Research Scientist 3

MIT • Cambridge, MA

1d • Onsite

About The Position

The Computer Science and Artificial Intelligence Lab is seeking a Research Scientist 3 to lead a research program on Provable AI Safety through Self-Proving Models. This role will advance AI safety research, a strategic priority at MIT, by building both the theoretical foundations and the practical implementation of AI systems that prove the correctness of their own outputs. The scientist will serve as a key member of a research team spanning theoretical computer science, machine learning, and the use of machine learning for enhancing mathematical discovery and formal verification. Responsibilities include setting the project's research direction, leading several of the program's parallel research tracks, and supervising PhD students, as well as publishing their research.

Requirements

• Ph.D. in Computer Science, Mathematics, or a closely related field.
• A track record of independent research contributions in theoretical computer science, machine learning, and/or formal verification, including foundational work on the Self-Proving Models framework.
• Demonstrated expertise in interactive proof systems, computational complexity, and machine learning techniques applied to language models.
• Excellent leadership experience and project management skills.
• Strong supervisory skills and experience mentoring graduate students across theoretical and applied work.
• Experience leading or contributing to cross-institutional research collaborations.
• Outstanding interpersonal and relationship management skills to relate to a wide diversity of faculty, staff, students, and external stakeholders, including funders and policy audiences.
• Ability to handle confidential information and sponsored research compliance matters with discretion and judgment.
• Proven written and verbal communication and presentation skills, including the ability to publish in top-tier venues in theoretical computer science, machine learning, and AI safety, and to present research to both technical and policy audiences.

Responsibilities

• Lead a research program on Provable AI Safety through Self-Proving Models.
• Advance AI safety research by building theoretical foundations and practical implementation of AI systems that prove the correctness of their own outputs.
• Serve as a key member of a research team spanning theoretical computer science, machine learning, and the use of machine learning for enhancing mathematical discovery and formal verification.
• Set the project's research direction.
• Lead several of the program's parallel research tracks.
• Supervise PhD students.
• Publish research.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: Ph.D. or professional degree