Description
Cerebras Systems is hiring an AI Hardware Performance Engineer to characterize, analyze, and optimize the performance of state-of-the-art AI models on its Cerebras WSE hardware. The role spans the hardware and software stack, including bringing up and optimizing new WSE generations, building kernel-level and end-to-end performance models, optimizing kernel microcode and compiler algorithms, debugging runtime performance, and developing performance-visualization tools. Candidates need a bachelor’s, master’s, or PhD in Electrical Engineering or Computer Science, strong computer architecture knowledge, low-level deep learning or LLM math exposure, at least three years of relevant experience, CPU/GPU simulator experience, performance-profiling experience, and comfort with C++ and Python.
