Compute Platform Engineer, License Scheduling - EDA Infrastructure

NVIDIA•Us, NC

2d•Hybrid

About The Position

NVIDIA is a leader in computer graphics, PC gaming, and accelerated computing. The Compute Platform team in Santa Clara, CA is responsible for flawless license scheduling to enhance the productivity and efficiency of engineering teams. This role is critical in optimizing resource prioritization, significantly impacting the compute budget and project success.

Requirements

BS in Computer Science, Electrical Engineering, or equivalent experience.
6+ years in EDA compute, HPC, or a comparable environment with contended licensed resources.
Practical experience working directly with a commercial license manager, including troubleshooting checkout behavior.
Working knowledge of a batch scheduler and understanding of how license state and job state interact.
Data proficiency in Python and SQL, with the ability to turn checkout logs into compelling arguments.

Nice To Haves

Direct experience with a dedicated license scheduling layer, including hierarchical project share build.
Proven track record of working through license allocation during a tape-out and understanding the intricacies involved.
Familiarity with the structure and pricing of commercial EDA licensing models.
Experience in influencing license procurement with evidence-based utilization data.

Responsibilities

Managing license scheduling policy throughout the estate, including hierarchical project shares, shared and non-shared distribution policies, and preemption on license rather than slot.
Modeling license demand against entitlement to enable procurement to negotiate from evidence rather than last year's peak.
Tuning allocation policy through tape-out crunch, ensuring flawless execution even when contention is highest.
Building telemetry systems to make license utilization legible, including feature-level checkout patterns, queue-time attribution, and waste reduction.
Partnering with CAD and project teams to translate business priorities into share configurations that hold up under load.