Kalepa Raises $75M Series B at $470M Valuation to Expand Insurance AI

An insurance submission can contain a straightforward application alongside years of loss records, spreadsheets, supporting documents and information that does not quite agree. Kalepa has raised a $75 million Series B to bring more of the work surrounding those submissions into connected AI workflows, extending its platform across operations, underwriting, pricing and portfolio management.

The financing values the company at $470 million and brings total funding to $91 million. Updata Partners led the round, with participation from existing investors IA Ventures and Inspired Capital.

Announced November 5, the investment follows a year in which Kalepa says both revenue and customer count tripled, while the volume of work running through its platform increased fourfold. The company also reports remaining cash-flow positive for more than two years. It plans to expand its AI product suite, enter new markets and double headcount in 2027.

Connecting the Work Around an Underwriting Decision

Kalepa describes its product as “Professional Grade AI” for insurance. The proposition is broader than generating a summary of an application: the platform performs and coordinates work across the steps required to understand a risk, price it and decide whether to insure it.

The company’s platform overview maps that sequence from submission ingestion through clearance, triage, risk analysis, rating, quote and bind, and portfolio management. An application first has to be classified and checked for missing information or conflicts. It then needs to be assessed against an insurer’s appetite and guidelines before the team can evaluate terms and pricing.

Those stages depend on one another. An incorrect field extracted at intake can affect a later rating calculation, while a missed condition in a supporting document can change how an underwriter views the exposure. Connecting the workflow gives a system a chance to carry the evidence and context forward instead of leaving people to reconstruct them at each stage.

The funding announcement says Kalepa’s insurance-specific AI harness coordinates the agents doing this work and adjusts the level of autonomy to the task and its risk. People participate where their expertise or approval is required, with outputs designed to be traceable and auditable.

From Mixed Documents to Evidence-Linked Data

Kalepa’s submission-ingestion documentation provides a more detailed picture of the first step. The platform accepts packages through email forwarding, portals and APIs, then classifies documents and extracts risk data.

The supported material includes ACORD insurance forms, spreadsheets, handwritten notes and loss runs—the records of prior claims that help an underwriter understand an applicant’s experience. Such inputs can vary substantially in structure and quality, even when they refer to the same business.

Kalepa describes document-linked evidence, checks for consistency across documents and confidence-driven controls as part of its validation process. That is a more useful operational design than treating extraction as a single answer that must simply be trusted: the reviewer needs a route back to the source and a way to identify conflicts.

For example, when two documents describe an exposure differently, collecting both values is only the beginning. The workflow must determine whether the discrepancy can be resolved automatically or should be presented to a person before it influences a decision.

An Ensemble Engine With Controls Around Automation

On its technology page for IT and AI leaders, Kalepa describes an Ensemble Engine combining probabilistic models with deterministic logic and layered controls. It presents the platform as API-first and modular, integrating with policy-administration systems, document platforms and internal data lakes.

That combination matters in insurance. Probabilistic models can help interpret varied documents and language, while deterministic logic can enforce an explicit rule or calculation. A useful system needs to connect the two and identify when the remaining uncertainty requires an underwriter’s judgment.

Kalepa says its product can be configured around underwriting rules and risk appetite without replacing the insurer’s core systems. The technical direction is to introduce AI within an existing operating environment, rather than make a new policy-administration system a prerequisite.

That also makes governance a workflow concern. The important questions include which tasks an agent may complete, which require approval, and whether the evidence behind an action remains available to the next person reviewing it.

Growth Claims Point to Productivity and Underwriting Discipline

The announcement names W.R. Berkley, Bowhead Specialty and AmRisc among Kalepa’s customers. Its broader market includes carriers, managing general agents, reinsurers and program administrators.

Kalepa reports customer results including more than 58% greater premium per underwriter, quotes delivered 30% to 60% faster, a doubled quote-to-bind rate and a 960-basis-point improvement in combined ratio. These are company-reported customer outcomes, not uniform results promised for every deployment.

The measures describe different parts of the business. Premium per underwriter concerns the amount of business handled, while quote-to-bind measures how often a quote becomes a policy. A combined-ratio improvement relates to underwriting performance; 960 basis points equals 9.6 percentage points, rather than a 960% change.

That distinction is important because processing more submissions is not automatically the same as selecting better risks. The commercial case for insurance AI has to connect operational speed with pricing discipline and the quality of the resulting portfolio.

“Insurance runs on the judgment of experienced professionals, and our Professional Grade AI is held to that same standard,” said Paul Monasterio, Kalepa’s co-founder and CEO, in the release. He said getting the division of work between people and AI right allows experienced staff to concentrate on complex risks, close calls and broker relationships.

New Capital and Leadership for a Broader Product Suite

Kalepa’s last financing was a $14 million Series A in 2021, led by Inspired Capital with participation from IA Ventures. The new round supports expansion across more of the insurance value chain, alongside the planned increase in hiring.

The company has added Garrett Schlesinger as Chief Technology Officer after 12 years at Affirm, where he helped build its real-time underwriting system. Other senior hires include Chief Revenue Officer David Meister, formerly of Addepar; Head of Product Audrey Howell, formerly of Palantir and Butterfly Network; and Head of Talent Tim Khoo-Jones, formerly of Block and Shopify.

Founded in 2018, Kalepa is pursuing a broader enterprise AI opportunity: connecting domain expertise with software that can execute and document work. Unite.AI’s coverage of governed enterprise AI decisions examines the same need to connect recommendations with permissions, evidence and outcomes.

For Kalepa, the next stage is to make that approach useful across more insurance tasks and markets. The $75 million Series B gives it resources to expand, while the operational test remains concrete: help insurers handle more work without losing the judgment and discipline that make the business sustainable.