Machine Learning Scientist III - Whole Trip AI

Expedia Expedia · Hospitality · Seattle, Austin Domain 11 - HomeAway, USA - California - San Jose, WA

Machine Learning Scientist III role focused on building and deploying end-to-end AI solutions for Expedia's travel platform, covering search ranking, recommendations, personalization, and optimization. The role involves developing scalable data pipelines, monitoring ML systems, and using evaluation frameworks to improve model performance.

What you'd actually do

  1. Design and implement end-to-end model pipelines to production across multiple product domains, including ranking, recommendations, search, and personalization
  2. Develop and maintain scalable data pipelines, data quality checks, and model monitoring to ensure reliability, performance, and responsible behavior of ML systems in production
  3. Collaborate with cross-functional partners (product, analytics, engineering) to translate ambiguous business needs into well-scoped ML projects, communicate findings, and influence decision making with data-driven insights
  4. Use A/B tests and offline/online evaluation frameworks to measure model impact and guide iterative improvement

Skills

Required

  • Python
  • ML frameworks and libraries
  • model development
  • model training
  • model evaluation
  • translate problem statements into ML tasks
  • design model and data structures
  • APIs
  • data models
  • AI-driven systems
  • AI/ML concepts

Nice to have

  • Graduate degree in a quantitative field
  • ML, optimization, or statistical modeling coursework/research
  • modern ranking & recommendation modeling approaches
  • optimizing ML systems in production
  • monitoring
  • alerting
  • retraining
  • model governance
  • performance
  • robustness
  • fairness
  • designing and improving ML architectures at scale
  • model selection
  • feature store design
  • low-latency, high-availability production systems
  • natural language search techniques
  • agentic workflows

What the JD emphasized

  • end-to-end model development
  • applying AI/ML concepts to real world products
  • safety and reliability

Other signals

  • end-to-end model pipelines
  • production deployment
  • scalable data pipelines
  • model monitoring
  • responsible AI
  • A/B tests
  • offline/online evaluation