Principal Machine Learning Scientist

Role Overview

You will be the technical anchor for Tripadvisor’s core discovery engine, leading machine learning strategy and execution for how users search, discover, and organize complex travel itineraries. The role bridges cutting-edge AI research and production-grade engineering, impacting outcomes like user engagement and booking conversion.

What You’ll Do

• Technical leadership & execution: Drive the ML roadmap for Search, Retrieval, Ranking, and Recommendation systems within the Trips vertical; translate business goals into scalable ML architectures.
• Algorithm innovation: Design, prototype, and scale next-generation ranking/recommendation models using sequential recommenders, representation learning, and deep multi-objective frameworks.
• System architecture & scalability: Oversee deployment of low-latency, high-throughput retrieval and ranking pipelines (e.g., multi-stage retrieval, vector search) supporting real-time processing over billions of travel data points.
• Cross-functional collaboration: Partner with Product Managers and Engineering/Data Science peers to optimize multi-task business objectives; serve as primary ML technical authority for the Trips vertical.
• Talent multiplier: Mentor and coach ML scientists; promote best practices across MLOps, rigorous A/B testing, data privacy, and code quality.

Requirements

• Education: Ph.D. or Master’s degree in CS, Machine Learning, Statistics, or another highly quantitative field.
• Experience: 8+ years building and deploying large-scale ML models in production, with demonstrated shipping impact at scale.
• Core expertise:
  • Retrieval & ranking (SOTA): Practical experience with Multi-Task Learning (MTL), Multi-gate Mixture-of-Experts (MMoE), or similar multi-objective architectures.
  • Sequential & temporal modeling: Building sequential recommendation systems for real-time session dynamics and long-term preferences.
  • Representation learning: Embedding generation and deep semantic retrieval; (multi-modal experience is indicated but not fully shown in the text snippet).