Member of Technical Staff, Interpretability

• Understand how data, activations, and weights affect model behavior, and use these insights to make AI more reliable
• Develop and scale interpretability techniques for large scale models
• Create infrastructure to support production and development workflows in interpretability

Work on interpreting and steering the behavior of frontier AI systems to ensure they are maximally beneficial for society.

• Strong software engineering skills
• Experience working with LLMs and some experience in interpretability
• Creative and enjoys ideas that are both scalable and useful
• Cares about both short-term impact and long-term insight
• Team player willing to pick up new skills to deliver results

Base salary plus equity, comprehensive medical, vision, and dental coverage, access to a 401(k) retirement plan, short & long-term disability insurance, life insurance, and various other discounts and perks.