Subject Matter Expert - Election Integrity, Violent Extremism, and Fraud
OUR MISSION
Reflection is a research lab making intelligence open and accessible for everyone to use, customize, and build on. We build open models that let anyone control their intelligence and help shape the future of AI. Our mission: make intelligence open and accessible to all.
ROLE OVERVIEW
As a Subject Matter Expert, Election Integrity, Violent Extremism, & Fraud, you are our internal authority on preventing, detecting, and responding to election-related manipulation (e.g., coordinated influence operations, voter suppression content, fabricated or misleading claims about voting), terrorist and violent extremist content (TVEC), and fraud and deception (e.g., scams, phishing, impersonation, financial fraud, and other deceptive schemes) generated or facilitated through Reflection's models and products. You will define how the lab thinks about these harms, translate that expertise into concrete enforcement workflows and detection strategies, and act as a critical gatekeeper ensuring our open releases meet our commitments in these areas.
You will own election integrity, violent extremism, and fraud evaluations, human grading, auto-grading, and enforcement end to end: standing up and scaling review workflows, serving as the primary point of contact for those who conduct content review, as well as conducting them yourself, driving model launch evals on novel or ambiguous cases, and partnering closely with our technical Safety team, Alignment, Policy, and Legal teams to close gaps as the threat landscape evolves. You will also represent Reflection's work in these areas externally with reporting bodies, NGOs, researchers, election officials, financial-sector and anti-fraud partners, and, where appropriate, law enforcement.
In this position you will regularly be exposed to and engage with content that may be violent, graphic, hateful, or psychologically disturbing, including terrorist and violent extremist material and depictions of real-world violence. Yo...