GPU Application Platform Engineer Graduate (Server Research and Development) - 2027 Start (PhD)

What you'll do

• Develop application benchmarks, tools, and performance optimizations for GPU/AI systems, including LLM training and inference.
• Analyze system bottlenecks and lead software-hardware co-design improvements; develop GPU/AI system TCO models.
• Evaluate emerging technologies and architectures with partners, suppliers, industry consortiums, and standards committees.

What they ask for

• Completing or recently completed a PhD in Electrical Engineering, Computer Engineering, Computer Science, or a related major.
• Thesis focused on GPU/AI platform architecture, application performance optimization, or software-hardware co-design.
• Deep understanding of computer architecture, especially GPU/AI SoC or platform architecture, interconnect fabric, and memory subsystems.
• Experience in GPU/AI application performance optimization or software-hardware co-design; proficiency in C/C++ and scripting languages such as Python.

Nice to have

• Understanding of LLM architecture and training and inference requirements on accelerators, memory, and networks.

Tools

GPU, AI, LLM, C/C++, Python

AI focus

Optimizing GPU/AI systems for giant-model training and inference, including LLM workloads.