ML Runtime & Kernel Engineer — High-Performance AI

We Love AI Jobs; Sunnyvale, Northern (CA, KY); Hybrid; USD 210,000 - 270,000; Full time

Job summary

Cerebras Systems in Sunnyvale, CA, is seeking a Runtime and Kernel Engineer to design and implement high-performance runtime components and kernels for Core ML algorithms on the Cerebras platform.

You will profile, optimize, and debug across ML frameworks, compilers, runtimes, and kernel layers, and translate research prototypes into robust demonstrations. Strong C++, Python, and GPU programming skills are essential for this hybrid role.

Qualifications

• Experience developing high-performance systems software and ML systems.
• Strong programming in C++ and Python with debugging and profiling skills.
• Familiarity with PyTorch or JAX and ML framework concepts.
• Ability to translate research ideas into robust software implementations.

Responsibilities

• Design and implement runtime components and high-performance kernels for Core ML algorithms.
• Profile and debug performance across ML frameworks, compilers, runtimes, and kernels.
• Build benchmarks, instrumentation, and automated tests for functionality and accuracy.
• Collaborate with Core ML researchers and engineers to evaluate design options and deliver end-to-end capabilities.

Skills

C++

Python

High-performance kernels

Runtime and compiler systems

Parallel programming

Performance profiling and optimization

PyTorch

JAX

CUDA

Education

Bachelor’s, Master’s, PhD, or equivalent experience

Tools

Triton

CUDA

Job description

Cerebras Systems in Sunnyvale, CA, is seeking a Runtime and Kernel Engineer to design and implement high-performance runtime components and kernels for Core ML algorithms on the Cerebras platform.

You will profile, optimize, and debug across ML frameworks, compilers, runtimes, and kernel layers, and translate research prototypes into robust demonstrations. Strong C++, Python, and GPU programming skills are essential for this hybrid role.