Post-silicon Validation and Methodology Engineer

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA

This role focuses on post-silicon validation and methodology for next-generation silicon, including GPUs and SoCs. It involves owning bring-up, validation, qualification, tuning, and productization plans. A key aspect is building and deploying AI-assisted lab workflows to automate telemetry, evaluate measurement data, and provide anomaly detection for faster root-cause analysis. The role requires strong programming skills (Python, C/C++) and proficiency in using AI tools for validation tasks.

What you'd actually do

  1. Own bring-up, validation, qualification, tuning, and productization plans for next-generation silicon — from first power-on through PVT sign-off.
  2. Partner across architecture, build, firmware, and software teams to define requirements for power management and clocking features, then drive coverage from pre-silicon through production.
  3. Build and deploy AI assisted lab workflows. These workflows automate bring-up telemetry and silicon measurement data evaluation. They provide anomaly detection on regression results and debug-triage tooling. This tooling reduces the time from observation to root-cause hypothesis. The team runs these tools during every bring-up, not as occasional scripts.
  4. Build the test infrastructure, characterization methodologies, and bring-up playbooks that shift post-silicon coverage left and raise velocity across programs.
  5. Lead root-cause analysis on the hardest HW/SW interaction issues — with the measurement field, instrumentation judgment, and hypothesis rigor to close them under schedule pressure.

Skills

Required

  • BS or MS in Electrical or Computer Engineering (or equivalent experience), plus 5+ years in silicon bring-up, validation, debug, or productization.
  • Deep fundamentals across digital development, microarchitecture, timing, clocking, power, noise, and control systems — and the ability to reason across the HW/SW boundary under real lab constraints.
  • Hands-on lab proficiency: oscilloscopes, logic analyzers, power analyzers, and the instinct to know which instrument answers which question during a bring-up.
  • Strong programming and scripting proficiency: Python, C/C++; experience building lab automation or test infrastructure that other specialists adopt and depend on — not just personal scripts.
  • Proficiency in the use of AI tools to accelerate silicon validation work — automated analysis of bring-up logs, regression data, or lab instrumentation output; anomaly detection on silicon measurement datasets; or LLM-assisted debug triage. We want to understand what you built, how you validated its trustworthiness, and where you decided the measurement had to stay manual.

Nice to have

  • Hardware proof of crafting: debug infrastructure you built in the lab, characterization methodologies adopted across programs, build DFT feature specs or in-system test suites you developed, or margin test flows that caught issues before production.
  • Proficiency in bring-up experience with GPU/SoC architecture
  • Experience crafting or scaling in-system test and DFT features for production silicon, with familiarity with fault models, DPPM, and RAS.
  • Build concrete examples of redesigning how a team debugs in the lab — faster triage, smarter hypothesis trees, automated measurement reporting — and the resulting increase in bring-up velocity or quality.

What the JD emphasized

  • AI assisted lab workflows
  • AI tools to accelerate silicon validation work
  • automated analysis of bring-up logs
  • regression data
  • lab instrumentation output
  • anomaly detection on silicon measurement datasets
  • LLM-assisted debug triage