About the job
Amazon's Community Intelligence Science & Engineering (CISE) team operates at the intersection of computational social science, artificial intelligence, and operational planning. We build and validate models that quantify how Amazon's operational presence impacts local communities, and embed those signals into core planning models that drive Last Mile, Middle Mile, and site-level decisions across our network. You'll work at the forefront of applied AI and causal inference, building systems that influence decisions affecting thousands of communities daily. This role offers a unique opportunity to shape how the world's most customer-centric company measures, forecasts, and reduces operational risk to the communities we serve.
Responsibilities
- Collaborate with operational science teams to integrate community risk signals into existing operational models and decision-making systems, with a focus on quantifying performance lift and defining integration architecture
- Design and execute experiments to measure how community-impacting operational policies affect business outcomes
- Build automated causal discovery systems leveraging knowledge graphs, LLMs, and document understanding to uncover relationships between operational policies and community outcomes
- Design and deploy production ML forecasting systems with extended prediction horizons using multi-modal data sources, including survey-based indices, geospatial risk features, and operational metrics
- Mentor junior scientists and contribute to building a research culture that balances high-risk, high-reward innovation with reliable product delivery
Qualifications
Minimum
- 3+ years of investigating the feasibility of applying scientific principles and concepts to business problems and products experience
- PhD, or Master's degree and 5+ years of quantitative field research experience
- Experience with big data technologies such as AWS, Hadoop, Spark, Pig, Hive etc.
- Knowledge of quantitative approaches (e.g., t-tests, regressions, ANOVAs, etc.)
- Knowledge of AWS platforms such as S3, Glue, Athena, Sagemaker
- Experience in standard machine-learning and statistical modeling tools and techniques (e.g. random forests, gradient-boosted regression, LASSO, logistic regression)
- Experience applying theoretical models in an applied environment
Preferred
- Experience converting research studies into tangible real-world changes
- Experience with discrete and continuous optimization methodologies and algorithms
- Experience applying quantitative analysis to solve business problems and making data-driven business decisions