Senior Performance and Manufacturability Architect

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA +3 · Remote

This role focuses on architecting system and silicon features to optimize power, performance, thermal efficiency, and manufacturability for GPUs and SoCs. It involves co-designing chip, system, and package features to enhance testability, yield, and manufacturing processes. The role leverages AI tools to accelerate engineering tasks while maintaining rigorous engineering judgment and operates across the full lifecycle from system architecture to post-silicon success.

What you'd actually do

  1. Architect system and silicon features to enhance performance, power efficiency, and thermal behavior.
  2. Drive improvements in V/F curves, Vmin, TGP, speed grading, and thermal envelopes through co-design.
  3. Design chip, system, and package-aware features that enhance testability, coverage, and yield.
  4. Define manufacturing-aware methodologies linking test, SRAM behavior, binning, and package constraints to product performance.
  5. Co-design test strategies and screening methods to reduce overkill, test time, and miscorrelation.

Skills

Required

  • Master’s degree (or equivalent experience) in Electrical Engineering, Computer Engineering, Computer Science, Systems Engineering, or related field.
  • 8+ years of experience in system architecture, silicon performance, manufacturing co‑design, or post‑silicon validation.
  • Deep understanding of DVFS, binning, power/thermal management, and performance trade‑offs in advanced GPUs or SoCs.
  • Ability to reason across circuit behavior, system constraints, and manufacturing realities.
  • Comfort with hands‑on lab work as well as abstract architectural reasoning.
  • Strong scripting and analysis skills (e.g., Python, Perl) for automation, modeling, and data‑driven decision making.
  • Clear technical communication and the ability to document and defend engineering decisions.

Nice to have

  • Proven record of improving real product performance through system–manufacturing co‑optimization.
  • Evidence of fast abstraction, strong pattern recognition, and deep systems thinking.
  • Proof of work: architectures you’ve designed, performance problems you’ve untangled, or complex trade‑offs you’ve driven to closure.
  • Ability to operate independently on hard, ambiguous problems—and collaborate clearly across functions.
  • Thoughtful use of AI to increase engineering velocity without lowering the technical bar.

What the JD emphasized

  • AI tools to accelerate engineering work
  • rigorous engineering judgment
  • strong systems thinking
  • comfortable with ambiguity
  • AI as a force multiplier
  • AI tools to accelerate engineering work
  • strong judgment on when to trust, verify, or override outputs
  • Proven record of improving real product performance through system–manufacturing co‑optimization
  • Thoughtful use of AI to increase engineering velocity without lowering the technical bar