This study introduces a hyperlocal tool for short-term extreme wind nowcasting (0-6hr forecasting), designed to enhance decision-making and risk mitigation for bridge operations. Bridges, susceptible to the adverse effects of extreme weather, demand reliable forecasts to proactively address potential hazards. These hazards include structural deformations from dynamic loads, ice-induced damage from freezing temperatures, vehicle overturning due to extreme winds, and reduced visibility due to fog, all of which could significantly compromise safety and structural integrity. The proposed model integrates advanced machine learning (ML) techniques with a blend of hyperlocal and global sensor data, facilitating real-time, accurate meteorological predictions. The framework functions by using real-time data from both local and global weather stations, to feed a supervised ML algorithm trained on historical data and specifically tailored for hyperlocal meteorological predictions. The system captures real-time weather conditions using local sensors situated on the target bridge - namely the 1624m main span Great Belt Bridge suspension in Denmark. Concurrently, data from a global weather station augments this localized information, providing a comprehensive view of impending meteorological threats. The ML forecast algorithm processes this data, leveraging historical records to enhance the model’s learning and predictive accuracy.