LLM AI Inference Performance Engineer

Intel; California (MO); Hybrid; USD 171,000 - 315,000; Full time

Job summary

Intel is seeking an AI Infrastructure Engineer to push LLM inference on next‑generation GPUs. You will profile stacks, write high‑performance kernels, and upstream optimizations to vLLM, SGLang, and PyTorch, shaping Intel hardware for GenAI workloads.

This hybrid role is based in the United States with primary location Santa Clara, CA, and includes multiple US sites. Strong C++/Python and HPC experience are required, with 4+ years in GPU computing or AI systems.

Qualifications

• 4+ years in GPU computing or HPC.
• Proficiency in C++ and Python.
• Bachelor's degree or higher in CS/SE/AI.

Responsibilities

• Profile end-to-end inference pipeline for running state-of-the-art LLMs on Intel GPUs.
• Profile, diagnose, and resolve cross-stack performance bottlenecks.
• Design, write, and optimize custom high-performance kernels for attention, MoE, quantization, and operator fusions.
• Upstream architectural improvements to vLLM, SGLang, and PyTorch; bridge hardware teams with the open-source community.
• Apply roofline analysis and profiling to shape future Intel GPU roadmaps.
• Show passion about AI infrastructure and performance optimization.

Skills

C++

Python

GPU computing

HPC

Education

Bachelor's degree in CS/SE/AI

Masters or PhD (advantage)

Tools

vLLM

SGLang

PyTorch

CUDA/CUTLASS

Job description

Intel is seeking an AI Infrastructure Engineer to push LLM inference on next‑generation GPUs. You will profile stacks, write high‑performance kernels, and upstream optimizations to vLLM, SGLang, and PyTorch, shaping Intel hardware for GenAI workloads.

This hybrid role is based in the United States with primary location Santa Clara, CA, and includes multiple US sites. Strong C++/Python and HPC experience are required, with 4+ years in GPU computing or AI systems.