Hybrid AI Inference Performance Engineer

Intel Corporation; Santa Clara (CA); Hybrid; USD 171,000 - 315,000; Full time

Benefits offered by this job

Hybrid work model

Competitive compensation

Job summary

Intel Corporation is seeking an AI Infrastructure Engineer to push LLM inference on Intel GPUs. You will optimize the end-to-end inference stack, profile bottlenecks, and develop high-performance kernels for attention, MoE, and quantization.

You will upstream improvements to vLLM, SGLang, and PyTorch while collaborating with architecture and compiler teams. Role requires a strong background in GPU computing, modern C++ and Python, and open-source contributions.

Qualifications

• BS in CS/Software/AI or related field with 4+ years of experience, MS/PhD optional.
• 3+ years in GPU computing or HPC with C++ and Python.
• Proven ability to read/modify complex systems-level code.

Responsibilities

• Own end-to-end optimization for running state-of-the-art LLMs on Intel GPUs.
• Profile bottlenecks, write high-performance kernels, and integrate improvements.
• Upstream contributions to vLLM, SGLang, PyTorch, and related stacks.

Skills

GPU computing

C++

Python

Performance optimization

Open-source contributions

Triton/CUDA/CUTLASS

DL/LLM inference

Education

Bachelor's degree in CS/Software/AI

Master's degree or PhD

Tools

vLLM

SGLang

PyTorch

llama.cpp

Job description

Intel Corporation is seeking an AI Infrastructure Engineer to push LLM inference on Intel GPUs. You will optimize the end-to-end inference stack, profile bottlenecks, and develop high-performance kernels for attention, MoE, and quantization.

You will upstream improvements to vLLM, SGLang, and PyTorch while collaborating with architecture and compiler teams. Role requires a strong background in GPU computing, modern C++ and Python, and open-source contributions.