Worldsphere.ai is the first AI startup laser focused on improving intensity predictions for tropical cyclones. 

Existing hurricane models are not very good at predicting rapid intensification. Rapid Intensification (RI) is becoming increasingly common because of rising ocean surface temperatures.
As we saw with Hurricanes Michael and Harvey, one day of rapid intensification can change the impact of a landfall event from minor to devastating. The number of people who die or are injured due to not evacuating on time can be significantly reduced if the strength of the hurricane is accurately predicted, especially when it is close to landfall. 

Deep learning has shown meaningful progress over traditional hurricane models in predicting RI. But it still has a long way to go. There is a clear opportunity to bring together the best minds in atmospheric science, meteorology and AI to attack this problem. Current hurricane models are run using supercomputers that utilize hundreds of thousands of CPUs and require huge energy consumption. Only a few wealthy countries can afford this setup! Deep learning based models can run on a laptop at home with a good GPU.

Worldsphere.ai

Org chart

Teams

Offices