About the job
At Amazon, our Demand Forecasting team is tackling one of the most ambitious challenges in applied time series research: building large-scale foundation models that generalize across an enormous and diverse catalog of products, geographies, and business contexts. This is not incremental modeling work. We are redefining what's possible in demand forecasting. Our team operates at a scale that is unmatched in industry. We run experiments across millions of products simultaneously, pushing the boundaries of what foundation models can learn from vast, heterogeneous time series data. We are also exploring novel data generation techniques that augment our already unprecedented dataset — opening new frontiers in model generalization and forecasting for products with limited or no sales history. The models you build here will ship to production and directly influence hundreds of millions of dollars in automated inventory decisions every week, labor plans for tens of thousands of employees, and Amazon's financial outlook. Beyond operational impact, this team contributes to the broader scientific community and advances the state of the art in time series foundation models.
Responsibilities
Design and run rigorous experiments at scale to evaluate and improve foundation model performance across hundreds of millions of products, geographies, and business verticals
Lead the end-to-end lifecycle of forecasting models — from research and experimentation through production launch — including defining success metrics, obtaining stakeholder sign-off, and managing rollout
Conduct online and offline labs to measure the real-world impact of forecast improvements beyond accuracy, including downstream supply chain, inventory, and financial outcomes
Develop and deploy production-grade deep learning and statistical models using Python, Scala, SQL, and related tools
Perform large-scale exploratory data analysis to uncover patterns, identify opportunities, and inform model development
Translate complex research findings into clear insights and recommendations for technical and non-technical stakeholders at all levels
Contribute to Amazon's scientific community and the broader research field through collaboration and publication in top-tier venues
Qualifications
Minimum
3+ years of machine learning, statistical modeling, data mining, and analytics techniques experience
3+ years of data querying languages (e.g. SQL), scripting languages (e.g. Python) or statistical/mathematical software (e.g. R, SAS, Matlab, etc.) experience
3+ years of data scientist experience
Bachelor's degree
Preferred
Master's degree, or PhD
2+ years of deep learning, computer vision, human robotic interaction, algorithms implementation experience
Experience processing, filtering, and presenting large quantities (hundreds of millions/billions of rows) of data
Experience with forecasting and statistical analysis
Natural curiosity and desire to learn