Senior AI Training Performance Architect
Core
Architecting and optimizing AI training workloads across the full hardware/software stack to maximize performance on NVIDIA's compute systems.
Role type
Senior IC AI Training Performance Architect
Builds
Production-quality software for NVIDIA's deep learning platform stack and MLPerf Training benchmark submissions.
Domain
AI/ML training, GPU architecture, high-performance computing
Deliverable
production ML models
Required skills
Deep learning and neural networks training, GPU architecture fundamentals, application performance tuning, processor/system-level performance modeling, C++, Python, CUDA
Preferred skills
PhD in CS/EE/CSEE, experience with proprietary processor/system simulators
Technologies
CUDA, C++, Python, MLPerf, NVIDIA DL frameworks, GPU simulators
Responsibilities
Profile and optimize AI training workloads on state-of-the-art hardware; identify and resolve performance bottlenecks in GPU-based AI training; implement production software across drivers to DL frameworks; develop tools to automate workload analysis and optimization workflows.
Seniority
Senior, hands-on IC
