Sr. Applied Scientist, AppStar Data Analytics & Engineering

Overview In this role you will design and deploy ML-driven solutions to prioritize and mitigate security risk at scale within Amazon’s DNA team. You’ll build production-grade ML pipelines and graph-based models that connect applications, services, teams, and vulnerabilities to inform security decisions. You will push scientific boundaries, validate approaches with rigorous experiments, and partner with engineers and TPMs to translate model outputs into actionable security insights. This is a high-impact opportunity to shape how Amazon identifies and addresses security risk across thousands of applications. Compensation / Benefitshealth insurance (medical, dental, vision)401(k) matchingpaid time offparental leaveRSUs and sign-on bonusescomprehensive benefits package ResponsibilitiesDesign, develop, and deploy ML models for application security prioritization, complexity scoring, and risk assessmentFrame ambiguous security problems into well-defined scientific challenges and propose novel approachesArchitect production-grade ML pipelines on AWS (S3, Glue, SageMaker, Neptune) including feature extraction, training, scoring, and deploymentDevelop and extend graph-based models capturing security relationships between apps, services, teams, and vulnerabilitiesDrive scientific agenda through research initiatives, experiments, and iterative model evaluationCollaborate with security engineers, data engineers, and TPMs to translate model outputs into actionable intelligence for security review programsEstablish and raise the bar for scientific rigor with peer reviews, reproducible documentation, and best practicesPublish results internally and externally at peer-reviewed venues when appropriate Key requirements3+ years of building machine learning models for business applicationsMaster's degree and 6+ years of applied research experience (or PhD)Experience programming in Java, C++, Python or related languageExperience with neural deep learning methods and MLcross-functional collaborationproblem framing and critical thinkingoral and written communicationneural deep learning methodsproduction ML pipelinesAWS services (S3, Glue, SageMaker, Neptune)