Data Scientist
Current•. Designed and implemented machine learning models using Python to predict inventory levels, resulting in an 8% reduction in stockouts and overstock situations.•. Used SARIMA models to automate shipment projections at the feature and factory levels for the next 2 years, improving accuracy by 15% and extending the forecasting period from 6 months to 2 years, including additional features such as individual factory projections which were previously unfeasible manually.•. Applied a combination of machine learning techniques such as Random Forest and XGBoost in Python to optimize supply chain operations, achieving a 7% increase in operational efficiency.•. Developed and implemented automated machine learning workflows that reduced the time spent on manual reporting by 40% and increased forecasting accuracy by 18%.•. Developed and maintained Python scripts for data preprocessing and feature engineering, improving model accuracy by 15% and reducing processing time by 10%.•. Adjusted 7 unique tools based on evolving needs from the supply planning team, ensuring dynamic support for ongoing and future projects, leading to a 20% increase in inventory management.•. Hosted regular review sessions with supply planning managers, fostering strategic alignment and continuous improvement, which improved supply chain responsiveness by 25%.•. Engineered complex SQL queries to extract, cleanse, and analyze large datasets from multiple sources, ensuring a 98% data accuracy rate for the supply chain inventory team.•. Designed and executed intricate SQL joins and subqueries to support comprehensive analysis of supply chain metrics, leading to a 25% increase in strategic planning effectiveness.•. Developed and maintained scalable SQL databases for inventory tracking, providing real-time insights into stock levels and turnover rates, contributing to a 15% reduction in inventory costs.
