Roostify has launched the Roostify Document Intelligence (RDI) Service for lenders – the first available service on its API platform – to instantly process and contextualize mortgage-related documents utilizing machine learning.
Buying a home requires a borrower to submit substantial and detailed paperwork to support the loan application, making the current manual process cumbersome, error-prone, expensive and difficult to scale. The lending industry spends millions of dollars each year processing these documents which include W-2s, pay stubs and borrower asset statements.
The RDI Service uses machine learning, developed in partnership with Google Cloud, to instantly process documents required for mortgage underwriting and deliver relevant data as needed to streamline the rest of the lending process. The service has three main capabilities: document classification, data extraction and data validation.
Through document classification, the service processes a document to identify the document type and confirm whether the document satisfies an associated condition generated by a loan origination system, automated underwriting system, or rules engine. With its data extraction capabilities, the RDI Service extracts the data contained in the document and returns it as human-readable, usable data via API. The data validation service matches extracted document data against loan application data to identify any potential inaccuracies, errors, or inconsistencies.
The Roostify Document Intelligence Service is readily available to all lenders as well as mortgage service providers and technology providers. It can support an infinite number of use cases benefitting from automated identification and extraction of data contained within the mortgage document set.
“Providing new services that utilize artificial intelligence is a big step towards improving the way borrowers experience the home buying journey, and the way lenders can simplify a process that despite rapid digitization is still cumbersome and complex,” says Rajesh Bhat, co-founder and CEO at Roostify.
RDI is currently able to process more than 25 of the most common borrower-provided documents (e.g., paystubs, asset statements, tax forms), and by the end of the year, will have upwards of 50 mortgage document types covered. Across the entire document set, the average confidence level is greater than 90%, with 35+ document types having a confidence level of over 95%, without manual review.
“With mortgage lending becoming increasingly competitive, every minute and dollar counts for both lenders and homebuyers,” adds Bhat. “As we look into the next decade, 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. The operational efficiencies the Roostify Document Intelligence Service enables serve as a foundation, and an absolutely fundamental one, for future innovations.”