Data Engineer, PXT Central Science

Amazon's People Experience and Technology Central Science (PXTCS) team uses economics, behavioral science, statistics, machine learning, and Generative AI to proactively identify mechanisms and process improvements that simultaneously improve Amazon and the lives, well-being, and value of work for Amazonians. PXTCS is an interdisciplinary team that combines the talents of science, engineering, and UX to build and deliver solutions that measurably achieve this goal — at a scale that touches over 1.5 million Amazonians worldwide. As a Data Engineer on PXTCS, you'll work side by side with economists, data scientists, software engineers, and applied scientists turning leading-edge ML and Generative AI models into reliable, scalable production systems. This is a rare chance to see your code directly shape how Amazon supports its workforce, spanning areas like benefits, compensation, recruiting, voice of employee, management practices, and organizational culture. We offer opportunities for builders to build and make history!

PXTCS combines economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements that improve both Amazon's operations and the experience of every Amazonian. Its engineering teams take science-driven insights and models — spanning areas like benefits, compensation, recruiting, voice of employee, management practices, and organizational culture — and turn them into production systems operating at Amazon's scale. PXTCS is an interdisciplinary group where engineering, applied science, and product work side-by-side, and where this team's output directly shapes how Amazon supports its workforce. Key job responsibilities PXTCS is looking for a data engineer with expertise in complex data environments. You will be responsible for enhancing our existing data architecture to further standardize metrics and definitions, building and testing new features, developing end-to-end data engineering solutions for complex analytical problems, and collaborating with economists, data scientists, and software engineers to translate data into actionable insights. Specific responsibilities include: • Data Pipeline Development: Design and maintain scalable data pipelines using native AWS services (Glue, EMR, Lambda); build monitoring and error handling for data workflows; optimize performance, reliability, and cost efficiency • Model Productionization & API Development: Develop and maintain APIs and data serving layers that productionize science models for downstream consumption; build batch and real-time inference pipelines • Data Integration & Quality: Build scalable feature extraction and processing frameworks for diverse data types; develop robust data quality and validation checks; create flexible schemas supporting evolving requirements • Cross-team Collaboration: Partner with economics, data science, and software engineering teams to translate analytical requirements into production-ready solutions; participate in technical design reviews and architecture discussions • Analytics & Infrastructure: Maintain layered data systems used by economists and scientists; build automated reporting solutions; work across multiple interconnected AWS accounts with security best practices About the team PXTCS combines economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements that improve both Amazon's operations and the experience of every Amazonian. Its engineering teams take science-driven insights and models — spanning areas like benefits, compensation, recruiting, voice of employee, management practices, and organizational culture — and turn them into production systems operating at Amazon's scale. PXTCS is an interdisciplinary group where engineering, applied science, and product work side-by-side, and where this team's output directly shapes how Amazon supports its workforce.