Hutsine Callicutt

Results-driven logistics professional with a proven track record at Amazon, consistently exceeding productivity targets and maintaining excellent order accuracy. Skilled in data analysis and process improvement, leveraging strong communication and teamwork abilities to enhance operational efficiency and customer satisfaction. Proficient in Python and supply chain analytics.

• Supply Chain Transportation Cost Optimization,Team Lead, Led a 4-member team in developing a transportation cost model for refrigerated seafood distribution across the United States.
• Coordinated project communication through GroupMe, assigned responsibilities, monitored milestones, and developed contingency plans to address project delays. Modeled transportation costs for 40,000-pound refrigerated truckload shipments across 50 U.S. states. Calculated cost-per-pound metrics for pricing, profitability, and landed-cost analysis. Evaluated hub-and-spoke distribution strategies to improve logistics efficiency and reduce transportation costs.
• Machine Learning & Operations Performance Analysis, Evaluated machine-learning-driven task assignment systems within a fulfillment center environment. Analyzed productivity data showing associates averaging 92 tasks/hour compared to machine-generated targets of 100 tasks/hour. Identified operational performance gaps ranging from 10–18 seconds per task. Developed recommendations targeting 95% compliance with productivity standards. Proposed coaching and workload-balancing strategies to improve operational performance.
• Statistical Learning: Employee Promotion Prediction Analysis, Analyzed a workforce dataset containing 100,000+ employee records. Evaluated 5 promotion-related predictors using multiple linear regression models. Conducted hypothesis testing, coefficient analysis, and regression diagnostics. Assessed model performance through R² analysis, residual analysis, and data visualization. Recommended alternative modeling approaches based on dataset limitations.
• Gas Price Prediction During Geopolitical Conflict, Built and compared 4 predictive models including Linear Regression, Lasso, Splines, and Random Forest. Analyzed gasoline price data across 50 U.S. states. Applied 5-fold cross-validation and RMSE evaluation methods. Improved model performance through feature engineering and trend analysis. Demonstrated the impact of data leakage on predictive modeling outcomes.