ML Systems Performance Engineer
Core
Drive end-to-end model inference speed and throughput by operating at the intersection of hardware and software, focusing on low-level kernel debugging, system-level analysis, and performance modeling.
Role type
ML Systems Performance Engineer
Builds
Performance models, optimized kernel micro code, compiler algorithms, and diagnostic tooling for the Cerebras WSE.
Domain
Hardware/Software intersection, specifically Cerebras Wafer Scale Engine (WSE) and high-performance computing (HPC).
Deliverable
production ML models | infrastructure
Required skills
Computer architecture, low-level deep learning / LLM math, performance profiling and debugging, C++, Python, CPU/GPU simulators, kernel optimization, HPC.
Preferred skills
null
Technologies
Cerebras WSE, C++, Python
Responsibilities
Build performance models to estimate ML model performance; Optimize and debug kernel micro code and compiler algorithms; Debug and understand runtime performance on the system and cluster; Develop tools and infrastructure to visualize performance data.
Seniority
Mid-level, hands-on IC