Vesta, an AI-native startup founded in 2020 that’s shaking up mortgage lending, has secured $30 million in fresh funding led by Conversion Capital. Other investors include Pennymac, New American Funding, Citi Ventures, and Andreessen Horowitz. The new capital follows a year of explosive growth: revenue has surged more than 12× year over year, and Vesta’s software now supports originating over $100 billion in loans annually. In total, Vesta has raised $85 million to date.
Speeding Up Loan Origination with Swarms of Agents
Mortgage loan processing in the U.S. often takes about 40 days and costs lenders roughly $11,000 per loan — most of those costs stem from manual work and delays in reviewing loan files. Vesta’s platform tackles both issues. It enables mortgage companies to deploy AI agents to handle discrete parts of the origination process. Initially, each agent works under human review; over time, some lenders allow agents to handle whole portions of the workflow autonomously — underwriting included. All agent activity is logged so decisions are auditable and compliant under regulatory rules.
Why Now Matters
The timing of the raise aligns with improvements in AI systems capable of managing complex, multi-stage tasks — something earlier generations of models couldn’t handle reliably. Vesta credits much of its recent leap to Sonnet 4.5 (developed at Claude), which has proven superior in obeying long-duration, user-specified workflows. The company also plans to introduce a mortgage-specific AI assistant that tracks workflows and performs tasks on behalf of lenders.
Among its customers are Pennymac and New American Funding. As Vesta expands, it faces competition from traditional mortgage tech providers — ones built before AI agents became central — as well as newer companies vying to do similar automation work. Vesta’s edge is that its tools are native to agents rather than layered onto legacy systems. Co-founder and CEO Mike Yu believes the firm’s current sub-5% market share is just a starting point for scaling. The goal now is to grow team, market presence, and product breadth.
AI in mortgage systems raises questions around liability: Vesta holds that underwriting decisions remain the lender’s responsibility. Its system records all inquiries, decision paths, and agent outputs for regulatory or internal audit. With agents increasingly capable, Vesta argues this transparency is essential.
Upcoming product lines include the aforementioned assistant, and improvements across the platform to allow more autonomous agent operation where safe and compliant. Vesta says it aims first to win over the rest of the U.S. mortgage industry, then follow customers wherever new use cases emerge.
What this means:Vesta’s funding round and product advances mark a turning point in mortgage fintech. As AI agents grow more sophisticated, they promise to reduce both time and cost burdens baked into legacy loan processes. But the shift comes with regulatory and operational challenges: transparency, model reliability, and liability stay front of mind. If Vesta can deliver on its agent-based automation without compromising compliance, it could force legacy players to adapt or fall behind.
