Floods are the most deadly natural disaster, affecting 250M+ people worldwide. They are most catastrophic in underdeveloped nations, where roughly 80% of fatalities occur. These countries lack sophisticated forecasting systems that can predict where severe flooding is likely. Oftentimes, people receive no warning before a flood, and have no time to prepare and evacuate. This film follows a small group of Google engineers from Tel Aviv using machine learning to design a forecasting system to predict a flood’s travel path. Using Google’s map data, computational power, and high-scale communications platforms, the team helped design a cost-effective system that yields more accurate alerts. In the film, we meet the team in India, where they’re running their first of many pilots to test their forecasting model. Their goal is to extend flood warnings from 1 day to 3, and to make this technology available for free to any country who needs it.