AI Systems Performance Engineer

What you'll do

• Benchmark and characterize AI inference workloads across AMD embedded and client platforms.
• Execute performance, power, and efficiency analysis for LLMs, VLMs, and computer vision applications.
• Develop and maintain automated benchmarking environments and performance dashboards.
• Configure and optimize AI workload execution environments using industry-standard frameworks and runtimes.
• Collect and interpret system-level performance traces using hardware and software profiling tools.
• Investigate workload bottlenecks involving compute, memory bandwidth, latency, and system resource utilization.
• Support the development of benchmarking methodologies for competitive analysis and product evaluation.
• Generate data-driven reports and present technical findings to engineering stakeholders.

What we're looking for

• Bachelor's or Master's degree in Computer Engineering, Electrical Engineering, Computer Science, or a related field.
• Experience with AI inference frameworks such as PyTorch, ONNX Runtime, TensorFlow, vLLM, llama.cpp, or similar technologies (preferred).
• Familiarity with modern AI model architectures including LLMs, VLMs, CNNs, and Vision Transformer (ViT) workloads (preferred).
• Strong Linux systems knowledge and experience working in command-line environments (preferred).
• Experience with performance profiling, workload tracing, and system analysis tools (preferred).
• Understanding of CPU, GPU, NPU, memory, and system-level performance interactions (preferred).
• Experience with Python, Shell scripting, or automation frameworks (preferred).
• Knowledge of AMD, x86, ARM, or embedded computing platforms (preferred).