Master’s or PhD in Computer Science/Engineering

PhD strongly preferred

Tools

CUDA

HIP

OpenCL

Triton/Gluon

CUTLASS

CK

LLVM

MLIR

PyTorch

Job description

Advanced Micro Devices is seeking a Principal Engineer to lead a hands‑on technical team focused on the performance and scalability of frontier AI workloads on AMD GPUs. You will guide long‑term technical vision, influence architecture decisions, and tackle the toughest challenges across the stack—from kernels to serving frameworks and distributed systems.

The ideal candidate will combine deep GPU architecture expertise with end‑to‑end AI systems knowledge and a proven track record of