Production-Grade Simulation Infrastructure Engineer

Slope; San Francisco (CA); On-site; USD 170,000 - 210,000; Full time

Job summary

OpenAI in San Francisco, CA, seeks a Sim Infrastructure Engineer to turn simulation systems into reliable, automated pipelines for model training, evaluation, and HIL validation. You will own automation, orchestration, and tool integration across sim tasks, building scalable CI/CD and end-to-end evaluation.

In-person presence required four days/week. You will collaborate with Sim Realism, Sim Environments, Research, and Ops to make simulation a repeatable, measurable part of ML and robotics

Qualifications

• Experience building CI/CD at scale.
• Proficiency in Python, C++, and Rust.
• Experience with distributed compute and GPU workloads.
• Familiarity with monitoring, observability, and KPIs for complex pipelines.
• Ability to design APIs and developer-oriented tooling.

Responsibilities

• Build and maintain presubmit checks, CI/CD pipelines for simulation code, environments, and tasks; ensure artifacts are testable and reproducible.
• Implement end-to-end automation for model evaluation in sim (SIL) and orchestrate HIL runs; compute metrics and dashboards.
• Create robust APIs/connectors so research and data systems can schedule and evaluate simulations; support RL rollouts and data collection.
• Scale scheduling, batching, and orchestration for tens of thousands of rollouts; optimize cloud/GPU runtime reliability.
• Produce metrics/tools for health, throughput, fidelity regressions, and cost; implement tests to catch regressions early.
• Version artifacts, ensure environment immutability, and manage resource quotas and costs across the sim farm.

Skills

CI/CD at scale

Distributed compute

Automation & observability

API design

Python/C++/Rust

Collaboration across teams

Tools

Kubernetes

CI systems

Distributed task queues

Python

C++

Rust

Job description

OpenAI in San Francisco, CA, seeks a Sim Infrastructure Engineer to turn simulation systems into reliable, automated pipelines for model training, evaluation, and HIL validation. You will own automation, orchestration, and tool integration across sim tasks, building scalable CI/CD and end-to-end evaluation.

In-person presence required four days/week. You will collaborate with Sim Realism, Sim Environments, Research, and Ops to make simulation a repeatable, measurable part of ML and robotics