I built a full-stack optimization engine that modeled how Arizona data centers could shift compute loads and cooling modes using real-time grid and weather data. Using Python, Pyomo, GLPK, NumPy, pandas, and PostgreSQL, the system evaluated trade-offs among cost, electricity use, and water consumption. The project achieved roughly 12.6 percent cost reduction and an estimated 467 million gallons of water savings annually in the competition scenario. This experience taught me how optimization, data engineering, and backend development work together to solve large-scale infrastructure problems.