Distinguished GPU Memory Simulation Architect

NVIDIA•Us, CA

About The Position

We are looking for a Distinguished Engineer to help define the future of GPU memory, on-chip interconnect, and architectural simulation at NVIDIA. In this role, we will rely on you to set technical direction for next-generation memory and Network-on-Chip (NoC) performance and functional modeling, including how we use AI-assisted methods to build, validate, and scale architectural models. We focus this role on long-term architecture strategy, simulation methodology, and product impact. We need someone who can connect GPU architecture, memory systems, interconnect behavior, workload analysis, and modeling infrastructure into a coherent view of future system performance. You will help guide multi-generation decisions, identify where simulation capability needs to evolve, and ensure our modeling investments produce meaningful architectural insight.

Requirements

Bachelor's, Master's, or PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related field, or equivalent experience.
18+ years of relevant professional experience, including a record of setting technical direction for complex architecture, modeling, or simulation systems.
Deep experience in GPU architecture, including memory systems, cache hierarchy, and on-chip interconnect design.
Strong performance modeling expertise, including experience building, scaling, and validating architectural simulation infrastructure.
Experience with large-scale software development and strong programming skills in C/C++, Python, or other scripting languages.
Sound judgment in ambiguous technical areas, with the ability to connect model behavior, workload characteristics, architectural trade-offs, and product requirements.
Clear communication, strong technical leadership, and experience influencing across multi-disciplinary engineering teams.
Helped define architecture strategy, improved simulation methodology, or mentored engineers into broader technical leadership roles.

Nice To Haves

Experience in parallel computing, datacenter architecture, large-scale interconnect architecture, or AI/HPC workload analysis.
Add value examples where you built or improved performance modeling frameworks, wrote and analyzed test cases at scale, or used AI tools for code development, validation, model generation, or architectural analysis.

Responsibilities

Define the strategy for next-generation GPU memory and NoC performance and functional models.
Guide how we use simulation, emulation, workload analysis, and AI-assisted model generation to evaluate architectural choices and predict system behavior.
Identify important bottlenecks and opportunities across GPU memory systems, on-chip interconnects, cache hierarchy, data movement, and workload execution.
Translate those insights into architecture recommendations, modeling priorities, and feature direction for future products.
Partner to establish modeling standards, validation approaches, common workloads, and decision frameworks that help teams compare architectural proposals with confidence.
Work across GPU architecture, memory, interconnect, software, verification, performance, and product teams to connect requirements with scalable technical solutions.
Communicate clear recommendations to senior technical and business leaders, including trade-offs, risks, assumptions, and expected performance impact.
Mentor senior engineers, strengthen modeling and architecture communities, and help influence the long-term direction of GPU architecture!