Data Scientist Ii, Middle Mile Transportation Science Team

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

Data Scientist II on the NASC & TOM Science team at Amazon, focusing on Operations Research, Machine Learning, and AI for middle mile transportation planning. The role involves designing, building, and shipping ML and optimization models end-to-end, from problem framing to deployment and integration with planning tools. Key responsibilities include developing forecasting and optimization solutions, building validation and monitoring pipelines, and partnering with engineering teams for productionization.

What you'd actually do

  1. Design and implement complex ML and optimization solutions (forecasting, MIP/LP, simulation, Deep learning / foundation model);
  2. Drive end-to-end delivery of scalable models — from data exploration and feature engineering through training, evaluation, deployment, and post-launch monitoring;
  3. Develop new modeling patterns and analytical frameworks for forecasting (multivariate, hierarchical, causal-DAG, model-chaining) and optimization;
  4. Build robust model validation, backtesting, and monitoring pipelines; identify and eliminate sources of leakage, bias, and silent failure;
  5. Define and own model performance metrics (e.g., WAPE) tied to business outcomes;

Skills

Required

  • Master's degree in Science, Technology, Engineering, or Mathematics (STEM)
  • 2+ years of data querying languages (e.g. SQL), scripting languages (e.g. Python) or statistical/mathematical software (e.g. R, SAS, Matlab, etc.) experience
  • 2+ years of machine learning/statistical modeling data analysis tools and techniques, and parameters that affect their performance experience
  • Experience with AWS services including S3, Redshift, Sagemaker, EMR, Kinesis, Lambda, and EC2
  • Proficiency in statistical modeling and machine learning — time-series forecasting, regression, tree-based methods, and deep learning.
  • Demonstrated ability to communicate technical results to non-technical business audiences.

Nice to have

  • 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 machine learning/statistical modeling data analysis tools and techniques, and parameters that affect their performance experience

What the JD emphasized

  • end-to-end Machine learning Operation cycle
  • deployed models that drive multi-million-dollar planning decisions
  • productionize models

Other signals

  • end-to-end Machine learning Operation cycle
  • deployed models that drive multi-million-dollar planning decisions
  • productionize models