Principal Software Development Engineer

Advanced Micro Devices, Inc•San Jose, CA

About The Position

AMD is seeking a GPU performance engineering lead to maximize the performance and efficiency of large-scale AI training, reinforcement learning, and inference workloads on AMD GPU platforms. The role drives innovation across the software-hardware stack, optimizing system throughput, scalability, and utilization for generative AI workloads. The position calls for deep expertise in GPU performance analysis, distributed systems, and machine-learning workloads, as well as the ability to influence architecture, software ecosystems, and organizational best practices. The successful candidate will work across layers—from kernels and runtimes to frameworks and distributed strategies—and help shape technical direction and future GPU architecture and software stack design.

Requirements

  • Deep expertise in GPU architecture and performance characteristics, including compute units, memory hierarchy, and interconnects such as PCIe, Infinity Fabric, and RDMA.
  • Strong experience with performance profiling tools, such as ROCm tools, Nsight-like systems, or custom profilers, and with bottleneck analysis.
  • Proficiency in Python and at least one systems language (C++, CUDA, or HIP), including debugging and low-level optimization.
  • Master’s degree, PhD, or equivalent experience in Computer Science, Computer Engineering, or a related field.

Nice To Haves

  • Proven experience optimizing large-scale distributed training workloads across thousands of GPUs.
  • Experience with frameworks such as Megatron-LM, Torchtitan, vLLM, Sglang, or equivalent.
  • Strong understanding of communication libraries and patterns, such as NCCL/RCCL, collective operations, and overlap of compute and communication.
  • Experience with compiler stacks, kernel optimization, or graph-level optimization.
  • Demonstrated technical leadership and ability to influence cross-functional teams.

Responsibilities

  • Lead performance optimization of large-scale AI training, reinforcement learning, and inference workloads on AMD GPU platforms in single-node and multi-node environments.
  • Identify and eliminate bottlenecks across compute, memory, and communication, including kernel efficiency, memory bandwidth, and network utilization.
  • Drive cross-stack optimizations spanning kernels, compilers, runtimes, communication libraries, and machine-learning frameworks.
  • Develop and apply advanced profiling, benchmarking, and performance-modeling methodologies.
  • Collaborate with hardware, compiler, and framework teams to influence next-generation GPU architecture and software stack design.
  • Contribute to and lead open-source efforts to improve ecosystem performance on AMD platforms.
  • Stay current with advances in large-scale systems and performance optimization techniques.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service