GPU Network & Distributed Systems Engineer

Advanced Micro Devices, Inc.; San Jose (CA); On-site; USD 150,000 - 190,000; Full time

Job summary

Advanced Micro Devices, Inc. (AMD) is seeking a motivated C++ developer to join the Collectives and Network Performance team in San Jose or Austin. You will develop, test and optimize collective communication and networking software for AMD GPUs, aiding multi-GPU and multi-node AI workloads.

You’ll profile workloads, run benchmarks, and collaborate with internal GPU libraries and ROCm teams while learning to map AI model architectures to software and hardware. This role is onsite.

Qualifications

• Understanding of collective communication concepts (e.g., all-reduce, all-to-all) and familiarity with RCCL, NCCL or MPI and networking basics (RDMA, NICs).
• Proficiency in C++ and Python; debugging, profiling and testing experience.
• Familiarity with GPU programming and distributed training or inference concepts.
• Strong Linux, computer architecture and networking fundamentals.
• Experience in performance engineering is a plus.

Responsibilities

• Develop & Test Features: implement, test and debug features in collective communication libraries and network software under senior guidance.
• Profile & Analyze Workloads: profile AI workloads on single-node and multi-node systems; identify performance bottlenecks.
• Run Benchmarks & Experiments: run performance experiments across AMD GPU generations and network configurations; summarize results.
• Contribute to Tooling: contribute to tooling for performance measurement, regression tracking and reporting.
• Support Model & Framework Enablement: support model enablement and out-of-box performance efforts across AI frameworks.
• Software Engineering Best Practices: follow code reviews, testing and documentation to deliver robust code.
• Learn & Grow: build expertise through mentorship, reviews, and hands-on projects.

Skills

Collective Communication

Networking Fundamentals

GPU Programming

C++

Python

Profiling & Debugging

Distributed Training/Inferences

Education

Bachelor's or Master's degree in CS/CE/EE

Tools

RCCL

NCCL

MPI

Job description

Advanced Micro Devices, Inc. (AMD) is seeking a motivated C++ developer to join the Collectives and Network Performance team in San Jose or Austin. You will develop, test and optimize collective communication and networking software for AMD GPUs, aiding multi-GPU and multi-node AI workloads.

You’ll profile workloads, run benchmarks, and collaborate with internal GPU libraries and ROCm teams while learning to map AI model architectures to software and hardware. This role is onsite.