Senior GPU Memory Architect
Core
Designing and optimizing GPU memory systems, micro-architecture, and management to maximize performance, power efficiency, area, and complexity for NVIDIA's AI, gaming, and autonomous vehicle solutions.
Role type
Senior IC GPU Memory Architect
Builds
High-performance GPU memory subsystems and management units
Domain
Semiconductor / Computer Architecture / GPU
Deliverable
production ML models
Required skills
GPU/CPU architecture, memory caching, interconnects, performance simulation, RTL simulation, silicon debugging, hardware memory management unit design, multi-core systems, memory coherency
Preferred skills
PhD in computer architecture, mentoring junior engineers, hardware prefetching design
Technologies
RTL simulation, silicon, high-level functional models
Responsibilities
Develop architecture and micro-architecture for GPU memory systems; Participate in performance simulation to improve memory access efficiency; Implement and maintain high-level functional and performance models; Analyze benchmarks and workloads to identify optimization areas; Define and perform experiments to study machine behavior; Create architectural specifications and customer-facing documents; Debug performance and functional issues across models, simulation, and silicon.
