I need an accurate, real-time forecasting solution that warns me the moment flood conditions begin to form. My focus is strictly on severe weather alerts rather than broad climate reporting, so the system must zero in on flood likelihood and intensity. Here is what I expect: • Data & Modeling – Pull live feeds from trusted meteorological and hydrological sources (NOAA, ECMWF, radar, river gauges, etc.), run short-lead models, and translate the output into clear flood-risk probabilities up to 48 hours ahead. • Alert Delivery – The instant your model crosses a predefined threshold, I should receive an alert (email, SMS, or API push). Alerts must arrive within five minutes and include probability, expected start/end times, and a confidence score. • Ongoing Visibility – Provide a lightweight dashboard or daily brief so I can track evolving risk on quiet days and verify model performance after each event. • Tech Stack – Python, GIS tools, and cloud services such as AWS or similar are preferred, but I am open to any robust setup that meets the timing and accuracy targets. When you reply, outline the datasets, algorithms, and validation methods you plan to use, along with a clear timeline for a working prototype.