This is a fully remote position, open to applicants in United States.

📋 Description

• Create, instrument, and manage the simulated environments in which agents learn.

• Establish the foundational elements, reward frameworks, interfaces, and telemetry for evaluating new capabilities under stress.

• Develop environments and set standards for data quality and reproducibility across extensive agent gyms.

⛳️ Requirements

• Proven experience in constructing RL environments or simulators (such as custom physics, multi-agent systems, and tool APIs) with a focus on determinism, performance, and observability.

• Strong intuition for crafting reward functions, scenario classifications, and QA processes that ensure signals remain aligned and free of drift.

• Ability to collaborate effectively across research and engineering teams to deliver practical, testable environments that adapt alongside model capabilities.