The news
Vesta, a San Francisco startup that builds mortgage origination software, has raised $30 million in a round led by Conversion Capital, TechCrunch reported on October 8, 2026. The company is selling AI agents, software that carries out multi-step tasks, to lenders that want to process home loans faster and more cheaply.
Lenders Pennymac and New American Funding, both Vesta customers, invested in the round, according to TechCrunch, as did Citi Ventures and Andreessen Horowitz. TechCrunch reported that the new money brings Vesta's total funding to $85 million.
Mortgage origination covers everything from taking an application to underwriting and closing the loan. TechCrunch reported that a mortgage takes about 40 days on average to close and costs about $11,000 per loan to originate. Vesta's agents automate parts of that work, offer a personal assistant for loan officers and track workflows.
According to TechCrunch, Vesta's agents can make underwriting decisions, but a human must approve them, and every action is recorded so lenders can audit it for compliance. The outlet reported that the agents run on Anthropic's Claude Sonnet 4.5 model.
Chief executive Mike Yu told TechCrunch that demand for the product has exploded over the past year and that the company plans to hire, push for market share and build new product lines. TechCrunch reported that Vesta's revenue rose 12 times year over year and that its market share remains under 5%. Vesta was co-founded in 2020 by Yu and Devon Yang, both former Blend employees, according to TechCrunch's 2022 coverage.
The numbers
- New funding
- $30 million
- Total raised
- $85 million
- Revenue growth
- 12x year over year
- Market share
- Under 5%
- Average time to close a mortgage
- About 40 days
- Average origination cost per loan
- About $11,000
Why CEOs should care
For lenders, the economics are the story. If origination runs about $11,000 per loan, as TechCrunch reported, even partial automation of document review and underwriting prep can move margins. CFOs evaluating agent vendors should ask for cost-per-loan and days-to-close data from comparable customers, not just pilot anecdotes.
Risk and compliance leaders should focus on the controls. Vesta's design keeps a human approver on underwriting decisions and logs every agent action. Ask any vendor how those logs are stored, whether examiners can review them, how fair-lending testing is done on agent recommendations, and what happens when the underlying model, here Claude Sonnet 4.5, is updated.
The investor list also matters. When customers like Pennymac and New American Funding put money into a vendor, it suggests deep commitment, but it can also tie the vendor's roadmap to its largest clients. Smaller lenders should ask how product priorities are set and whether their own needs will get equal weight.
The bigger picture
Financial services firms are moving from AI chat tools to agents that take actions inside core workflows, and mortgages are a natural target because the process is document-heavy and slow. Vesta competes with legacy provider ICE Mortgage Technology and AI-native rival Xpanse, according to TechCrunch. The company's last widely reported round was a $30 million Series A led by Andreessen Horowitz in January 2022, TechCrunch reported at the time, when total funding stood at $35 million.
What’s next
With under 5% market share, Vesta's test is whether it can win more large lenders from incumbents. Watch for new lender customers, measured results on closing times, and how regulators treat agent-made underwriting recommendations.
What “Fact-checked” means
Fact-checking means testing a story’s facts against the evidence before it is published. This story went through at least two separate checks before this version was published.
- What we checked
- Its names, figures, dates, job titles, quotes and who said what were checked against the story’s sources, including its main source where it could be opened. The headline was checked for accuracy and overstatement.
- How
- A first check reviewed the whole story. If it passed, a second, skeptical check went back to the sources to look for mistakes in the most important facts. If a check flagged the story, it was edited to fix the problems found, and a separate re-check then reviewed the whole story again.
- Who
- The checks are made by our newsroom, as steps kept separate from the writing, under rules set by our editor, Hussein Mukhtar. A story the checks still flag is held for the editor, who decides whether it is fixed, published or dropped.
- If something is wrong
- “Fact-checked” does not mean error-free. If a material error is found after publication, we correct the story and add a note saying what changed. Report an error
