NVIDIA
GPU System Performance Architect
$224K–$356K
US, CA, Santa Clara +2 · Full-time · Posted Sep 14, 2026
- Location
- US, CA, Santa Clara
US, CO, Boulder
US, OR, Hillsboro - Experience
- 10 years of relevant work or research experience.
- Education
- Master's or PhD in a relevant discipline such as Computer Science, Electrical Engineering, or Computer Engineering (or equivalent experience).
About the job
Design and analyze GPU-accelerated processor and system architectures for cloud computing, evaluating performance, cost, and efficiency and informing system design across hardware and software.
What you'll do
- Develop processor and system architectures for GPU-accelerated cloud computing.
- Build analytical models, simulators, and data visualization tools to assess performance, cost, and efficiency.
- Analyze hardware-software interactions in data center deployments and collaborate with software, product, and business teams.
What they ask for
- Excellent mathematical and analytical skills and deep knowledge of computer architecture.
- Experience with analytical performance modeling, simulation, profiling, and analysis.
- Strong communication, organizational, and interpersonal skills; ability to work as part of a team.
Nice to have
- Background in data center/cloud computing design, including hardware for virtualized environments.
- Expertise in data analysis and visualization; familiarity with GPU computing, parallel programming models, or PCIe and IO-centric workload optimization.
Pay
224,000 USD - 356,500 USD (base salary)
Tools
GPU, PCIe
AI focus
The role develops GPU-accelerated computing systems used for deep learning and AI; the posting also states that NVIDIA uses AI tools in its recruiting processes.
Model-written summary · the original posting is authoritative