Data Scientist I, Fma

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

Data Scientist role focused on building and owning machine learning models for offer ranking and recommendation systems on Amazon. The role involves feature engineering, production deployment, A/B experimentation, data mining, and partnering with engineers to operationalize models at scale for a high-traffic, low-latency environment. The goal is to improve customer experience and seller competition by predicting the best offer in real-time.

What you'd actually do

  1. Build and own machine learning models that rank and recommend offers to customers across Amazon's product catalog, from feature engineering through production deployment
  2. Design and analyze A/B experiments to measure the impact of algorithm changes on customer experience, seller competition, and business outcomes
  3. Mine large-scale datasets to identify patterns and signals — seller behavior, pricing dynamics, fulfillment performance — that improve how we predict the best offer for each customer
  4. Translate ambiguous business questions into well-defined data science problems, and communicate findings clearly to engineers, product managers, and leadership
  5. Partner with software engineers to operationalize models at scale, ensuring they perform reliably under high-traffic, low-latency conditions

Skills

Required

  • data querying languages (e.g. SQL)
  • scripting languages (e.g. Python)
  • statistical/mathematical software (e.g. R, SAS, Matlab, etc.)
  • creating or contributing to mathematical textbooks, research papers, or educational content
  • Master's degree in Science, Technology, Engineering, or Mathematics (STEM)

Nice to have

  • Ph.D. in Science, Technology, Engineering, or Mathematics (STEM)
  • statistical packages and business intelligence tools such as SPSS, SAS, S-PLUS, or R
  • machine learning concepts and their application to reasoning and problem-solving
  • clustered data processing (e.g., Hadoop, Spark, Map-reduce, and Hive)
  • working with or evaluating AI systems
  • applying quantitative analysis to solve business problems and making data-driven business decisions
  • effectively communicating complex concepts through written and verbal communication

What the JD emphasized

  • operating at internet scale
  • directly affect what customers buy
  • getting the answer right really matters
  • operationalize models at scale
  • high-traffic, low-latency conditions
  • consequential algorithms
  • enormous impact
  • shipping things that matter
  • not taking shortcuts that erode customer trust

Other signals

  • ranking models
  • recommendation systems
  • real-time systems
  • A/B testing
  • large-scale data