Role Purpose
We are looking for a hands-on Data Scientist to strengthen our analytics pipelines that support planning, forecasting, and inventory optimization across the HP supply chain network. This role bridges data science modeling and analytics operations, ensuring that analytical work moves smoothly from exploration to deployment and supports better planning and decision making. The position works closely with other data scientists, data engineers, and planning teams to design, test, and scale model-driven solutions in Databricks and GitHub.
Key Responsibilities
- Build and maintain reproducible analytical workflows in Databricks using Python, PySpark, and MLflow.
- Apply feature engineering best practices, including lag features, rolling averages, and proper handling of data cutoffs.
- Translate prototypes into parameterized, reusable code that supports weekly or monthly production runs.
- Develop and evaluate models for short-term forecasting, attach rate tracking, and inventory buffer analysis.
- Link model outputs to business metrics such as forecast accuracy, bias, and service level.
- Work with planners and business teams to turn exceptions, backlog, or service issues into data features or analytical logic.
- Ensure version control and reproducibility through GitHub and configuration files.
- Support integration of model results with dashboards and decision support tools.
- Partner with data engineering to validate data pipelines and maintain data quality.
- Collaborate with domain data scientists to convert recurring manual logic into automated processes.
- Participate in code reviews and documentation to ensure consistency and knowledge sharing.
- Identify redundant or manual processes and refactor them into shared functions or libraries.
- Contribute to internal best practices for model reproducibility, documentation, and analytics transparency.
Required Skills And Experience
Strong programming skills in Python, SQL, Databricks, GitHub, and MLflow. Experience with Pandas, PySpark, and Scikit-learn. Demonstrated ability to build regression, classification, or time-series models for business applications. Familiarity with feature engineering techniques such as lag, rolling, and categorical encoding, and awareness of methods to prevent data leakage. Experience with workflow automation, reproducibility, and parameterized scripts. Ability to explain analytical results and model behavior in business terms.
Nice to Have
Knowledge of supply chain or planning data such as forecast, backlog, and inventory. Experience with Power BI or similar visualization tools.
Who You Are
You are a problem solver who enjoys turning analytical ideas into practical solutions. You are comfortable working across data and business contexts and can connect statistical concepts to operational outcomes. You are methodical, organized, and eager to help others build analytical and coding capability.
