Senior GPU Compute Scheduler – Scale AI Clusters

NVIDIA AI; Redmond (WA); On-site; USD 152,000 - 288,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA in Redmond is seeking a Scheduling Engineer to design and implement features for GPU compute clusters, enabling large multi-node AI workloads. You will focus on resource fairness, GPU occupancy, resilience, and performance across DL pipelines.

The role requires 5+ years in systems programming with C/C++, Go, Python, Bash, Linux, and experience with container tech (Docker/Singularity/Podman) and batch schedulers (SLURM, K8s). Equity and benefits are offered.

Qualifications

• Bachelor’s degree in CS, EE or related field or equivalent experience.
• 5+ years of work experience.
• Strong knowledge of batch scheduling (SLURM or K8s batch schedulers).
• Experience in C/C++, Go and scripting (Python, Bash).

Responsibilities

• Design and develop new scheduling features for GPU clusters.
• Develop batch workload management and orchestration services.
• Provide support to staff and end users for batch scheduler issues.
• Build and improve ecosystem around GPU-accelerated computing.
• Performance analysis and optimization of deep learning workflows.
• Develop large-scale automation solutions.
• Root cause analysis and corrective actions.
• Identify and fix problems before they occur.

Skills

C/C++

Go

Python

Bash

Linux

Docker/Singularity/Podman

Batch scheduling

Performance tuning

Communication

Education

Bachelor’s degree in Computer Science, Electrical Engineering or related field

Tools

SLURM

Kubernetes batch schedulers (Kueue, Volcano)

Job description

NVIDIA in Redmond is seeking a Scheduling Engineer to design and implement features for GPU compute clusters, enabling large multi-node AI workloads. You will focus on resource fairness, GPU occupancy, resilience, and performance across DL pipelines.

The role requires 5+ years in systems programming with C/C++, Go, Python, Bash, Linux, and experience with container tech (Docker/Singularity/Podman) and batch schedulers (SLURM, K8s). Equity and benefits are offered.