Senior Manager, Silicon Speed Productization — Silicon Co-Design Group

NVIDIA•Us, CA

8d•Hybrid

About The Position

NVIDIA’s Silicon Co-Design Group (SCG) sees every silicon program from first power-on to high-volume production. We sit at the intersection of architecture, design, marketing, operations, and productization, turning silicon behavior into products that deliver industry-defining performance, power efficiency, yield, and quality. We are hiring a Senior Manager to lead our Silicon Speed Productization team. In this role you will own simulation-to-silicon speed correlation and determine whether each product delivers its intended frequency across voltage, process, temperature, workload, and lifetime conditions. This is not a coordination role or a validation-only role. You and your team will identify where predictions diverge from measured silicon, drive the hardest issues to root-cause closure, and establish the technical basis for product limits, guardbands, binning, and production sign-off! Your organization will lead senior individual contributors and a first-line manager at the boundary between what was designed and what was built. Architecture, design, product, and senior leaders will use your data and judgment to make program decisions. Success in this role means that speed risks are raised early, correlation improves with every generation, and programs ship at speed-of-light schedules. The exceptional leader also applies AI deliberately—with demonstrated workflow impact!

Requirements

BS, MS, or PhD in Electrical Engineering, Computer Engineering, Systems Engineering (or equivalent experience)
12+ overall years of experience in silicon characterization, post-silicon validation, timing or performance analysis, including 5+ years leading technical teams
Proven management experience developing senior engineers and first-line managers
Deep technical foundation in digital design, computer architecture, static timing analysis, critical-path behavior, clocking, power and timing interactions, semiconductor process variation, sampling, and statistics.
Hands-on expertise with silicon speed characterization, simulation-to-silicon correlation, voltage-frequency margining, guardbanding, and process-voltage-temperature and binning dependencies for high-performance silicon
Demonstrated ability to drive ambiguous, system-level issues to root-cause closure across the company
Proficiency with data-analysis and automation tools such as Python, JMP, SQL, or equivalent, with the judgment to distinguish correlation from causation in large silicon datasets
Strong written and verbal communication; able to translate complex technical issues into decision-ready options for leadership

Nice To Haves

Evidence that you traced a frequency miss to a specific critical path, microarchitectural interaction, clocking condition, or process corner and drove a productized fix through confirmation.
Experience building reusable correlation, margining, or characterization methodologies adopted across silicon programs or advanced process nodes.
Applied AI tools to characterization or debug workflows to lift team velocity
Comfort navigating ambiguity, setting technical direction at speed-of-light schedules, and aligning global teams around a clear plan of record.

Responsibilities

Close the prediction-to-silicon loop: Own speed characterization and correlation from pre-silicon planning and first power-on through production sign-off across GPU, CPU, and SoC programs
Resolve the hardest speed failures: Drive root-cause closure when a frequency corner does not hold, minimum operating voltage is too high, a critical path escapes timing analysis, or workload behavior diverges from the model
Define product limits with evidence: Refine methodologies for silicon margining, process-voltage-temperature characterization, guardbanding, aging, and speed binning, and translate the results into product requirements and production decisions
Left-shift learning: Feed measured-silicon insights back into timing models, critical-path predictions, test content, design practices, and future architectures so that each program begins with a stronger correlation baseline
Give leadership the clarity to act: Convert characterization data, model deltas, yield trends, and program risks into crisp, decision-ready options for leadership on plan-of-record, qualification, production, and ramp decisions
Scale analysis and coverage: Build closed-loop data, automation, and AI-assisted workflows for test orchestration, anomaly detection, triage, and reporting that improve coverage and cycle time
Build the organization NVIDIA depends on: Hire, develop, and retain a high-performing, globally-distributed team; grow engineers into recognized subsystem experts and develop first-line managers into independent leaders