Handshake
Strategic Projects Lead, Safety
$160K
New York, NY · On-site · Full-time · Posted Sep 16, 2026
- Experience
- 2+ years; the posting lists related technical/analytical fields as alternatives.
About the job
Leads large-scale AI safety and red-teaming programs, using adversarial testing to identify model vulnerabilities and help frontier labs strengthen policies and evaluations.
What you'll do
- Own and deliver safety evaluations and red-teaming programs across harm categories, labs, and policy domains.
- Design adversarial testing methods, including jailbreak testing, and extend evaluations to agentic AI systems.
- Coordinate expert testing teams, partner with lab policy leads, and report findings and recommendations.
What they ask for
- 2+ years of experience in trust & safety, red-teaming, security research, policy enforcement, or a related technical/analytical field.
- Strong project and workforce management skills, including leading teams against tight timelines and changing scope.
- Working familiarity with adversarial testing concepts such as jailbreak techniques or prompt-based exploits, or ability to learn quickly.
- Strong analytical problem-solving, communication, and stakeholder management skills.
Nice to have
- Experience red-teaming, jailbreaking, or adversarially evaluating LLMs or agentic AI systems.
- Expertise in high-severity policy harms, testing or evaluation programs, security research, offensive security, or AI safety research.
Pay
$160K (base salary; compensation type not specified in posting)
Tools
LLMs
AI focus
Adversarially tests frontier AI models, including in agentic settings, to identify safety risks and improve policies and evaluation coverage.
Model-written summary · the original posting is authoritative