About the job
As an Applied Scientist, you will develop and refine forecasting models that power Zillow’s view of the housing market at national, regional, and local levels. You will collaborate closely with other applied scientists, data scientists, economists, and engineers to turn large-scale housing and economic data into actionable insights for senior leaders and partners across the company.
Responsibilities
No responsibilities listed.
Qualifications
Minimum
2+ years of experience working with time series and/or spatial forecasting problems, ideally including hierarchical or panel data settings in an academic or industry environment.
Solid foundation in traditional econometric methods and modern machine learning techniques relevant to forecasting.
Hands-on experience preparing and processing large-scale time series or panel datasets, including data cleaning, preprocessing, and feature engineering; familiarity with distributed computing frameworks (such as Spark) is a plus.
Ability to design and apply appropriate model evaluation and validation approaches for time series forecasting, including backtesting, cross-validation strategies, and forecasting accuracy metrics.
Proficiency with software development best practices and tools (for example, version control systems such as Git and cloud-based platforms for deploying and monitoring models).
Preferred
Demonstrated interest in leveraging AI tools (e.g., LLMs, copilots, automation frameworks) to improve research productivity and forecasting workflows.
Passion for experimenting with emerging technologies and incorporating them into day-to-day analytical work to increase efficiency and impact.
Strong interest in the housing market and the economic, demographic, and policy factors that influence housing outcomes along with the ability to explain complex quantitative concepts and tradeoffs to diverse audiences.