Senior AI Kernel Engineer — GPU Inference & Kernel Optimization

Modular; United States; Hybrid; USD 198,000 - 286,000; Full time

Benefits offered by this job

Premier insurance plans

401k matching

Flexible paid time off

Team building events

Job summary

Modular is seeking a Senior AI Kernel Engineer to lead design and optimization of high-performance kernels for AI inference on GPUs. This role involves working at the intersection of hardware and AI systems, ensuring efficient implementations at scale. The ideal candidate will have over 5 years of experience, strong C/C++ skills, and expertise in GPU programming. The position offers competitive compensation, including stock options and world-class benefits, with the office located in Los Altos, CA or the option to work remotely.

Qualifications

• 5+ years of experience in performance-critical systems or kernel development.
• Strong proficiency in C/C++ and low-level programming.
• Extensive hands-on experience with GPU kernel programming (CUDA, HIP, or equivalent).

Responsibilities

• Design, implement, and optimize performance-critical kernels for AI inference workloads.
• Lead kernel-level optimization efforts across various hardware environments.
• Analyze performance using profilers, hardware counters, and translate insights into improvements.

Skills

Performance-critical systems

C/C++ proficiency

GPU kernel programming

GPU architecture understanding

Problem-solving skills

Education

5+ years experience

Tools

CUDA

HIP

Job description

Modular is seeking a Senior AI Kernel Engineer to lead design and optimization of high-performance kernels for AI inference on GPUs. This role involves working at the intersection of hardware and AI systems, ensuring efficient implementations at scale. The ideal candidate will have over 5 years of experience, strong C/C++ skills, and expertise in GPU programming. The position offers competitive compensation, including stock options and world-class benefits, with the office located in Los Altos, CA or the option to work remotely.