Are you passionate about solving complex business problems at scale through Generative AI? Do you want to build intelligent systems that reason, act, and learn from minimal supervision? Are you excited about taking innovative AI solutions from proof-of-concept to production? If so, we have an exciting opportunity for you on Amazon's Trustworthy Shopping Experience (TSE) team. At TSE, our vision is to guarantee customers a worry-free shopping experience by earning their trust that the products they buy are safe, authentic, and compliant with regulations and policy. We give customers confidence that Amazon stands behind every product and will make it right in the rare chance anything goes wrong. We do this in close partnership with our selling partners and empower them with best-in-class tools and expertise required to offer a high-quality selection of compliant products that customers trust. As a Senior Applied Scientist, you will lead the development of next Gen AI solutions to automate complex manual investigation processes at Amazon scale. You will work on some of the most fascinating challenges in applied AI—building systems that reason and act autonomously, learn rich representations from structured and relational data without extensive labels, adapt rapidly from limited examples, improve through feedback and interaction, seamlessly connect visual and textual understanding, and compress complex model capabilities into efficient, deployable systems. Your innovations will deliver significant impact to cost-of-serving customers while maintaining the highest standards of trust and safety. This role offers end-to-end ownership—from initial research and proof-of-concept through production deployment. You will see your innovations serving hundreds of millions of customers within months, not years. Key job responsibilities • Design and build next-generation agentic AI systems that think, plan, and act—capable of multi-step reasoning, dynamic tool use, and autonomous task execution that transforms how investigations are conducted • Invent novel solutions across the AI frontier: autonomous reasoning, self-supervised representation learning, few-shot adaptation, feedback-driven optimization, and multimodal intelligence • Develop deep expertise in research areas strategic to the organization while maintaining solid understanding across adjacent domains • Identify and frame ill-defined customer and business problems, devising new research methodologies using a customer-obsessed scientific approach • Tackle ambiguous, high-impact challenges where the problem itself isn't yet fully defined—and shape the science roadmap that solves them • Automate complex investigation workflows involving unstructured text, documents, images, symbols, and rich relational data—directly impacting hundreds of millions of customers • Anticipate and articulate key scientific challenges of current and future customer needs, proactively presenting interventions to address them • Drive ideas from whiteboard sketch to production system—prototype rapidly, iterate relentlessly, and deploy solutions that scale • Engineer efficient, production-ready systems by distilling models into lightweight, cost-effective deployments without sacrificing capability • Build a proven track record of repeatedly delivering innovative, impactful scientific solutions into production • Write clear, compelling narratives and documentation that enable others to understand, reproduce, and build upon your work • Provide architectural guidance for AI systems—whether building from scratch or transforming existing solutions • Write significant portions of critical-path code that form the backbone of complex systems, setting the standard for technical excellence • Maintain deep knowledge of team solutions and proactively drive utilization and improvement upon state-of-the-art techniques • Independently assess emerging technologies and make sound decisions on adoption for your systems • Champion engineering best practices and conduct rigorous peer reviews that raise the bar for the entire team—your solutions, code, and designs set the example for others • Influence team strategy by contributing to roadmaps, goals, priorities, and scientific approach • Lead and participate in science reviews for your team and adjacent teams • Build consensus by communicating effectively, harmonizing discordant views, and resolving contentious technical debates • Actively participate in hiring top scientific talent and mentor fellow scientists—improving their skills and accelerating their ability to deliver impact • Drive the team's scientific agenda and role model the pursuit of publications at peer-reviewed venues when appropriate A day in the life No two days look the same. You might spend one morning diving deep into experiment results, then pivot to whiteboarding a new approach with engineers in the afternoon. Some days you'll be hands-on coding a proof-of-concept; others you'll present findings to leadership or collaborate with investigators to understand their workflows. You'll regularly navigate ambiguity—translating loosely defined business problems into concrete science roadmaps. What remains constant: the opportunity to see your innovations move from idea to production and directly impact how Amazon protects customers. About the team The TSE Ops is responsible for the human-in-the-loop products and technology used in risk investigations at Amazon. We are building the next generation of AI-powered investigation systems—combining autonomous reasoning, multimodal understanding, and continuous learning to create intelligent solutions that adapt and improve over time. Our mission is to: • Reduce the cost of performing investigations through intelligent automation • Optimize the investigator experience where manual interventions are needed • Leverage state-of-the-art technology and GenAI to deliver transformative products We value scientific rigor with pragmatic execution—where promising ideas move from whiteboard to production rapidly, and where you'll have the freedom to explore novel approaches while delivering tangible business outcomes.
