About the job
As a Senior Applied Scientist, you will tackle complex, real-world challenges that directly shape Zillow’s virtual touring experiences. You’ll have a broad impact by defining and solving ambiguous problems, developing innovative models, and collaborating with cross-functional teams to deliver features that enhance how users explore homes. Your work will drive the next wave of immersive, user-centric experiences at Zillow.
Responsibilities
Frame and solve complex perception problems using scientific and engineering best practices.
Collaborate with product, engineering, and design teams to translate user needs into research questions and solutions.
Design, implement, and iterate on machine learning and computer vision models for structured understanding of spaces.
Develop robust evaluation pipelines and experiments to measure and improve model performance.
Integrate models into production systems, ensuring reliability and scalability.
Monitor and improve deployed models based on real-world data and user feedback.
Mentor and support team members in modeling, evaluation, and research practices.
Communicate findings and technical decisions clearly to both technical and non-technical partners.
Qualifications
Minimum
5+ years of experience as an applied or research scientist working on machine learning or computer vision with real-world data.
Proficiency in Python and at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX), with a track record of building and deploying models.
Experience shipping production ML systems, including data pipelines, deployment, monitoring, and iteration.
Strong understanding of probability, statistics, and experimental design, with the ability to apply these to practical evaluation strategies.
Demonstrated ability to work with noisy, imperfect datasets and design robust solutions for challenging edge cases.
Preferred
Experience with geometry-heavy or spatial understanding problems, or multi-modal/sensor-fusion challenges, is a plus.
Proven ability to communicate complex technical ideas to both technical and non-technical audiences, and to collaborate effectively in cross-functional teams.
Prior success in ambiguous, evolving problem spaces or zero-to-one environments is valued.
Contributions to the broader ML or computer vision community (e.g., publications, patents, open-source) are a plus.