Senior AI Security Researcher

Carnegie Mellon University • Pittsburgh, PA

22h

About The Position

The CERT Division of Carnegie Mellon University's Software Engineering Institute (SEI) is seeking an AI Security Researcher to advance research in AI security. As part of the Threat Analysis Directorate, the researcher will work with security experts to assess AI-system vulnerabilities, evaluate defenses, reverse engineer AI systems and models, and identify areas where further security research is needed. The role involves collaboration with network defenders, software developers and vendors, security researchers, AI practitioners, policymakers, university faculty, and government representatives.

The researcher will develop approaches for analyzing AI-system robustness and apply them to understand vulnerabilities and attacker tradecraft. Work includes designing analysis methods and tools, evaluating security tools and practices, studying vulnerability disclosure, shaping AI-security best practices, and developing models, tools, and datasets to characterize threats. The researcher will publish findings and use results to support testing, evaluation, and transition of technologies from government-funded research programs, while also identifying ways AI can improve cybersecurity research. The role includes co-authoring proposals, conducting studies, and presenting findings to sponsors, decision-makers, and academic audiences.

The SEI is a nonprofit federally funded research and development center at Carnegie Mellon University. The position is based in Pittsburgh, Pennsylvania, and requires occasional travel, approximately 5%, to SEI and CMU locations, sponsor sites, conferences, and offsite meetings. Candidates must undergo a background check and obtain and maintain a Department of War security clearance.

Requirements

• Bachelor's degree in machine learning, cybersecurity, statistics, or a related discipline and 10 years of experience; or master's degree in those fields and 8 years of experience; or Ph.D. in those fields and 5 years of experience.
• Deep interest in AI/ML and cybersecurity, intellectual curiosity, and a desire to make an impact beyond one's organization.
• Practical experience applying cybersecurity knowledge to vulnerability research, analysis, disclosure, or mitigation.
• Experience advising on a range of security topics based on research and expert opinion.
• Familiarity with implementing and applying AI/ML techniques to practical problems.
• Familiarity with common AI/ML software packages and tools, such as NumPy, PyTorch, TensorFlow, or ART.
• Knowledge of or familiarity with reverse-engineering tools, such as NSA Ghidra or IDA Pro.
• Experience with Python, C/C++, or low-level programming.
• Experience developing frameworks, methodologies, or assessments to evaluate technology effectiveness and robustness.
• Excellent oral and written communication skills, particularly communicating technical topics to non-experts.
• Ability to travel occasionally, approximately 5%.
• Ability to pass a background check and obtain and maintain a Department of War security clearance.

Nice To Haves

• Candidates with strong technical proficiency in either AI/ML or cybersecurity are welcome, provided they demonstrate intellectual agility and commitment to accelerated learning in the role.

Responsibilities

• Develop approaches for analyzing the robustness of AI systems.
• Apply analysis approaches to understand AI-system vulnerabilities and how attackers adapt their tradecraft to exploit them.
• Reverse engineer malicious code in support of high-impact customers.
• Design and develop analysis methods and tools.
• Identify and address emerging and complex threats to AI systems and participate in the broader security community.
• Study and influence AI-security and vulnerability-disclosure ecosystems.
• Evaluate the effectiveness of tools, techniques, and processes developed by industry and the AI-security research community.
• Examine and shape assumptions underlying current AI-security best practices.
• Develop thought models, tools, and datasets to characterize threats to and vulnerabilities in AI systems, and publish results.
• Use research results to support testing, evaluation, and transition of technologies developed by government-funded research programs.
• Identify opportunities to apply AI to improve existing cybersecurity research.
• Co-author research proposals, execute studies, and present findings and recommendations to sponsors, government and industry decision-makers, and academic conferences.
• Mentor and cross-train others and share knowledge within the broader community.
• Travel occasionally to support the SEI's mission.

Benefits

• 8% monthly retirement contribution without requiring an employee contribution.
• Tuition benefits at Carnegie Mellon University and other institutions for employees and their dependent children.
• Flexible work arrangements.
• Paid parental leave.
• Paid military leave.
• Annual professional development opportunities, including conferences, training, or certification.
• Reimbursement for membership in professional societies.
• Relocation assistance for eligible candidates.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior