CoDesign & NextGen Performance Engineer

About the role

Cerebras Systems builds the world's largest AI chip, which delivers training and inference speeds over 10 times faster than GPU-based cloud services. This role focuses on characterizing, analyzing and optimizing the performance of state-of-the-art AI models running on Cerebras' hardware, working across the hardware and software stack to identify bottlenecks, improve computational efficiency and help shape the design of the next-generation AI architecture and software.

Key responsibilities

• Bring up and optimize performance on new generations of the Cerebras WSE.
• Build performance models (kernel-level, end-to-end) to estimate the performance of state-of-the-art and customer ML models.
• Optimize and debug kernel microcode and compiler algorithms to improve ML model inference speed, throughput and compute utilization.
• Debug and understand runtime performance on the system and cluster.
• Develop tools and infrastructure to visualize performance data collected from the Wafer Scale Engine and the compute cluster.

Required profile

• Bachelor's, Master's or PhD in Electrical Engineering or Computer Science.
• Strong background in computer architecture.
• 3+ years of experience in a relevant domain such as computer architecture, CPU/GPU performance, kernel optimization or HPC.
• Experience working on CPU/GPU simulators.
• Exposure to performance profiling and debugging on any system pipeline.

Required skills

• Understanding of low-level deep learning / LLM math.
• Comfort with C++ and Python.
• Strong analytical and problem-solving mindset.