The opportunity
Safety Research at OpenAI is responsible for ensuring our most capable models can be safely deployed in the real world. The Safety Training team develops the research and engineering methods that shape model behavior through post-training, reinforcement learning, evaluation, and deployment feedback.
What you'll 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 that make models safer,: more reliable, and more aligned with user intent
Build evaluations and measurement infrastructure for safety harms,: high-stakes reliability failures, policy adherence, jailbreak robustness, multimodal risks, agent behavior, and domain-specific risks
Translate ambiguous safety problems into concrete research plans, datasets,: training runs, evals, and product-facing improvements
Analyze model failures such as hallucination, unsafe tool use, prompt: injection, privacy leakage, over-refusal, under-refusal, sycophancy, and deceptive or misleading behavior
Partner with domain experts to design safety training approaches for areas: like cyber, privacy, youth safety, wellbeing, and other frontier-risk domains
What they're looking for
- Collaborate with research and deployment teams to ensure safety improvements: are incorporated into OpenAI’s core model training and product launches
- Have hands-on experience training, fine-tuning, or post-training language: models, multimodal models, or agentic systems
- Have experience with RL, RLHF, RLAIF, DPO, PPO, reward modeling, preference: optimization, instruction tuning, or related post-training methods
- Care deeply about shaping model behavior in real-world systems, not only publishing research
