AI Safety Policy Evaluator, Violence & Threats at Handshake

Department: HAI Delivery Ops

Compensation

$55 – $120/hr

Setup: On-site
Location: Seattle, Washington
Type: Full-time
Level: not_specified
Posted: Sep 2, 2026

Description

Handshake AI is hiring an AI Safety Policy Evaluator focused on Violence & Threats to evaluate user requests, AI model responses, and conversation history involving violence, weapons, threats, and dark fiction. The role distinguishes fictional, educational, historical, defensive, and defensive violence from requests seeking real-world capability or intent to harm, assesses borderline cases, writes rationales, probes policy boundaries, identifies gaps, and participates in calibration. It is an ongoing full-time W-2 assignment with TCWGlobal supporting Handshake AI, onsite in Seattle, Monday through Friday, with a planned start date of September 21, 2026, and compensation of $55-$120 per hour.

Required Skills

Content moderation
Threat assessment
Calibration
Adversarial prompting
Rationale writing

Additional Information

Required Previous Experiences: Relevant depth in violent fiction or media, military or emergency response, crisis or threat assessment, trust and safety, content moderation, or closely related policy work
Preferred Qualifications: Published or produced work involving violence
Military, law enforcement, corrections, private security, or armed professional experience
Crisis intervention, threat assessment, forensic or clinical psychology, or violence prevention
Firearms, martial arts, or other weapons disciplines
AI evaluation, red teaming, data annotation, RLHF, trust and safety, or content moderation
Professional experience evaluating language model outputs
Calibration, inter-rater agreement, or adjudication workflows