About this role
Design and build ML and LLM-powered solutions for Amazon's customer memory and personalization systems, owning end-to-end delivery from problem formulation and modeling to offline/online experimentation and production deployment at scale. Work areas include fact extraction, memory quality and lifecycle, temporal reasoning, grounded personalization, and tradeoffs between precision, recall, latency, and coverage.
Skills for this role
CC++PythonJavaPerlMachine LearningDeep LearningLarge Language Models (LLMs)Information RetrievalRecommender SystemsNatural Language ProcessingPersonalizationSparkAWSDistributed SystemsInformation ExtractionKnowledge RepresentationTemporal ReasoningOffline ExperimentationOnline ExperimentationProduction DeploymentScalabilityLow-latency Systems