Senior GPU Architect - Performance and Yield Optimization
NVIDIAJob Title
Senior GPU Architect - Performance and Yield Optimization
Role Summary
Design and optimize GPU architectures to improve manufacturing yield while preserving performance and architectural simplicity. The role partners across architecture, design, silicon, operations, and software teams to evaluate tradeoffs and move selected proposals into production.
Experience Level
Senior - typically requires 10+ years of experience in GPU, CPU, SoC, or complex processor architecture.
Responsibilities
Deliver architectural approaches and tooling that improve product yield, performance, and implementation efficiency.
- Analyze how manufacturing defects affect architectural resources and define methods to isolate or disable affected regions while preserving functionality.
- Develop software tools and models that capture architectural, performance, and product requirements as rules and constraints.
- Create optimization techniques to explore large configuration spaces and identify viable product configurations.
- Design approaches for multi-die architectures to improve silicon utilization while meeting product and performance goals.
- Evaluate yield, performance, area, cost, complexity, and product flexibility; collaborate across teams to implement chosen solutions.
Requirements
Core technical skills and proven experience required to execute the responsibilities.
- 10+ years of experience in GPU, CPU, SoC, or complex processor architecture.
- Strong understanding of GPU architecture, execution pipeline, and interactions across major GPU subsystems.
- Solid computer-architecture fundamentals and ability to reason about disabling, isolating, or reconfiguring resources.
- Proficiency in C++ and/or Python and experience building architectural models, simulators, optimization frameworks, or analysis tools.
- Experience translating architecture and product requirements into rules, constraints, algorithms, and executable analysis for complex optimization or configuration problems.
- Ability to quantify trade-offs among performance, yield, area, cost, and complexity and influence cross-functional decisions.
Nice-to-have:
- Hands-on GPU or large-scale SoC architecture experience.
- Experience with constraint-based, combinatorial, or mathematical optimization techniques.
- Background in silicon yield, defect tolerance, harvesting, redundancy, repair, or configurable processor architectures.
- Experience with multi-die/chiplet architectures and using silicon or manufacturing data to drive design choices.
Education Requirements
BS, MS, or PhD in Computer Engineering, Computer Science, Electrical Engineering, or a related field, or equivalent practical experience.
About the Company
Company: NVIDIA
Headquarters: Santa Clara, California, USA
NVIDIA is a global leader in accelerated computing, renowned for its innovative solutions in AI and digital twins that transform diverse industries. The company specializes in networking technologies, providing end-to-end InfiniBand and Ethernet solutions for servers and storage that optimize performance and scalability. NVIDIA serves sectors such as high-performance computing, enterprise data centers, and cloud computing, constantly reinventing its products and services to stay ahead in the market.
