below marketmost similar roles pay hereabove market
This role pays more than
62%
of similar roles. Most pay
$169,050–$235,750
— the blue band above.
At the midpoint, this role pays about
$215k
versus about
$202k
for comparable roles.
Based on 240 similar postings.
Employer
About Amd
AMD (Advanced Micro Devices) is a semiconductor company that develops high-performance processors, graphics cards, and adaptive computing solutions for gaming, data centers, and embedded markets. Industry: Semiconductors
Amd currently has
500 open roles
on FindRole.
Listed pay typically runs
$172,000–$258,000
across 500 roles with salary data.
Distributed Systems Software Engineer, Collectives joins the Collectives and Network Performance team to develop, test, and optimize collective communication and networking software for AMD GPUs. The role involves implementing features in collective communication libraries, profiling performance bottlenecks for LLM, diffusion, and generative AI workloads, and running experiments across various GPU generations and network configurations. Key responsibilities include contributing to performance measurement tooling and supporting model enablement across AI frameworks. Candidates must possess proficiency in C++ and Python, with a solid foundation in Linux environments, computer architecture, and networking concepts like RDMA. The position requires experience with collective communication concepts such as all-reduce, and familiarity with libraries like RCCL, NCCL, or MPI. The work focuses on improving training and inference performance across multi-GPU and multi-node systems.
Implement, test, and debug features in collective communication libraries and network software.
Profile LLM, diffusion, and generative AI workloads to identify performance bottlenecks on multi-node systems.
Run performance experiments across various AMD GPU generations and network configurations.
Summarize experimental results and contribute to performance measurement and regression tracking tools.
Support new model enablement and out-of-box performance efforts across various AI frameworks.
Coordinate with internal teams and open-source communities to get software changes reviewed and integrated.
Follow software engineering best practices including code reviews, testing, and documentation to deliver maintainable code.
What we're looking for
Bachelor’s or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent.
Proficiency in C++ and Python with experience in debugging, profiling, and writing tests (preferred).
Solid foundation in Linux environments, computer architecture, and networking concepts.
Professional experience in software development with exposure to performance engineering, systems software, networking, or GPU programming (preferred).
Understanding of collective communication concepts and familiarity with libraries such as RCCL, NCCL, or MPI (preferred).
Familiarity with networking fundamentals such as RDMA, NICs, and interconnects (preferred).
Foundational knowledge of GPU architecture and experience with HIP, CUDA, or similar GPU programming (preferred).
Familiarity with AI frameworks such as PyTorch, JAX, vLLM, or SGLang and distributed training/inference concepts (preferred).
CUDA
C++
Python
Deep Learning
Distributed Computing
PyTorch
JAX
TensorRT
vLLM
Triton
XLA
NCCL
MPI
UCX
Kernel Optimization
Model Optimization
FP8
INT8
Generative AI