Data Scientist (AI Preparedness)

Quick Facts

• Role: Data Scientist (AI Preparedness)

Description

Help build, evaluate, and continuously improve mitigations that prevent extreme harms from AI systems. The role goes beyond “running evals”: you’ll create mitigation intelligence and monitoring systems to detect issues early, measure effectiveness over time, and reduce both over-blocking (unnecessary friction) and under-blocking (missed harm).

Responsibilities

• Evaluate and improve mitigation systems, including classifiers and detection pipelines across domains (e.g., biosecurity, cybersecurity, emerging risk areas)
• Diagnose false positives and false negatives using deep error analysis, root cause investigation, and mitigation recommendations
• Build monitoring and measurement frameworks to track mitigation effectiveness over time and across user segments and use cases
• Identify trends in over-blocking vs. under-blocking, quantify customer impact, and propose prioritized interventions
• Develop insights from customer feedback, complaints, and usage patterns to detect shifts in adversarial behavior and failure modes
• Expand risk monitoring into new areas (e.g., cybersecurity threats, model loss-of-control or sabotage scenarios) with domain experts
• Communicate results to technical and executive stakeholders with crisp narratives, decision-ready metrics, and clear tradeoffs

Requirements

• Significant experience in data science or applied analytics in high-stakes domains (security, trust & safety, abuse prevention, fraud, platform integrity, reliability)
• Strong foundations in experimentation, causal thinking, and/or observational inference; ability to design robust measurement under imperfect data
• Fluency in SQL and Python (or equivalent) for analysis, modeling, and building monitoring workflows
• Experience building metrics, dashboards, and operational monitoring that meaningfully changes outcomes
• Track record driving cross-functional impact with engineering, product, and research partners
• Experience with classifier evaluation, calibration, thresholding, and error analysis at scale
• Familiarity with detection systems in adversarial settings (evasion, distribution shift, feedback loops)
• Genuine interest in AI safety, alignment, and catastrophic risk prevention
• Cybersecurity data science experience is strongly preferred (e.g., threat modeling, adversarial dynamics, abuse patterns, security telemetry)
• Trust & Safety experience is helpful, but not required

Benefits

Build mitigation intelligence and monitoring systems with broad importance by helping prevent extreme harms from frontier AI systems.