Building on the strength of its strategic partnership with Google Cloud, digital mortgage technology provider Roostify is offering a new service that uses machine learning to help lenders streamline the underwriting process by automating document classification, data extraction, and data validation.
As the “first of many” application programming interfaces (APIs) Roostify says it’s developing for a new API platform, Roostify Document Intelligence Service is available to any lender — even if Roostify isn’t their mortgage technology partner.
“Document processing may not be the shiniest, or most exciting piece of technology imaginable for improving the mortgage process,” said Roostify cofounder and CEO Rajesh Bhat, in a blog post. “However, as we look towards a truly digital mortgage in the future, the operational efficiencies this kind of technology enables serve as a foundation, and an absolutely fundamental one, for future innovations.”
In the decade ahead, Bhat expects that innovations in “AI, machine learning, and other technologies will lead to a home lending industry that is virtually unrecognizable from today’s, and the foundations that we lay now are just the start of that process.”
Harnessing Google Cloud’s Lending DocAI machine learning platform, Roostify Document Intelligence Service starts out by classifying the documents uploaded by borrowers and loan officers in support of mortgage applications, providing instant feedback if they upload incorrect or illegible documents.
RDI Service uses artificial intelligence and machine learning to extract and validate information from more than two dozen of the most common borrower-provided documents, including paystubs, asset statements and tax forms, with an average confidence level of more than 90 percent.
By the end of the year, Roostify expects to have more than 50 mortgage document types covered, Bhat said, helping lenders eliminate steps, expedites the loan eligibility process, and deliver more pristine data to the loan origination system.
Because it’s accessed via API, lenders can integrate RDI Service anywhere in the underwriting process that works for them, the company said.
Roostify’s head start with Google Cloud’s Lending DocAI
Even before Google Cloud made Lending DocAI “generally available” to lenders in April, Roostify got a head start as a Google Cloud strategic partner last fall.
In announcing a Google Cloud Technology Partner of the Year Award this summer, San Francisco-based Roostify said it had captured more than 10 million documents through its digital lending platform, “with monthly averages continuing to rise.”
“AI and machine learning will continue to be a key component of Roostify’s ongoing efforts to meet the growing demand for innovative products and services that can enact meaningful improvements on lenders’ operations,” the company said.
In a blog post detailing the collaboration between Roostify and Google Cloud, the companies said Roostify is also using Lending DocAI to power Roostify Beyond, “a robust suite of AI-powered solutions that enables home lenders to create intelligent experiences from start to close.” The companies said Roostify Beyond is being piloted by early adopters, and will be rolled out later this year.
Cloud providers Amazon Web Services and Microsoft Azure have also developed proprietary AI and machine learning capabilities that can be customized for lenders.
Black Knight has teamed up with AWS on an AI-powered mortgage solution, Underwriter Assist, which extracts data from documents like W2s and pay stubs, and then uses AIVA, Black Knight’s artificial intelligence solution, to analyze it.
Finastra has partnered with Azure to make its core banking solution, Fusion Phoenix, and its mortgage solution, Fusion Mortgagebot, available on the cloud. Last month Finastra announced it’s moving the rest of its lending solutions to the company’s private Azure cloud, where they’ll be accessible through a portal, Total Lending Home.