Applied Scientist, People eXperience Technology Central Science (PXTCS)

The Central Science Team within Amazon's People Experience and Technology org (PXTCS) uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, well-being, and the value of work to Amazonians. We are an interdisciplinary team, which combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal.

We are looking for an Applied Scientist to build models at the intersection of prediction, causal inference, and optimization. These models are the science foundation for operational workforce planning at Amazon"s scale - predicting who stays, who shows up, and suggesting which levers Amazon should use to ensure smooth business operations and the best employee experience.

As an Applied Scientist, you will own the forecasting and prediction models. Your forecasts go straight into planning cycles that commit real hiring, so accuracy, calibration, and being able to explain a number to business partners all matter for success. You will work with other scientists and economists on the team, so you will need an interdisciplinary mindset and an eye on how causal inference and forecasting interact. You will also collaborate with engineers to put models into production, so you should be comfortable owning a live system and not only an analysis.

Key job responsibilities

• Design and develop forecasting and prediction models that feed Amazon"s workforce planning and optimization systems.
• Build models and algorithms from prototype to production-level systems, and support them once planners depend on them.
• Partner with the team"s economists to connect forecasts with causal estimates of policy levers.
• Extend models to new sites, shifts, and business lines as the platform expands.
• Translate ambiguous business problems into modeling approaches, and drive the technical design with planning, engineering, and operations partners.

A day in the life

• Analyze data to investigate a forecast miss or model performance, and identify improvements
• Brainstorm modeling approaches with fellow scientists and economists on the team
• Build and backtest new features for your model
• Run a simulation or experiment to evaluate your model"s performance
• Meet with planning, compensation, or finance partners to review requirements, data, design, or other project decisions
• Review code changes or a design document from a fellow scientist or engineer
• Write and present a document covering method, results, and recommendations