AI agents aim to cut mortgage labor costs and speed loan closings

Vesta, an AI-native software startup, announced it closed a $30 million funding round led by Conversion Capital to expand its platform that uses AI agents to automate mortgage loan origination. Lenders like Pennymac and New American Funding participated in the round, joined by Citi Ventures and Andreessen Horowitz. CEO Mike Yu says demand for the product has exploded in the last year, and the company has helped originate more than $100 billion in loans annually.

Vesta describes its system as a swarm of AI agents that can take on tasks across the loan process. The idea is to let humans start a task with an agent, have the agent handle portions of the workflow, and then scale up as lenders gain confidence in automation. Yu notes that the dogs of the process, reviewing documents, data entry, and underwriting inputs, are precisely the kinds of repetitive tasks that agents can handle, potentially shortening the typical 40-day mortgage closing window and lowering the roughly $11,000 per loan in labor-related costs.

While Vesta emphasizes automation, Yu stresses lender accountability. “Companies remain responsible for underwriting decisions, regardless of the software or AI agents they use, and all actions and reasoning behind a decision are recorded for compliance and auditing purposes.” The technology relies on increasingly capable AI models, including a recent generation that Yu says adheres better to instructed workflows over longer time horizons.

Vesta isn’t just a single tool in a vacuum. It sits alongside traditional mortgage platforms and other AI-native contenders, competing with incumbents that aren’t built around agents from the ground up. Yu argues the advantage lies in engineering for agents from the start rather than layering AI onto legacy systems. The company’s plan now is to grow its sales and push deeper into mortgage origination, then extend as customers drive demand.

What this means for lenders and borrowers is a future where more steps in the mortgage process could be automated, potentially speeding approvals and reducing human bottlenecks. For consumers, that could translate to shorter wait times and fewer manual handoffs, though it also raises questions about how decisions are audited and how sensitive data is handled.

In practical terms, lenders will watch how quickly Vesta’s agents can reliably handle underwriting tasks, document review, and workflow management at scale, while keeping compliance and risk controls solid. The big question is whether fast closings come with tradeoffs in accuracy or reviewer oversight. For now, Vesta sees a clear path: expand the pool of customers, broaden the range of handled tasks, and keep refining agent performance with the latest AI models.

In addition to the new funding, the company is signaling intent to broaden its product lineup beyond general loan origination to include more helper roles for mortgage issuers, such as a dedicated assistant for tracking workflows and tasks throughout the loan lifecycle.