Data Scientist I, Customer Delivery Excellence Science

Amazon Amazon · Big Tech · Bellevue, WA · Research Science

This role focuses on improving global logistics and delivery experiences by building and validating predictive and classification models using machine learning techniques. It involves feature engineering, exploratory data analysis, and partnering with operations teams to implement ML solutions within Amazon's fulfillment and delivery network.

What you'd actually do

  1. Build and validate predictive models for delivery time estimation using historical delivery data, weather patterns, and traffic information
  2. Implement classification models to identify delivery exceptions and risk factors using established ML frameworks
  3. Apply feature engineering techniques to extract meaningful signals from transportation and logistics data
  4. Conduct exploratory data analysis on delivery performance metrics to identify improvement opportunities
  5. Create data visualizations and reports to communicate findings to operations partners

Skills

Required

  • Bachelor's degree or above in a quantitative field such as statistics, mathematics, data science, business analytics, economics, finance, engineering, or computer science
  • Experience communicating technical concepts to a non-technical audience
  • 1+ years of experience building supervised learning models (regression, classification) from problem definition through deployment
  • Proficiency in Python or R for data manipulation and statistical analysis, including libraries such as pandas, scikit-learn, or equivalent
  • Strong SQL skills for data extraction and transformation from relational databases
  • Experience with model evaluation techniques including cross-validation, performance metrics (RMSE, AUC, precision/recall), and statistical testing

Nice to have

  • Experience working with or evaluating AI systems
  • 2+ years of experience in data science or machine learning roles
  • Experience with transportation, logistics, or supply chain optimization problems
  • Familiarity with Amazon SageMaker, AWS services, or similar cloud ML platforms
  • Experience with gradient boosting frameworks
  • Knowledge of time-series forecasting or geospatial analysis techniques

What the JD emphasized

  • 1+ years of experience building supervised learning models (regression, classification) from problem definition through deployment
  • Experience communicating technical concepts to a non-technical audience

Other signals

  • improving global logistics
  • advanced machine learning
  • predictive models
  • classification models
  • feature engineering
  • delivery exceptions
  • risk factors