AI Security Researcher — Remote & Hybrid Opportunities

RAND; Oregon; Hybrid; USD 132,000 - 339,000; Full time

Job summary

RAND’s Center on AI, Security, and Technology (CAST) is seeking researchers to advance the technical foundations of AI safety. This three-year appointment invites candidates with deep expertise in CS theory, cryptography, formal methods, programming languages, systems security, or ML systems to conduct independent, original research.

Researchers will lead projects from problem formulation through publication, communicating results to both technical and policy audiences.

Qualifications

• Strong track record of independent, original research.
• Deep expertise in at least one area such as theoretical computer science, cryptography, formal methods, programming languages, computer security, systems, computer architecture, or machine learning systems.
• Strong programming or systems-building ability in Python, Rust, Go, or C/C++.

Responsibilities

• Conduct original research at the intersection of AI and security.
• Lead complex research projects from conception through completion.
• Communicate results to technical and non-technical audiences; publish in peer-reviewed venues.
• Collaborate with multi-disciplinary teams across RAND and external partners.

Skills

Independent research leadership

Deep expertise in AI/CS/ML

Programming in Python/Rust/Go/C++

Effective communication

Collaborative work in distributedteams

Education

PhD in CS/EE/Math/Physics or related

MS in relevant field with 3+ years experience

BS in relevant field with 5 years experience

Tools

Python

Rust

Go

C/C++

Job description

RAND’s Center on AI, Security, and Technology (CAST) is seeking researchers to advance the technical foundations of AI safety. This three-year appointment invites candidates with deep expertise in CS theory, cryptography, formal methods, programming languages, systems security, or ML systems to conduct independent, original research.

Researchers will lead projects from problem formulation through publication, communicating results to both technical and policy audiences.