Developed predictive models to analyze turnover among 12,000+ employees, focusing on high-risk salaried leadership roles within rapidly scaling renewable energy projects.

Engineered 70+ features from HR records, such as promotion and rehire, and used Random Forest and Logistic Regression to distinguish voluntary vs. involuntary exits with actionable insights.

Created an interactive Tableau dashboard to flag at-risk employees by project, tenure, and promotion history to empower HR to take proactive retention steps.

Recommended structured onboarding, mentorship programs, and ways for employees to move within the company.

Tools: Python, Scikit-learn, Pandas, Tableau, Seaborn, Excel