Job ID 10419077 - Amazon Kuiper Manufacturing Enterprises LLC Build the scientific intelligence layer powering Amazon's satellite manufacturing system. We are looking for a Senior Applied Scientist to lead the development of models that transform fragmented manufacturing, test, quality, and operational data into a unified, closed‐loop intelligence system that directly improves how satellites are built. Export Control Requirement: Due to applicable export control laws and regulations, candidates must be a U.S. citizen or national, U.S. permanent resident (current Green Card holder), or lawfully admitted into the U.S. as a refugee or granted asylum. Key Responsibilities Lead the design, training, and deployment of machine learning models, including LLM‐based systems, retrieval models, and task‐specific models Translate ambiguous, real‐world manufacturing problems into well‐defined scientific problems, modeling approaches, and evaluation criteria Train, fine‐tune, and evaluate models using large‐scale, noisy, and heterogeneous datasets with incomplete or delayed ground truth Develop models over partially observed systems spanning test data, inspection signals, quality records, supplier data, and knowledge systems Invent and extend approaches for problems such as anomaly detection, root‐cause inference, multimodal learning, and generative AI under real‐world constraints Define evaluation frameworks that capture real‐world failure modes, distribution shift, and decision risk, and use them to drive model iteration Make principled trade‐offs between model complexity, data quality, and generalization, and justify when to extend or depart from state‐of‐the‐art approaches Work closely with engineering teams to deploy models into production systems with monitoring, feedback capture, and continuous retraining Build closed‐loop learning systems where model outputs influence design, manufacturing, and test decisions Influence scientific direction across teams and mentor scientists and engineers A Day in the Life You may start by partnering with Quality, Manufacturing, and engineering teams to define and scope a training dataset for a root‐cause prediction model, curating labels from historical cases. You then design and execute experiments to train and fine‐tune models, comparing approaches across architectures, features, and data slices. Later, you analyze benchmark results, identify failure modes, bias, and generalization gaps, and refine evaluation datasets to better reflect real‐world edge cases. You iterate on model design and data quality before deploying the highest‐performing model into a production workflow with monitoring, feedback capture, and retraining. About the Team Leo Intelligence Technologies (LIT) is the centralized AI team within Leo Satellite Build Systems. We build the shared foundation for AI across Production Operations, including governed data assets, models, retrieval systems, evaluation frameworks, and knowledge services. We operate on real‐world systems where model outputs directly influence physical outcomes. We treat evaluation, data quality, and model behavior as first‐class problems, and hold a high bar for rigor, auditability, and production readiness. Our work sits at the center of a shift toward AI‐native manufacturing, where data, models, and feedback loops continuously improve production outcomes. Basic Qualifications 3+ years of building machine learning models for business application experience PhD, or Master's degree and 6+ years of applied research experience Programming experience in Java, C++, Python or related language Experience with neural deep‐learning methods and machine learning Experience training and evaluating machine learning models on large‐scale, real‐world datasets Experience applying statistical analysis and experimentation to measure model performance and drive improvements Experience working with engineering teams to deploy machine learning models into production systems Preferred Qualifications Experience training and deploying LLM‐based systems, retrieval‐augmented generation (RAG), or agentic workflows Experience designing evaluation frameworks for production AI systems, including safety, grounding, and regression testing Experience building closed‐loop or feedback‐driven ML systems Experience working with ambiguous problem spaces and inventing novel modeling approaches Experience influencing scientific direction across teams and mentoring other scientists Experience in manufacturing, aerospace, robotics, or other complex physical‐world systems Experience working with governed data environments, compliance constraints, or access‐controlled systems Experience building systems where model outputs directly drive operational or physical‐world decisions Amazon is an equal opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status. Base salary range: $167,100.00 – $226,100.00 USD annually (El Segundo, CA; Long Beach, CA; Bellevue, WA). Benefits include health insurance, 401(k) matching, paid time off, and parental leave. Learn more about our benefits at https://amazon.jobs/en/benefits #J-18808-Ljbffr