Policy Design Manager, Conventional Weapons
AnthropicGenerative AI, company
San Francisco, United States$245,000 - $285,000 USDManager
GIC
Coatue
Google
Lightspeed Venture Partners
ICONIQ Capital
Amazon
Legal & Compliance
About the role
TL;DR
Owns conventional weapons policy, threat models, and enforcement systems for AI safety.
- •Anthropic's Safeguards organization builds policies and systems to define and enforce limits on AI usage.
- •In this role, you will own our conventional weapons work, defining boundaries between acceptable and harmful requests in a domain where components are dual-use.
- •Key Responsibilities Own and maintain Anthropic's conventional weapons policy.
- •Build threat models and evaluations to measure AI contributions to weapons development.
- •Partner with engineering to implement policy into model guardrails and detection systems.
- •Serve as the subject-matter expert for escalations involving conventional weapons content.
- •Requirements Deep, applied expertise in weapons systems and ability to translate technical evidence into policy judgments.
- •Experience in a relevant setting (e.g., defense research, weapons system design).
- •Ability to write clear policy and explain complex technical topics to non-specialists.
- •Understanding of legal frameworks governing weapons and their transfer.
Required skills
OpenAI APIAnthropic APIPythonBash
Nice-to-have skills
LLMsNLPComputer Vision
Domain expertise
defenseai
Tech stack
PythonBashOpenAI APIAnthropic APIGitLinuxJiraConfluenceSlack