Data Scientist

Public Storage•Glendale, CA

5d•$155,000 - $170,000•Hybrid

About The Position

Public Storage is expanding its creative team to enhance its consistent and engaging visual brand presence. We are seeking a Data Scientist to establish trust and confidence across the organization by designing, building, and deploying ML models. This role involves translating ambiguous business problems into well-scoped analytical solutions, writing production-grade Python and SQL, and collaborating with data engineering on pipeline architecture and model deployment. The Data Scientist will also leverage LLMs and modern AI tooling, mentor junior team members, own model documentation, and monitor deployed models for performance.

Requirements

Bachelor's/Master's in a STEM field (statistics, economics, CS, engineering, applied math, or similar)
Expert in SQL and Python in real-world
Strong verbal communication skills: ability to effectively communicate cross-functionally
Expert-level SQL and Python in production settings
Strong applied statistics — comfortable in frequentist and Bayesian frameworks
Experience with the modern ML stack: scikit-learn, XGBoost, PyTorch or equivalent; experiment tracking (MLflow, W&B, or similar)
MLOps fundamentals: versioning, model registries, scheduled retraining, monitoring
Git-based workflows (GitHub or GitLab) with code review habits

Nice To Haves

Ph.D. in a quantitative field a plus, equivalent experience considered: 5+ years in a production data science role
Alternative to education, 6+ years of experience as contributor/leader
Demonstrated track record shipping models that drove measurable business outcomes
Familiarity with dbt or orchestration tools (Airflow, Prefect, etc.)
Exposure to web behavioral data and customer lifecycle modeling (churn, propensity, LTV)

Responsibilities

Exceptional verbal and written skills to convey ideas, problems and solutions
Establish trust and confidence as the data scientist, up, across, and down the organization
Design, build, and deploy ML models across structured and unstructured data
Translate ambiguous business problems into well-scoped analytical solutions with clear trade-offs documented
Write production-grade Python and SQL; contribute to shared codebases with reproducibility and refactorability in mind
Collaborate with data engineering on pipeline architecture, feature stores, and model deployment patterns
Leverage LLMs and modern AI tooling where appropriate, sound judgment on when not to
Mentor analysts and junior data scientists through code review, whiteboarding, and hands-on pairing
Own model documentation, versioning, and knowledge artifacts (Confluence, GitHub)
Monitor deployed models for drift and degradation; refresh proactively, not reactively