Fellow AI Performance & Reliability Engineer (Hybrid)

Advanced Micro Devices; San Jose (CA); Hybrid; USD 180,000 - 240,000; Full time

Job summary

AMD seeks a Fellow Software Engineer to advance AI performance and reliability across training and inference workloads. You will optimize AI systems spanning models, frameworks, and hardware, collaborating with customers and internal teams to improve throughput and stability at scale.

The role demands deep expertise in AI software stacks, performance analysis, and system-level thinking, with strong programming skills in Python and C++. Hybrid work in California or Washington is supported.

Qualifications

• PhD or master's in AI/ML, CS, or related field.
• Experience profiling and optimizing AI workloads.
• Strong foundations in computer architecture and systems performance.

Responsibilities

• Profile and optimize AI model training and inference workloads.
• Improve model throughput, latency, memory efficiency, scalability, and reliability.
• Collaborate with customers to understand requirements and reproduce issues.
• Document performance findings and best practices.

Skills

Strong software engineering

Profiling & optimizing

Python

C++

PyTorch

TensorFlow

JAX

Debugging

Communication

Customer collaboration

Education

PhD or MS in CS/ML

Tools

CUDA

ROCm

HIP

Triton

XLA

MLIR

NCCL

Job description

AMD seeks a Fellow Software Engineer to advance AI performance and reliability across training and inference workloads. You will optimize AI systems spanning models, frameworks, and hardware, collaborating with customers and internal teams to improve throughput and stability at scale.

The role demands deep expertise in AI software stacks, performance analysis, and system-level thinking, with strong programming skills in Python and C++. Hybrid work in California or Washington is supported.