AI Systems Engineer - Scalable Training Infra

OpenAI; San Francisco (CA); On-site; USD 160,000 - 210,000; Full time

Job summary

OpenAI is seeking an AI Systems Engineer to scale infrastructure behind training and evaluation workflows. You will own projects from bottleneck analysis to deployment, combining distributed systems engineering with performance optimization and close collaboration with researchers.

You might build shared grading services, improve resource allocation, or bring a new training stack into production—directly accelerating research progress and reliability across OpenAI's pipelines.

Qualifications

• Strong software engineering fundamentals and experience building or operating large-scale distributed systems.
• Experience in ML infrastructure, inference systems, GPU performance, or infrastructure tooling.
• Highly self-motivated and comfortable taking ownership of open-ended problems.
• Enjoy debugging across system boundaries and using measurements to guide improvements in performance and reliability.

Responsibilities

• Build and operate infrastructure for large-scale training and evaluation, improving reliability, throughput, and resource efficiency.
• Develop shared inference and grading platforms with automated capacity management, health monitoring, and visibility into performance.
• Improve compute scheduling and resource allocation to reduce idle GPU time and help workloads recover quickly from failures.
• Diagnose bottlenecks across training, inference, and orchestration, and work across teams to improve end-to-end performance.
• Build self-service tools, automated validation, and observability to help researchers launch experiments and compare results with less manual intervention.

Skills

Distributed systems

Software engineering

ML infrastructure

Performance optimization

Ownership of open-ended problems

Job description

OpenAI is seeking an AI Systems Engineer to scale infrastructure behind training and evaluation workflows. You will own projects from bottleneck analysis to deployment, combining distributed systems engineering with performance optimization and close collaboration with researchers.

You might build shared grading services, improve resource allocation, or bring a new training stack into production—directly accelerating research progress and reliability across OpenAI's pipelines.