Principal / Senior GPU SW Performance Engineer — Post‑Training in San Jose, California | Advanced Micro Devices, Inc

Amd

Hybrid Actively hiring
San Jose, CA · United States Posted 65 days ago $240,000–$240,000 / year
View original post Log in to save

At a glance

AI generated

TL;DR

As a senior software engineer on the AMD Instinct GPU team, you will drive performance for post-training workloads by optimizing finetuning and reinforcement learning (RL) training solutions across data loaders, kernels, distributed training, and compilers. Your daily tasks include enhancing throughput, memory efficiency, and stability in multi-GPU/multi-node environments while contributing efficient kernels and targeted graph-level optimizations. You will profile, diagnose, and resolve bottlenecks using standard tooling to prevent regressions in continuous integration (CI) systems, ensuring reproducible pipelines and documentation are adopted by internal teams and external developers. Ideal candidates have experience with GPU performance engineering for deep learning on ROCm/HIP or similar platforms, hands-on expertise with SFT, LoRA, and RL-based training at scale, strong PyTorch skills, proficiency in Python and C++, and a track record of turning profiles into fixes and documenting results.

Skills

Python PyTorch C++ ROCm HIP AMD Instinct GPUs Distributed training Multi-GPU/multi-node SFT LoRA RL-based training torch.distributed FSDP ZeRO Distributed systems Collective communication libraries CI/CD

What you'll do

  • Lead performance optimization for finetuning and RL training on AMD GPUs.
  • Enhance throughput, memory efficiency, and stability in multi-GPU/multi-node setups.
  • Develop efficient kernels and graph-level optimizations for deep learning frameworks.
  • Profile and resolve bottlenecks using standard tooling to prevent CI regressions.
  • Ship reproducible pipelines and documentation for internal and external use.

What we're looking for

  • Proven GPU performance engineering for deep learning (ROCm/HIP, Triton, etc.)
  • Hands-on experience with SFT, LoRA, and RL-based training at scale
  • Strong PyTorch expertise including torch.distributed, FSDP/ZeRO or equivalent
  • Proficient in Python and C++; capable of reading/writing kernels
  • Experience optimizing multi-GPU/multi-node training and communication patterns
  • Track record of profiling, diagnosing, and resolving performance bottlenecks

Market check

Salary context

This $240,000–$240,000 range sits above 85% of similar postings on FindRole.

Peer median band

$168,000–$241,500

Median floor and ceiling across peers.

Typical midpoint (25–75%)

$168,500–$235,750

Middle half of comparable postings.

Based on 240 comparable postings.

* 240 is the maximum number of comparable postings sampled.

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 80 open roles on FindRole.

Listed pay typically runs $176,400–$176,400 across 80 roles with salary data.

Most-posted roles

View all roles at Amd

More like this

Similar roles