Applied Scientist, Robotics Delivery and Packaging Innovation; RDPI

Applied Scientist, Robotics Delivery and Packaging Innovation (RDPI)

Do you want to be part of a team that’s revolutionizing Amazon’s fulfillment and packaging technology? Are you ready to optimize systems that process tens‑of‑millions of customer packages daily with the lowest cost to serve and a defect‑free customer experience? Do you have a passion for solving complex science challenges and building a sustainable e‑commerce experience? The Robotics Delivery and Packaging Innovation (RDPI) team is seeking an Applied Scientist who will join experts in Machine Learning, Statistics, Operations Research, Computer Vision, and Generative AI to break new ground in the world of automated packaging solutions.

The RDPI team owns mission‑critical automation and packaging solutions that impact billions of customer shipments annually across Amazon’s global marketplaces. We manage billions of dollars in material spend and packaging labor costs while driving significant reductions in carbon emissions. This is an exciting opportunity to work on large‑scale automation challenges that directly impact customer experience, operational efficiency, and environmental sustainability at one of the world’s largest e‑commerce companies.

• Leverage generative AI technologies to develop scalable solutions for automated product compatibility and safety assessments (e.g., evaluating product shipping compatibility, safety requirements, and packaging configurations).
• Develop advanced AI models by extracting predictive features from multiple data sources (product/packaging images, product descriptions, sensor data, geospatial data) to forecast package‑related damages and optimize packaging decisions based on customer preference prediction.
• Develop and implement computer vision solutions to automate packaging workflows and detect product/packaging defects in real‑time operations.
• Build causal inference models to capture the downstream impacts of different packaging designs and delivery experience.
• Design and implement robotic control algorithms to optimize machine efficiency and meet diverse business objectives.

Scientists on our team work daily with dedicated product and engineering partners to bring innovative solutions from concept to production. You will divide your time between deep technical work—building models, analyzing results, and iterating on algorithms—and collaborative activities such as design reviews, stakeholder presentations, and cross‑functional planning. You will also have opportunities to support science initiatives across the broader RDPI organization (1500+ people), partnering with diverse teams to solve high‑impact problems and scale your solutions across Amazon.

We are a team of scientists with diverse technical backgrounds spanning Machine Learning, Operations Research, Causal Inference, and Econometrics. We tackle complex, high‑impact problems that directly influence Amazon’s strategic decisions and financial performance. Our solutions typically require combining multiple methodologies, and you will work collaboratively with other scientists while partnering closely with product and engineering teams to bring your innovations into production systems.

You’ll have the opportunity to grow your expertise across disciplines while delivering measurable business impact at scale.

• PhD, or Master’s degree and 4+ years of CS, CE, ML or related field experience.
• Experience programming in Java, C++, Python or related language.
• Experience building machine learning models or developing algorithms for business application.
• Experience implementing algorithms using both toolkits and self‑developed code.
• Experience applying theoretical models in an applied environment.

• PhD, or a Master’s degree and experience in state‑of‑the‑art deep learning model architecture design, deep learning training and optimization, and model pruning.
• Experience in patents or publications at top‑tier peer‑reviewed conferences or journals.
• Experience designing experiments and statistical analysis of results.
• Experience with deep generative models (GANs, VAEs,…