AI Systems Engineer - Scale Training & Inference

United States Digital Space LLC; United States; Remote; USD 140,000 - 190,000; Full time

Job summary

United States Digital Space LLC is seeking an AI Systems Engineer to scale infrastructure behind our training and evaluation workflows. You will own projects from bottleneck identification to deployment, working with researchers and engineers to shorten the path from experiments to validated models.

You will build and operate distributed systems, improve resource management, and create self-service tools for researchers to run experiments reliably and efficiently.

Qualifications

• Strong software engineering fundamentals.
• Experience building or operating large-scale distributed systems.
• Experience in ML infrastructure, inference systems, GPU performance, or similar tooling.
• Highly self-motivated with ownership of open-ended problems.
• Ability to debug across system boundaries using measurements.

Responsibilities

• Build and operate infrastructure for large-scale training and evaluation, improving reliability and efficiency.
• Develop shared inference and grading platforms with automated capacity management and monitoring.
• Improve compute scheduling and resource allocation to reduce idle GPU time.
• Diagnose bottlenecks across training, inference, and orchestration; work across teams on end-to-end improvements.
• Build self-service tools and observability to help researchers launch experiments with less manual toil.

Skills

Software fundamentals

Distributed systems

ML infra experience

Ownership of open-ended problems

Performance measurements

Job description

United States Digital Space LLC is seeking an AI Systems Engineer to scale infrastructure behind our training and evaluation workflows. You will own projects from bottleneck identification to deployment, working with researchers and engineers to shorten the path from experiments to validated models.

You will build and operate distributed systems, improve resource management, and create self-service tools for researchers to run experiments reliably and efficiently.