Kalepa raises $75 million to grow insurance AI with task-by-task human oversight

Kalepa has secured $75 million to bring AI into more insurance work, but its pitch stops short of handing every decision to machines. The Series B, led by Updata Partners at a reported $470 million valuation, backs software that adjusts its independence to the task and brings in people when judgment or approval is needed.

Existing investors IA Ventures and Inspired Capital also participated. Kalepa says the proceeds will fund a broader AI product portfolio, entry into additional markets and a doubling of its workforce during 2027. The money is intended to take its technology into more parts of insurers’ operations, beyond the underwriting decisions already central to its business.

Fresh capital after five years

The New York company last raised a $14 million Series A in 2021, led by Inspired Capital with IA Ventures participating. The new financing brings its total funding to $91 million. That five-year gap separates this investment from the earlier round; it is not a rapid follow-on financing.

Kalepa says it has been cash-flow positive for more than two years. It also reports that revenue and its client base each tripled over the preceding year, while the volume of work processed through the platform grew fourfold. Those growth figures describe the business entering the financing, rather than results of the expansion the new money will support.

Its customers include W. R. Berkley, Bowhead Specialty and AmRisc. Kalepa serves insurance carriers as well as managing general agents, reinsurers and program administrators. The funding therefore supports software sold to businesses handling insurance risk, pricing and portfolios, rather than a consumer insurance assistant.

Autonomy assigned by task

Kalepa describes a platform spanning operations, underwriting, pricing and portfolio management. Underwriting is the work of assessing a proposed insurance risk. Its system coordinates AI agents—software carrying out multi-step tasks—and determines how much independence each task receives according to its risk and complexity. People remain involved where expertise or approval is required.

A Bowhead Specialty case study described by Tech Funding News gives a concrete example. When a broker submits a proposed policy, Kalepa gathers research, data and underwriting guidelines into one workspace and flags missing information. That places the software inside the preparation and review of an insurance decision, not just the drafting of a response.

The company also says outputs are checked and the work remains traceable and auditable. Those controls are part of its stated design for expanding AI’s role while preserving a record that insurance professionals can review.

Insurance runs on the judgment of experienced professionals, and our Professional Grade AI is held to that same standard. That means insurers can look at each task and ask who does it best, and at what cost. Sometimes that’s AI and sometimes it’s a person.

Paul Monasterio, Kalepa co-founder and CEO

The performance figures need care

Kalepa reports substantial customer gains, although the published accounts differ on two headline measures. FinTech Global gives 30% to 60% more premium per underwriter and a 58% reduction in quoting times. Tech Funding News instead gives more than 58% higher premium per underwriter and quotes delivered 30% to 60% faster.

The accounts agree on a reported 960-basis-point improvement in combined ratios: a 9.6-percentage-point reduction in the share of premium spent on claims and expenses. These remain company-reported customer outcomes. The conflicting assignments of the speed and productivity figures leave the precise size of those two claimed benefits unresolved.