below marketmost similar roles pay hereabove market
This role pays more than
91%
of similar roles. Most pay
$174,600–$242,862
— the blue band above.
At the midpoint, this role pays about
$300k
versus about
$209k
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.
As a Principal Distributed Systems Software Engineer, Collectives, you will join the Collectives and Network Performance team to define and deliver industry-leading AI pre-training and distributed inference performance on AMD GPUs. You will lead the design and development of collective communication libraries, optimizing algorithms like all-reduce and all-to-all across scale-up and scale-out networks. Your daily work involves performance tuning, profiling, and analyzing large-scale LLM, diffusion, and generative AI models on multi-node systems. You will utilize C++, Python, and Linux to collaborate with hardware, compiler, and runtime teams while integrating optimizations into frameworks like PyTorch, JAX, vLLM, and SGLang. This role solves critical bottlenecks in distributed training by mapping model architectures to low-level software and hardware through hardware-software co-design and performance modeling.
Define the goals, scope, and roadmap for AMD’s collective communication and network optimizations.
Lead the design and development of collective algorithms for scale-up and scale-out networks.
Perform performance tuning and profiling for large-scale AI models on multi-node systems.
Participate in the hardware-software co-design of future scale-up networks and NICs.
Develop frameworks and tools for performance estimation, modeling, and regression tracking.
Integrate optimizations into internal GPU libraries and open-source frameworks like PyTorch and JAX.
Provide technical guidelines on network load balancing, workload scheduling, and model sharding.
Mentor junior team members and provide strategic direction through code reviews and technical guidance.
What we're looking for
Master’s and/or PhD degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
Industry-leading expertise in collective communication and network performance.
Deep expertise in C++ and Python with experience in performance tuning, debugging, and robust test design (preferred).
Deep knowledge of network, NIC, and GPU hardware architecture.
Experience mapping model architectures to low-level software and hardware.
Experience optimizing AI workloads for performance and resource utilization across large-scale, heterogeneous compute environments (preferred).
Experience integrating optimized communication into ML frameworks such as PyTorch, JAX, vLLM, and SGLang (preferred).
Senior level of industry experience in technical software development focusing on performance engineering, communication libraries, and networking (preferred).