SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Reflection AI is seeking a Subject Matter Expert to serve as the internal authority on election integrity, violent extremism, and fraud prevention within its AI research lab. In this role, you will define how the organization approaches these critical safety domains, translate expertise into concrete enforcement workflows and detection strategies, and act as a gatekeeper for model releases.
You will own election integrity, violent extremism, and fraud evaluations end-to-end: standing up and scaling review workflows, serving as the primary point of contact for content reviewers, conducting reviews yourself, driving model launch evaluations on novel cases, and partnering with Safety, Alignment, Policy, and Legal teams. You will also represent Reflection's work externally with reporting bodies, NGOs, researchers, election officials, financial-sector partners, and law enforcement.
Key responsibilities include: establishing standards, definitions, and escalation thresholds for these harm categories; partnering with Safety research and evaluation teams to create automated evaluations; managing day-to-day content evaluation workflows including task routing, queue management, and SLA monitoring; onboarding and training review partners; reviewing flagged content to drive enforcement decisions; partnering with engineering on detection models and abuse-prevention systems; validating models meet risk thresholds before release; developing internal documentation and decision trees; identifying emerging misuse patterns; coordinating external reporting obligations; and staying current on evolving legal frameworks and threat-detection technology.
The role requires deep expertise in election integrity, disinformation/influence operations, counterterrorism/violent extremism, or fraud and deception. You should have direct experience in trust & safety, content moderation operations, or policy enforcement; experience managing content review operations at scale; proficiency in SQL and data analysis; strong cross-functional communication skills; sound judgment under pressure; and familiarity with content classification frameworks, influence-operation taxonomies, fraud typologies, and relevant legal frameworks (EU Digital Services Act, EU TCO Regulation, election transparency laws, consumer protection regulations).
Note: This role involves regular exposure to violent, graphic, hateful, and psychologically disturbing content, including terrorist material and depictions of real-world violence, as well as deceptive and fraudulent material. Reflection prioritizes team member wellbeing in this area.
