RLHFPythonPyTorchModel AlignmentLLMsAI SafetySuperalignment

Role Briefing: Research and engineer scalable alignment, rule-based rewards, and automated self-critique techniques to make frontier multimodal models beneficial, honest, and robust against jailbreaks.

About OpenAI

OpenAI's mission is to ensure that artificial general intelligence (AGI) benefits all of humanity. The Alignment & Safety research team invents methodologies to steer, align, and evaluate frontier foundation models.

What You'll Do

Engineer scalable reinforcement learning from human feedback (RLHF) and reinforcement learning from AI feedback (RLAIF) training pipelines.

Develop automated synthetic evaluation environments to probe models for deceptive alignment, hallucination tendencies, and dangerous capabilities.

← Return to All Tech Positions