Research Engineer, AI Risk Modeling
A 501(c)(3) nonprofit think tank that convenes technology, policy, and industry leaders to develop actionable responses to emerging security challenges, including AI risk, cybersecurity, and strategic stability.
Funding history
Investors
About Institute for Security and Technology
Institute for Security and Technology (IST) is an active nonprofit policy and research organization focused on technology-driven security challenges. Its current work includes AI risk reduction, AI and nuclear command/control/communications, cybersecurity, ransomware, critical-infrastructure resilience, and nuclear crisis communications.
Skills
Candidate Availability
About the Role
You will design statistical models, measurement frameworks and evaluation protocols that characterize risk dynamics and capability thresholds across AI systems. You will develop research methodologies using causal inference, Bayesian methods and longitudinal analysis, build data collection pipelines, document findings for technical and policy audiences, and work with lab safety teams and external reviewers to refine measurement approaches.
Requirements
- Advanced degree (MS/PhD) in Mathematics, Statistics, Quantitative Methods, Computer Science, or Machine Learning, or equivalent experience
- 2–4 years of experience in research design, statistical modeling, or quantitative analysis
- Strong proficiency in at least one statistical programming language (e.g., Python, R, Julia) and probabilistic modeling frameworks
- Experience with Bayesian methods, time-series analysis, and dynamic modeling
- Ability to design rigorous experiments or observational studies with clear causal identification strategies
- Ability to work independently and communicate technical concepts to mixed audiences
- Legally able to work in the United States without sponsorship
- Successful completion of a background check
Responsibilities
- Contribute quantitative expertise to design research methodologies for AI risk modeling projects, including causal inference, Bayesian methods, and longitudinal analysis
- Develop and validate statistical models that characterize risk dynamics, intervention effectiveness, and capability thresholds
- Design and implement technical measurement frameworks and data collection pipelines
- Support development of evaluation protocols and testing methodologies that generalize across models, tasks, and organizations
- Assist in translating technical findings into modular frameworks for lab piloting, policy reference, and adoption
- Document methodology and findings clearly for technical and policy audiences
- Collaborate with the AI Center team on AI security and safety projects
- Collaborate with lab safety teams and external reviewers to refine measurement approaches and align evaluation protocols
Benefits
- Comprehensive health, dental, and vision insurance for you and all your dependents
- 401(k) plan with up to 4% matching
- 12 weeks paid parental leave
- Unlimited PTO after the introductory period
- Potential stipends for education, home office, commuting, and wellness
- Potential for partial relocation support for those moving to the Bay Area
- Potential for bonuses
