The loan origination software company reports customer efficiency gains, while its financing announcement leaves valuation and measurement details undisclosed
Vesta says it has raised a $30 million Series B round, bringing its total funding to $85 million, as it promotes a loan origination system designed for lenders and AI agents to work together. The October 8 company release identifies Conversion Capital as the lead investor. The firm also led Vesta’s seed round, according to the announcement.
The financing includes lender customers New American Funding, Pennymac and nbkc bank, alongside Citi Ventures, FirstKey Mortgage, Navitas Capital, Andreessen Horowitz, Zigg Capital and Parker89, First American Financial’s venture arm. Customer participation gives the announcement a business angle beyond a conventional venture funding update: some organizations using the platform are also investing in its supplier. The release does not disclose individual investment amounts or Vesta’s valuation.
For U.S. mortgage lenders, Vesta’s pitch centers on the expense and staffing involved in processing loans. The company says lenders already on its platform, or moving to it, originate more than $100 billion annually. That figure describes the lending activity of the combined customer group, including those transitioning; it is not Vesta revenue or a stated volume of loans already processed entirely through its system.
Pennymac provides the release’s principal performance example. According to the announcement, the lender replaced its legacy origination system with Vesta and reported a 25% decrease in its operational cost to originate, along with approximately 50% efficiency gains for loan officers. The release says those results preceded activation of AI agents, an important distinction when separating platform changes from subsequent AI automation.
The supplied material does not explain the measurement periods, calculation methods or other operational changes that might have accompanied Pennymac’s migration. The figures should therefore be understood as customer-reported results presented in Vesta’s announcement, not a controlled comparison or a prediction for every lender. Nor do they establish that borrowers received lower rates or fees.
Jensen Szakaly, Pennymac’s managing director of mortgage fulfillment, described the rationale this way: “The mortgage industry has spent years adding technology on top of a manual process, and cost to originate has gone up, not down… What Vesta put in front of us was a different premise. We think the large majority of routine production work should be automated over time, and getting there requires a system built for that rather than one retrofitted for it. Vesta is that system, and our teams have adopted it well.”
Vesta also says approximately 40% of tasks completed on its platform are currently handled by AI agents and automated workflows. That combines two forms of automation rather than attributing the entire share to AI agents. A task-based percentage is also different from the share of complete loans processed automatically, which the announcement does not provide.
Cofounder and CEO Mike Yu offered a description of the product: “Vesta is the only LOS where people and AI agents work the same file in the same system… For example, when loans come in, lenders can choose to have AI agents run document review, data validation, and cross-checks. The AI agents record their decision making and escalate anything that needs a person.”
Yu’s exclusivity claim is the company’s position, not an independently established market comparison. His description identifies optional agent functions and human escalation, but the release does not detail accuracy testing, audit procedures or the criteria governing those escalations.
The financing gives Vesta additional backing for its business, while the customer examples explain its commercial appeal. The key distinction is between reported operational gains and broader promises of automation: the announcement offers evidence attributed to one customer, but leaves long-term economics and results across the customer base undisclosed.
