Senior GPU Performance Engineer — Low-Level Kernel Mastery

Intel; Santa Clara (CA); Hybrid; USD 195,000 - 276,000; Full time

Benefits offered by this job

Stock bonuses

Health benefits

Retirement plan

Flexible hybrid/remote work

Job summary

Intel's Software and AI (SAI) organization is seeking a highly skilled software engineer to contribute to the development and low-level optimization of oneDNN, a cross-platform open-source performance library that underpins modern AI workloads.

This role focuses on high-performance GEMM, convolution, and attention kernels, GPU kernel generation, memory-traffic optimizations, and close collaboration with hardware and compiler teams on next-generation Intel GPUs. Hybrid work is available.

Qualifications

• Education: BSc, MSc, or PhD in Computer Science, Computer Engineering, Mathematics, Physics, or a highly technical related field
• 5+ years of professional software development experience with expert-level modern C++
• 2+ years of GPU kernel optimization (via SYCL/DPC++, OpenCL, CUDA, or HIP) or 5+ years of similar low-level CPU optimization
• Strong foundations in computer architecture, cache hierarchies, memory subsystems, and parallel programming paradigms

Responsibilities

• Kernel Development and Architecture: Develop high-performance GEMM, convolution, and attention kernels for AI workloads
• Low-Level Optimization: Implement fusion and memory-traffic optimizations to maximize hardware utilization
• Performance Modeling and Profiling: Build analytical and empirical performance models for kernel dispatch and tuning
• Hardware and Software Co-Design: Co-design GPU primitives and kernel architectures for next-generation Intel GPUs
• Infrastructure and Validation: Improve validation, benchmarking, and CI infrastructure for performance-critical GPU workloads

Skills

Modern C++

GPU kernel development

Performance optimization

Parallel programming

Education

BSc/MSc/PhD in CS/CE/Math/Physics

Tools

SYCL/DPC++

OpenCL

CUDA

HIP

Job description

Intel's Software and AI (SAI) organization is seeking a highly skilled software engineer to contribute to the development and low-level optimization of oneDNN, a cross-platform open-source performance library that underpins modern AI workloads.

This role focuses on high-performance GEMM, convolution, and attention kernels, GPU kernel generation, memory-traffic optimizations, and close collaboration with hardware and compiler teams on next-generation Intel GPUs. Hybrid work is available.