Open-Source AI Workload Engineer — GPU Performance

Socket.dev; San Jose (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

AMD is seeking an influential engineer to optimize AI workloads on AMD GPUs. This is a founding member role shaping the future of AI computing solutions and driving performance at scale.

You will work across kernel and framework layers and collaborate with software and hardware teams to deliver significant improvements. The ideal candidate has a bias for action, thrives in fast-paced environments, and can apply first-principles thinking to complex problems.

Qualifications

• Bachelor's, Master's, or Ph.D. in Computer Engineering, Computer Science, Electrical Engineering, or related technical field.

Responsibilities

• Optimize AMD GPU performance for open-source LLM frameworks such as vLLM and SGLang.
• Analyze and resolve performance bottlenecks in AI workloads at the kernel level.
• Integrate AMD software stacks (ROCm, ATen) into open-source frameworks including PyTorch, JAX, and Triton.
• Collaborate cross-functionally with software and hardware engineering teams to identify gaps and deliver performance improvements.
• Build strong technical relationships with internal teams and external partners.

Job description

AMD is seeking an influential engineer to optimize AI workloads on AMD GPUs. This is a founding member role shaping the future of AI computing solutions and driving performance at scale.

You will work across kernel and framework layers and collaborate with software and hardware teams to deliver significant improvements. The ideal candidate has a bias for action, thrives in fast-paced environments, and can apply first-principles thinking to complex problems.