AI wildfire risk models used to underwrite California wildfire reinsurance and commercial excess-and-surplus property insurance.
Kettle runs three machine learning models — ignition, spread and building vulnerability — over roughly 130 terabytes of satellite, weather, real estate and utility data, extracting about 70 risk indicators to simulate wildfire outcomes at parcel level. The company states its models identified 97.23% of California's burned properties over a four-year backtest. It deploys that view of risk through parametric wildfire reinsurance capacity and a commercial E&S property book, with 'Fire in a Parcel' as its flagship offering, and runs a team of roughly 25.
- Headquarters
- United States
- Employees
- 11–50
- Business model
- Services
- Ownership
- Venture-backed
No queue positions recorded under this name. Expected for software, capital and supply chain companies, which do not hold interconnection requests.
- · Curated research
Written profiles are hand-entered and not independently verified. Queue figures are computed from ISO interconnection data and reflect requested capacity, not built capacity.