RLHF Research Engineer – Multimodal Personalization

OpenAI; San Francisco (CA); Hybrid; USD 100,000 - 180,000; Full time

Job summary

A leading AI research organization is looking for a Research Engineer/Scientist to join their Future of Computing Research team in San Francisco. The role focuses on developing methodologies for personalized multimodal AI systems and includes responsibilities such as building reward models and conducting long-horizon evaluations. Candidates should have strong backgrounds in machine learning and be excited about designing experiments that align AI behaviors with user values. The position offers a hybrid work model and relocation assistance.

Qualifications

• Strong background in machine learning research, with experience in RLHF and reward modeling.
• Comfortable across the stack, from data generation to training runs.
• Excited by nuanced behavioral objectives and multimodal AI.

Responsibilities

• Develop RLHF and post-training methods for multimodal models.
• Prototype and iterate quickly on training recipes and evaluation suites.
• Collaborate closely with safety researchers for aligned systems.

Skills

Machine learning research

RLHF (Reinforcement Learning from Human Feedback)

Reward modeling

Preference optimization

Human-in-the-loop evaluation

Job description

A leading AI research organization is looking for a Research Engineer/Scientist to join their Future of Computing Research team in San Francisco. The role focuses on developing methodologies for personalized multimodal AI systems and includes responsibilities such as building reward models and conducting long-horizon evaluations. Candidates should have strong backgrounds in machine learning and be excited about designing experiments that align AI behaviors with user values. The position offers a hybrid work model and relocation assistance.