Expert AI Safety Data Annotator – National Security & Policy Background | Remote (US Time Zones Preferred)
Why This Role Matters
At Taskify, we believe AI safety fundamentally depends on high-quality human data. Advanced AI systems cannot self-evaluate—they need skilled humans who can apply structured judgment to complex, nuanced scenarios. We’re building a flexible team of Safety Specialists, including technical and non-technical experts, to annotate and evaluate AI behaviors and ensure system safety
What You’ll Do
- Produce precise, high-quality human data by annotating AI outputs on safety criteria such as bias, misinformation, disallowed content, and unsafe reasoning
- Apply harm taxonomies and guidelines consistently, even for ambiguous cases
- Document your reasoning clearly to improve annotation guidelines over time
- Collaborate with AI researchers and engineers to power safety research, model refinement, and risk audit
Who are You
- Experienced in national security, defense operations, government policy, or high-stakes communications.
- Skilled in crisis management, policy briefings, congressional or executive communications, or managing large-scale operations.
- Expert at structured risk framing: identifying vulnerabilities early and communicating them clearly to decision-makers.
- Comfortable applying this mindset to emerging AI system risk
- Past roles might include Intelligence Officer/Analyst, Policy Advisor, Strategic Communications Specialist, Crisis Operations Manager, National Security Analyst, Defense Policy Analyst, Government Relations Officer, Operations Planner, or Legislative Analyst.
Success in This Role Looks
- Delivering accurate, consistent annotations even in ambiguous situations
- Surfacing risks early that automated tools overlook.
- Providing feedback that improves guidelines and taxonomies.
- Producing data that directly strengthens AI model safety and compliance
Why Join Taskify?
- Apply your expert experience to a rapidly growing, high-impact AI safety.
- Work at the frontier of AI safety, where your risk assessment skills are directly relevant
- Join a mission-driven team dedicated to making AI systems safer, trustworthy, and aligned with human values.
About Taskify
Taskify connects top experts with leading AI labs and organizations pioneering the future.
Additional Details
- Fully remote; US time zones preferred.
- Open to candidates in the US, UK and CA
- Full-time or part-time flexible engagement.
- Compensation varies by project scope, expertise required, and content sensitivity; processed via Stripe Connect.
- Taskify is an equal opportunity employer providing reasonable accommodations.
