The Pulse
Kastle Raises $24M to Put AI Workers Inside Lending Systems
Kastle raises a $24 million Series A led by Insight Partners to expand its AI workforce platform for banks and lenders. The company says its agents have processed more than $1.8 billion in transactions while operating across existing lendin
AI.info Team ·
Kastle says its AI agents have processed more than $1.8 billion in transactions, a figure the consumer-lending startup is using to support a larger claim: banks can deploy software workers inside existing operational systems without replacing the infrastructure those institutions have spent years building.
The San Francisco company announced on September 17 that it raised $24 million in Series A funding led by Insight Partners. Y Combinator and Commerce Ventures, which backed Kastle previously, continued their participation. Fifth Wall, along with a group of founders and financial-services executives, joined as new investors.
Kastle plans to use the money to expand its engineering, product and go-to-market teams, deepen its platform and accelerate deployments with banks and other financial institutions across North America. The company’s initial focus is consumer lending, where loan inquiries, payments, servicing and collections often move between legacy systems, contact centers and manual operating teams.
Kastle announced the financing in a company release distributed through PR Newswire.
A $1.8 billion test of AI inside lending operations
Kastle is not pitching its platform as a replacement for a bank’s core systems. Its agents are designed to work across existing infrastructure, execute high-volume workflows and keep systems of record current. The company says that approach lets institutions add AI capacity without undertaking a multiyear migration to a new technology stack.
The platform is aimed at work that lenders already perform through established processes and controls. Kastle describes a hybrid model in which AI agents handle repeatable activity while employees take on cases requiring judgment, expertise, empathy and customer relationships.
That distinction matters in lending because the work is both repetitive and regulated. A system that answers a borrower’s question but cannot update the relevant record, follow an approved procedure or transfer a complicated case to a person would add another layer of software rather than remove work from the operation.
Rishi Choudhary’s “third path” for banks
Rishi Choudhary, Kastle’s co-founder and chief executive, framed the company’s proposition as an alternative to two unattractive choices: accepting the limitations of legacy operations or replacing the systems that run the business.
“For years, enterprises have effectively been presented with a false choice: accept the limitations of legacy operations or endure a long and risky replacement of the systems at the heart of their business,” said Rishi Choudhary, Co-Founder and CEO of Kastle. “Kastle creates a third path. We give financial institutions an AI workforce that can operate across the systems they already have, so they can capture the benefits of AI now—not five years from now.”
Choudhary said Kastle believes its model can become a way for large institutions to adopt AI while keeping existing operational controls in place. The release does not identify the banks using the platform, but it says Kastle is already powering work across some of the largest enterprises and banks in the world.
Agents for servicing, collections and compliance
Kastle’s website lists specialized agents for loan-officer assistance, collections, compliance, servicing and loss mitigation. The company says its agents can resolve borrower inquiries across voice, SMS, email and chat, and complete approved transactions rather than simply route customers into a queue.
The platform connects with a lender’s contact-center, payment-processing and loan-management systems. Kastle says its agents map to existing standard operating procedures, while monitoring tools track performance and customer satisfaction and allow teams to make updates without code.
Compliance is central to the product’s pitch. Kastle identifies the Fair Debt Collection Practices Act, the Truth in Lending Act, the Real Estate Settlement Procedures Act and the Telephone Consumer Protection Act among the regulatory frameworks incorporated into its platform. The company’s site also says its agents are designed to keep work inside workflows that lending teams have approved.
Insight Partners backs a narrower enterprise AI bet
Rebecca Liu-Doyle, a managing director at Insight Partners, said financial institutions need systems that complete work rather than create another interface for employees to manage.
“Financial institutions do not need another layer of software that creates more work for their teams. They need AI that can reliably complete the work while ensuring compliance,” said Rebecca Liu-Doyle, Managing Director at Insight Partners.
Insight’s investment places Kastle among a group of startups targeting the operational bottlenecks that make financial-services AI difficult to deploy. The company is choosing a narrow starting point—consumer lending—rather than presenting a general-purpose agent platform for every department in a bank.
Kastle’s earlier company profile from Y Combinator describes the startup as founded in 2024 by Choudhary and Nitish Poddar. That profile says the company began with AI voice and SMS agents for mortgage lending, including payment collection, loan questions and qualification of new inquiries.
The next expansion is measured in deployments
Kastle’s funding announcement gives no valuation, revenue figure or customer count. It also does not disclose how many agents are active or how the reported transaction total is distributed across lenders and use cases.
The company’s immediate plan is more concrete: hire across engineering, product and sales, improve the platform and bring its agents to more banks and financial institutions in North America. For Kastle, the financing turns the $1.8 billion processed through its systems into a case for expanding from individual lending workflows toward a broader AI workforce for banking operations.