Staff Scientist, Applied AI

Staff Scientist, Applied AI Department: Data Science Team: Data Scientist Location: San Francisco, United States Type: Full-Time **About the Role** As a Scientist on the Uber Applied AI team, you will help build intelligent systems that help to power various personalization applications at Uber. While application teams are focused on specific features, this team develops horizontal intelligence that can help enhance multiple systems. Examples of the team's past work include semantic information about our Eats merchandising catalog, multi-modal content enhancements, etc. **What You'll Do** - Lead planning and prioritization by identifying high-impact segments in the Uber ecosystem of the market the AI team should be building for. - Understand how personalization systems work on multiple teams and identify opportunities by looking at the data. - Develop measurement frameworks to evaluate the quality of these foundational AI investments and use that to guide future direction. - Design, execute, and analyze large-scale experiments. - Present findings and recommendations to senior technical and business leaders. **Basic Qualifications** - Ph.D., M.S., or Bachelors degree in Statistics, Economics, Operations Research, or other quantitative fields. - Minimum 5 years of industry experience as a Data Scientist or equivalent role (3+ if holding a Ph.D.) - Expertise in experimental design and statistical analysis for A/B testing and causal inference. - Proficiency in SQL & Python for exploratory data analysis and modeling. - Strong analytical skills, with experience extracting user behavior insights from large datasets. - Experience with data visualization and dashboarding tools (e.g., Tableau, Matplotlib). **Preferred Qualifications** - Expertise in synthesizing complex technical analyses into clear insights to influence product direction. - Prior experience working in personalization or recommendation systems spaces. - Exceptional communication skills across technical, non-technical, and executive audiences. - Comfortable with leading ambiguous horizontal workstreams involving multiple teams of