Our Technology
Kettle outperforms the industry by using proprietary machine learning algorithms that use billions of lines of weather, fuel, satellite, anthropogenic, and ground truth data.
Contact Us
Kettle is an insurer built for the modern world
Kettle outperforms the industry by using proprietary machine learning algorithms that use billions of lines of weather, fuel, satellite, anthropogenic, and ground truth data.
We use groundbreaking technology to deliver better, more affordable protection to people and their communities
This includes data from NASA, NOAA, and the European Space Agency.
We process the collected raw data using an extract, transform, and load (ETL) pipeline to build our deep neural network models.
This model generates over 2 million wildfire footprints at a 100 meter resolution.
Our proprietary models capture the temporal and spatial correlations of risk as well as their physical drivers.
Kettle’s technology includes a family of models- the Ignition Model, Contagion Model, Building Vulnerability Model, Fire Edge High Resolution Downscaling Model & Parcel Risk Assessment Model to predict wildfire risk.
Using 42 million grid-level high resolution simulations to generate wildfire ignitions and spreading patterns, Kettle accurately captures the distribution of wildfire risk and property damage.
Using the Ignition Model, Kettle divides California into 419,000 micro grids - each 0.5 square mile.
We train our Genesis Model by discretizing the state of California into grids and analyzing the factors that contribute to a wildfire ignition in a specific month.
Wildfire Spread PaperThe accuracy score of Kettle’s Genesis Model is
89.2%
Kettle’s spread model uses AI to emulate the physics and randomness of a wildfire event.
Our deep long short term memory convolutional neural network captures both temporal and spatial correlation of wildfire spread, leading to its state-of-the-art performance in predicting the spread of wildfire.
Wildfire Spread Paper