Senior Kernel Optimization Engineer for AI Hardware

Cerebras; United States; Remote; USD 140,000 - 180,000; Full time

Job summary

Cerebras Systems seeks a Kernel Engineer to push the limits of AI and HPC kernels on our massive parallel processor. You will design, implement, optimize, and validate high-performance kernels that fully leverage Cerebras hardware, contributing to world-class throughput and efficiency.

You will collaborate with hardware architects and software teams to scale kernels, libraries, and tooling for state-of-the-art neural networks and simulations, driving performance across CPU/GPU-less architecture.

Qualifications

• Experience with kernel development for ML/HPC workloads.
• Strong knowledge of C++ and Python; debugging skills.
• Familiarity with low-level assembly or CSL languages is a plus.

Responsibilities

• Develop design specs for ML and linear algebra kernels for Cerebras WSE System.
• Develop and debug kernel libraries using low-level assembly and CSL language.
• Evaluate performance with models and analysis to inform design decisions.
• Create unit and system tests to verify kernel functionality.
• Collaborate with chip and system architects to optimize instruction sets and IO.

Skills

C++

Python

Debugging

Education

Bachelor's in CS/CE/Math

MS/PhD in related field

Job description

Cerebras Systems seeks a Kernel Engineer to push the limits of AI and HPC kernels on our massive parallel processor. You will design, implement, optimize, and validate high-performance kernels that fully leverage Cerebras hardware, contributing to world-class throughput and efficiency.

You will collaborate with hardware architects and software teams to scale kernels, libraries, and tooling for state-of-the-art neural networks and simulations, driving performance across CPU/GPU-less architecture.