Research Scientist - CoT, Science of Deep Learning

OpenAI is seeking a Research Scientist for their Chain-of-thought Interpretability Team, part of the Science of Deep Learning team. This role focuses on developing innovative machine learning techniques to monitor models' latent thinking, particularly in chain-of-thought (CoT) reasoning. The ideal candidate will work on technical approaches to determine when model CoTs are monitorable, faithful, and legible, and explore interventions to improve or degrade monitorability.

Key responsibilities include:

• Developing innovative machine learning techniques
• Collaborating with peers across the organization
• Advancing a critical pillar of OpenAI's mission
• Discovering simple, generalizable ideas that work well at large scale
• Contributing to a broader research vision that unifies the entire company
• Writing bug-free ML code and working with complex codebases

The ideal candidate should:

• Be excited about OpenAI's mission and eager to contribute to building safe, beneficial AGI
• Have a background in statistical machine learning, physics, mathematics, or another theoretically and empirically rigorous field
• Be passionate about building, running, and studying AI systems at the largest scales
• Enjoy fast-paced and collaborative research environments

OpenAI offers a competitive compensation package, including a salary range of $295K – $440K plus equity. They are committed to diversity, equal opportunity, and providing reasonable accommodations to applicants with disabilities.

Join OpenAI in shaping the future of technology and ensuring that the benefits of AI are widely shared.