Researcher, Safety Training

The role

Researchers will develop methods to improve model behavior after pretraining. This includes working on RLHF, RLAIF, DPO, PPO, and safety evaluations to make models helpful and aligned.

What you will do

• Conduct research on safety post-training methods including RLHF, RLAIF, DPO, PPO, reward modeling, adversarial training, synthetic data, and scalable oversight
• Develop and implement model training interventions
• Build evaluations and measurement infrastructure for safety harms and reliability failures
• Translate ambiguous safety problems into research plans, datasets, and training runs
• Analyze model failures such as hallucination, unsafe tool use, and prompt injection
• Partner with domain experts to design safety training approaches
• Collaborate with research and deployment teams to incorporate safety improvements

Required qualifications

• Hands-on experience training, fine-tuning, or post-training language models, multimodal models, or agentic systems
• Experience with RL, RLHF, RLAIF, DPO, PPO, reward modeling, preference optimization, or instruction tuning
• Strong research judgment and strong engineering ability
• Comfort running experiments, analyzing model behavior, and working with large-scale ML systems
• Experience in one or more relevant domains: agents, tool use, multimodal models, cyber/security, reliability, uncertainty calibration, factuality, evals, privacy, youth safety, wellbeing, or applied a
• PhD in machine learning, computer science, or a related field, or equivalent practical research or engineering experience

Nice to have

• Experience improving production model behavior through post-training or deployment feedback
• Experience building safety evals, automated graders, red-teaming systems, or monitoring pipelines
• Experience working with human feedback, expert feedback, synthetic data, or policy-grounded training data
• Experience in safety-critical or abuse-sensitive domains

About the company

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity.