Applied Scientist II, Identity Security & Abuse Prevention
Core
Design, build, and own production ML systems to detect abuse patterns, classify threats, and automate enforcement across Amazon's identity and authentication landscape.
Role type
Senior Applied Scientist (Identity Security & Abuse Prevention)
Builds
Production ML systems for abuse detection, anomaly detection, threat classification, and automated enforcement
Domain
Cybersecurity, Identity Security, Fraud Prevention
Deliverable
production ML models
Required skills
Machine learning system design, anomaly detection, graph neural networks, temporal modeling, causal inference, statistical validation, A/B testing, Java, C++, Python, algorithms and data structures, numerical optimization, data mining, parallel and distributed computing, high-performance computing
Preferred skills
Unix/Linux, large scale distributed systems (Hadoop, Spark), fraud investigation, LLM fine-tuning, reward modeling, RAG systems, knowledge graphs, memory-augmented architectures
Technologies
GenAI, LLMs, AI-agent architectures, Hadoop, Spark
Responsibilities
Design and deploy production ML systems for abuse pattern detection and automated enforcement; frame ambiguous security problems into scientific questions and drive them to production; monitor, diagnose, and retrain existing detection models; build graph-based entity analysis and identity resolution systems; execute rigorous experiments to measure model performance; architect and deploy GenAI/LLM solutions for investigation automation; contribute to the scientific roadmap and publish research findings; partner with investigators and engineers to translate operational insights into model features
Seniority
Senior, hands-on IC with mentorship responsibilities