CoDesign & NextGen Performance Engineer

Job Title

CoDesign & NextGen Performance Engineer

Role Summary

This role focuses on characterizing, analyzing, and optimizing the performance of state-of-the-art AI models running on Cerebras hardware. The engineer will work across hardware and software (kernel, compiler, runtime, cluster) to identify bottlenecks, improve computational efficiency, and influence next-generation architecture and software systems.

Experience Level

Mid-level. Requires approximately 3+ years of relevant experience in computer architecture, CPU/GPU performance, kernel optimization, or HPC.

Responsibilities

Primary responsibilities include:

• Bring up and optimize performance on new generations of the Cerebras Wafer Scale Engine (WSE).
• Build kernel-level and end-to-end performance models to estimate model performance.
• Optimize and debug kernel microcode and compiler algorithms to improve inference speed, throughput, and compute utilization.
• Investigate and resolve runtime and cluster-level performance issues.
• Develop tools and infrastructure to collect, analyze, and visualize performance data from the WSE and compute clusters.

Requirements

Must-have skills and experience:

• Strong background in computer architecture.
• Familiarity with low-level deep learning / LLM math.
• 3+ years of relevant experience (architecture, CPU/GPU performance, kernel optimization, or HPC).
• Experience with CPU/GPU simulators.
• Experience with performance profiling and debugging across system pipelines.
• Proficiency in C++ and Python.
• Strong analytical and problem-solving mindset.

Nice-to-have:

• Experience with compiler optimization, kernel microcode development, or cluster-scale performance tooling.

Education Requirements

Bachelor's, Master's, or PhD in Electrical Engineering, Computer Science, or a related technical field (as listed in the posting).

About the Company

Company: Cerebras

Headquarters: Sunnyvale, CA, USA

Developer of wafer-scale AI accelerators, Cerebras designs the Wafer Scale Engine (WSE)—one of the world’s largest AI chips—to deliver high-speed training and inference solutions for model labs, enterprises, and AI-native startups.