Applied Scientist

Zillow

1d•$132,400 - $222,600•Remote

About The Position

Join our dynamic team, leveraging a comprehensive database of over 100 million homes, transactions, and users to predict and analyze housing market trends. As an Applied Scientist specializing in traditional econometrics and machine learning with a focus on time series and hierarchical forecasting, you will play a critical role in shaping our understanding and forecasting of the housing market. As an Applied Scientist you will actively participate in Collaborative Forecast Development: Work alongside other applied scientists, data scientists and economists to develop hierarchical forecasts for the housing market at various regional levels: Feature Extraction: Identify economic and demographic driving forces of housing market growth across regions and extract features that predict housing market trends 1 to 2 years in the future. Model Deployment: Collaborate closely with the engineering team to deploy models into production environments. Forecast Production: Contribute to monthly forecast production, communicating forecast performance and the housing market outlook to business partners and senior leadership team. Scenario Modeling: Develop scenario models that capture a wide range of housing market outcomes and contribute to a company-wide stress-testing framework. This role has been categorized as a Remote position. “Remote” employees do not have a permanent corporate office workplace and, instead, work from a physical location of their choice, which must be identified to the Company. U.S. employees may live in any of the 50 United States, with limited exceptions.

Requirements

Strong interest and understanding of the housing market and the economic factors that influence it.
Passionate about working on innovative solutions to time series, panel data, and hierarchical forecasting problems, conducting independent research, and applying research methods to real-world problems.
Strong foundation in traditional econometric methods and modern machine learning techniques.
Proficient in data cleaning, preprocessing, and feature engineering for time series data.
Experience with large-scale datasets and familiarity with distributed computing frameworks such as Spark.
Ability to design and implement robust model evaluation and validation strategies, including cross-validation and backtesting.
Experience with metrics for time series forecasting accuracy and performance assessment.
Proficient in cloud-based platforms and tools for deploying and monitoring machine learning models.
Proficient in building AI agent to automate manual work
Excellent verbal and written communication skills, capable of conveying complex concepts to a broad audience.
2+ years of proven experience working in time series/spatial forecasting space.
Master’s or PhD degree in Mathematics, Statistics, Economics, Econometrics, Physics, Earth Sciences, or a related scientific field

Responsibilities

Develop hierarchical forecasts for the housing market at various regional levels.
Identify economic and demographic driving forces of housing market growth across regions and extract features that predict housing market trends 1 to 2 years in the future.
Collaborate closely with the engineering team to deploy models into production environments.
Contribute to monthly forecast production, communicating forecast performance and the housing market outlook to business partners and senior leadership team.
Develop scenario models that capture a wide range of housing market outcomes and contribute to a company-wide stress-testing framework.