GPU Kernel Engineer: Scale AI with Equity

Sciforium; San Francisco, Northern (CA, KY); Hybrid; USD 180,000 - 260,000; Full time

Benefits offered by this job

Medical, dental, vision insurance

401k plan

Competitive salary and equity

Job summary

Sciforium is seeking a GPU Kernel Engineer to push performance on modern accelerators. You will design and optimize custom GPU kernels, from low-level development to integrating ops in ML frameworks used for large-scale training and inference.

Ideal candidates have 5+ years in GPU kernel development, strong C++/Python skills, and deep CUDA/ROCm expertise. Collaboration with ML researchers and hardware teams is required to advance compute performance.

Qualifications

• 5+ years of GPU kernel development or HPC experience.
• Strong programming in C++, Python, and ML frameworks.

Responsibilities

• Design, implement, and optimize custom GPU kernels.
• Profile and optimize end-to-end ML operation performance for large-scale workloads.
• Integrate kernels into PyTorch, JAX, and internal runtimes.
• Develop performance models and identify bottlenecks.

Skills

CUDA/ROCm

C++

Python

GPU kernel development

ML frameworks

PTX/GPU ASM

JAX Pallas

LLM workloads

Education

Bachelor’s/Master’s/PhD in CS/CE/EE/Applied Math

Tools

Triton

JAX Pallas

Job description

Sciforium is seeking a GPU Kernel Engineer to push performance on modern accelerators. You will design and optimize custom GPU kernels, from low-level development to integrating ops in ML frameworks used for large-scale training and inference.

Ideal candidates have 5+ years in GPU kernel development, strong C++/Python skills, and deep CUDA/ROCm expertise. Collaboration with ML researchers and hardware teams is required to advance compute performance.