Applied Scientist, Amazon Precision Match

Amazon Amazon · Big Tech · Seattle, WA · Machine Learning Science

The Applied Scientist will design, develop, and optimize ML-based matching algorithms for a high-scale customer service routing system. This involves feature engineering, running simulations and A/B tests, building real-time inference systems, and developing retraining infrastructure. The role focuses on connecting customers with the best service options using recommendation and real-time ML inference techniques.

What you'd actually do

  1. Design, develop, and optimize ML-based matching algorithms that pair customers with optimal CSAs based on contact complexity, intent, and CSA skill profiles.
  2. Build and iterate on feature engineering pipelines across CSA-level (skills, tenure, sentiment handling), contact-level (intent, complexity, urgency), and customer-level (language, communication style) attributes.
  3. Run offline simulations on large-scale historical contact data and design statistically rigorous A/B experiments to validate matching improvements.
  4. Develop real-time low-latency scoring and inference systems for production contact routing.
  5. Address the cold start problem for new CSAs and build continuous model retraining infrastructure using production feedback.

Skills

Required

  • building models for business application
  • PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience
  • programming in Java, C++, Python or related language
  • algorithms and data structures
  • parsing
  • numerical optimization
  • data mining
  • parallel and distributed computing
  • high-performance computing

Nice to have

  • Unix/Linux
  • professional software development

What the JD emphasized

  • ML-based matching algorithms
  • real-time ML inference
  • large-scale experimentation
  • low-latency scoring and inference systems
  • continuous model retraining infrastructure

Other signals

  • ML-based matching algorithms
  • real-time ML inference
  • large-scale experimentation
  • low-latency scoring and inference systems
  • continuous model retraining infrastructure