Senior GPU Capacity Planner

Crusoe•Sunnyvale, CA

5d•$160,000 - $195,000

About The Position

At Crusoe, the Capacity Planning & Efficiency team operates at the critical intersection of all Crusoe products, services, and underlying physical infrastructure. As the Senior GPU Capacity and Optimization Planner, you will own the day-to-day management, allocation, and strategic bin-packing optimization of Crusoe Cloud’s high-performance GPU fleet. This is a highly cross-functional, high-impact role that bridges the gap between our technical infrastructure and our Go-To-Market (GTM) engine. You will work closely with Sales, Customer Success, Solutions Engineering, and Fleet Management to translate commercial customer pipelines into requirements within the datacenter. Your mission will be to maximize resource utilization, eliminate fragmentation across accelerated computing clusters, and ensure our AI/ML customers receive seamless performance, directly supporting Crusoe’s mission to align the future of computing with environmental sustainability.

Requirements

3+ years of direct experience in Infrastructure Capacity Planning, Technical Product Management, or Systems Engineering with heavy exposure to machine-level resource scaling.
Experience working within a hyperscaler cloud environment (e.g., AWS, GCP, Azure, Oracle Cloud) or a specialized, large-scale AI/accelerated compute cloud fabric.
Solid foundational understanding of GPU topologies (e.g., NVIDIA H100/B200 ecosystems).
Proven ability to communicate complex infrastructure and physical layout constraints clearly to business stakeholders like Sales and Customer Success, while translating commercial requirements back to hardware engineers.
Bachelor’s or Master’s degree in Computer Engineering, Computer Science, Operations Research, Industrial Engineering, Data Science, or an equivalent quantitative field.

Nice To Haves

Deep technical familiarity with distributed AI training frameworks and multi-tenant cloud storage dynamics.
A passion for green energy and aligning massive computing scales with environmental sustainability initiatives.

Responsibilities

Develop and execute advanced bin-packing strategies to optimize GPU resource allocations within our data centers, minimizing fragmentation and maximizing cluster utilization.
Act as the primary technical capacity partner to Sales, Customer Success, and Solutions Engineering, reviewing upcoming pipeline demands and mapping them to physical hardware constraints.
Collaborate tightly with Fleet Management, Infrastructure Engineering, and Data Center Operations to ensure maximum uptime of our customer workloads.
Design and implement models to track real-time GPU cluster headroom, workload densities, and allocation velocities to prevent supply bottlenecks.
Synthesize commercial demand signals and large-scale AI training/inference architectural trends to inform hardware placement and scheduling.
Partner with Core Software Engineering to translate manual allocation processes into scalable, automated programmatic scheduling and visualization tools.