Extreme weather events have become the norm of our day-to-day lives! We hear long running heat waves, flash floods, extreme droughts, fires, and failed monsoons. These extreme weather events have an unprecedented impact on psychological and quality of life on the population, especially the poor and underprivileged will bear most of the impact in terms of loss of economic opportunities and un-sustained livelihood. The result is increased undernourishment and food insecurity. The preparedness of future democratic governments and public distributed systems rest in harnessing prognostic markers from data using advanced analytics from past weather events. The well-prepared governments apply multi-modal interdisciplinary governmental data systems to perpetually analyze and infuse departments to be ready to tackle any potential disruptions to the lives of citizens, especially poor and marginalized parts of the societies that could fall into vicious cycle of poverty-health-and-loss of economic opportunities. The paper proposes innovative Machine Learning Models to address food security concerns to the extreme weather events.