295K–295K USD / year
Tabletop Exercise Lead, The Anthropic Institute
OtherHybrid — San Francisco, CA
Published on 2026-10-17
These details were extracted automatically from the original listing. They may not be complete or fully up to date — it's worth checking the original job post.
About this role
Build and lead Anthropic Institute’s tabletop exercise program, using simulations to study how people and institutions might respond to consequential AI-related scenarios. Set the research agenda, standardize methods and measurement across exercises, and publish findings for policymakers, researchers, and other decision-makers.
About the company
Anthropic builds reliable and interpretable AI systems, including the Claude family of models, for businesses and developers. The company focuses on AI safety research and practical AI assistants for enterprise and consumer use.
The team
The Anthropic Institute (TAI), which includes the Frontier Red Team, Societal Impacts, Economic Research, and the Rule of Law team.
Stack
ClaudeAI agentsLanguage models
What you'll do
- Set the research agenda, scenarios, and audiences for exercises
- Design exercises as empirical studies and standardize data capture across runs
- Develop measurement methods, including assessments, decision logs, transcript analysis, and after-action reports
- Publish findings and brief senior stakeholders inside and outside Anthropic
- Hire, manage, and develop a team and facilitator bench
- Build a reusable scenario library and facilitator kit
- Direct tools for exercises, including Claude-driven simulations
What we're looking for
- Experience designing and running tabletop exercises, wargames, or simulations for senior decision-makers, with published work on the subject
- Training in a discipline involving simulation, experimental methods, or structured analysis
- 10+ years of experience in national security, government, technology policy, or strategy
- Strong grasp of AI capabilities and safety research
- Experience managing and developing a team
- Extensive use of Claude and other AI tools
- Commitment to AI safety and Anthropic’s public benefit mission
Nice to have
- Published research on wargaming, simulation, or crisis decision-making
- Exercises that informed policy or crisis-response decisions
- Experience running multi-iteration or experimental wargames and analyzing results across runs
- Experience using AI agents or language models in simulations
- Experience leading a growing research program or team
- Trusted relationships across national security, government, policy, and AI communities
- Widely read publications on AI, emerging technology, strategy, or crisis decision-making
Benefits
- Flexible working hours
- Generous vacation and parental leave
- Optional equity donation matching
- Office collaboration space