In a conversation with HousingWire’s Allison LaForgia, Indecomm CEO Rajan Nair discussed how 25 years of experience in mortgage technology and due diligence have shaped the company’s approach to AI-powered quality control (QC), the importance of repeatable results and the launch of AuditGenius | ai.
For Nair, successful innovation begins with understanding whether technology delivers meaningful improvements for lenders. “For me, the core for value is: Does it make sense for the customer, and does it deliver enduring benefits and tangible and measurable benefits?” Nair said.
That customer-first philosophy has guided Indecomm’s focus on mortgage operations, where technology and human expertise work together. “Our focus is really on the middle office and the back office. That’s where we do really well.”
Why mortgage QC requires a different approach to automation
While automation has advanced throughout mortgage origination, Nair explained that QC presents distinct challenges because auditors must evaluate work that has already been completed. “The QC auditor is expected to be at a higher skill level than even the underwriter,” Nair said. “What’s expected from them in terms of the result, that quality essentially is the key.”
For Nair, the challenge isn’t simply whether AI can perform an audit, but whether auditors will trust the results enough to incorporate the technology into their workflows. “Can you get an automation solution that the auditors can trust, embrace and adopt?”
Rather than automating traditional checklists, Indecomm has focused on connecting QC processes to underlying mortgage requirements, using its own QC operations to validate new capabilities. “A lot of what is out there in the market has focused more on trying to automate a checklist. We’ve gone away from that.”
Building trust through consistent AI results
Consistency is especially important in a highly regulated industry where auditors must defend their conclusions to lenders, investors and regulators. “If you cannot replicate that same loan, the same outcome, I mean, you basically just completely lost trust.”
Nair noted that auditors naturally approach their work with skepticism, making repeatable results essential to AI adoption. “The auditors are inherently suspicious people. I don’t mean in a negative sense. They’re forensic in their attitude and their approach.”
He cautioned that AI solutions that rely primarily on foundational models and checklists may show promising test results but fail to deliver consistent performance in production. “It works in a lab environment, but as you start releasing it, you don’t see the repeatability and the consistency of the results.”
Controlling AI starts with better data
When evaluating AI-powered QC platforms, Nair encouraged lenders to look beyond the technology itself and examine how information is extracted, structured and processed. “The data is often the most ignored aspect of any automation in our industry. The data is not uniform.”
Indecomm addresses that challenge through IDXGenius, its AI-powered data extraction solution, which supports the company’s broader automation strategy. “We extract over 17,000 data points across a single mortgage, but it is that too is repeatable and predictable.”
Nair also emphasized that AI should perform much of its work before an auditor begins reviewing a loan, rather than generating new outputs during each review. “You want to do all the AI work upfront for the authoring or the preparation of the system. You want minimal AI running in production, live in real time.”
That controlled approach is central to Indecomm’s newly launched AuditGenius AI platform, which builds on the company’s established QC technology. “AuditGenius has been in production for 20 years. This is really the true, completely AI-first model, the business model that we’ve deployed.”
Making every audit finding explainable
Beyond repeatability, lenders must also understand and demonstrate exactly how an AI-powered audit reached a particular conclusion. “They should be able to show them a glass box,” Nair said. “You need to be able to get down right down to the document, to the page, to the field.”
That transparency extends beyond identifying errors. Nair said lenders should also be able to explain cleared findings and potential false positives. “Every aspect of the QC that has been done, you must be able to drill down to the last minute data field, essentially.”
When asked how quickly lenders should be able to produce that information, Nair’s response was straightforward. “Instantaneously.”
AI’s future is about expanding human capabilities
Looking ahead, Nair believes AI’s greatest opportunity lies in helping mortgage professionals accomplish more, rather than focusing exclusively on reducing headcount. “There is going to be a harmonious existence with AI between the AI models and the human. The human is going to be able to do a lot more than they were able to do before.”
As AI continues to evolve, he sees opportunities for lenders to increase productivity, improve decision-making and develop products that traditional mortgage technology has historically constrained. “These AI models are just going to keep getting more and more powerful. They’re going to be able to do things that humans would take several minutes to do. They’ll take seconds.”
“The mortgage industry is constrained by its traditional technologies, and it cannot deliver more products. I think everyone will benefit eventually.”
