Artificial intelligence has become the mortgage industry’s most powerful and most misunderstood new ally. While some lenders chase hype, others are quietly transforming operations with measurable, data-driven automation and AI. Few leaders bridge both the business and technology sides of this transformation as fluently as Rajan Nair, CEO of Indecomm, a provider of the Genius AI suite that powers intelligent automation across the loan lifecycle. In this conversation, Nair discusses the difference between hype and habit, why automation and AI is a marathon, and how lenders can build trust in AI, one workflow at a time.
HousingWire: Many lenders say they’re “automating” and “leveraging AI” but progress still feels incremental. What’s holding lenders back?
Rajan Nair: Lenders often have a dozen or so different systems, and many don’t talk to each other. Adding automation or AI without standard processes or connected systems can make things even more confusing.
Also, a lot of lenders are cautious because past transformations have gone over budget and caused new problems. This is understandable but unfortunate because if you don’t use it at all, you are likely going to fall behind. I think it is important to get past that fear.
One way to tackle both challenges is to map out how data flows through the origination process and across all systems. Seeing the full picture makes it easier to fix gaps and streamline operations. Then, the addition of AI will make sense and make an impact.
HW: You’ve mentioned research showing most AI pilots fail. How does Indecomm avoid that trap?
Nair: We start with process design, not model design. We map how the loan truly moves through the lender’s ecosystem, identify friction points, and automate those slices first.
We also own the “last mile,” focusing on the 20% of work that models often miss. Because 20% of missing documents doesn’t equal 20% of effort, it equals half your cost. We commit to that last-mile accuracy so our clients see real ROI, not theoretical results.
HW: What’s one area where you think GenAI will quietly make the biggest impact in the next year?
Nair: GenAI has endless possibilities, but I think its biggest impact will be in enhancing loan decisioning and reviews across the entire loan lifecycle, from pre-funding through post-closing, servicing, and capital markets. Gen AI isn’t just processing data; it’s spotting patterns and highlighting potential issues that might otherwise go unnoticed. This gives lenders a smarter way to approach underwriting and loan reviews, turning insights into action. When you add pragmatic AI to the mix, it delivers practical, reliable solutions that improve processes and outcomes.
HW: You’ve described Indecomm’s approach as “glass-box AI.” How do you make that real for users?
Nair: Transparency builds trust. Our Genius suite of automation solutions allows the user to have full visibility into the data points and documents that influenced the output like a condition or a finding. This level of visibility is critical in driving adoption. We don’t expect users to trust the machine blindly. Instead, we show them why it reached that conclusion. When they can see the logic and verify the source, confidence follows naturally.
HW: When you talk with CEOs about AI strategy, what separates successful adopters from those who stall out?
Nair: The successful ones assign ownership. They don’t delegate AI to IT; they make it a business transformation. They define metrics, touches per file, decision time, defect rate, and measure relentlessly. For those who stall to embrace AI should be mindful that you can’t bolt AI onto a broken process. You have to fix the process first, or automation just runs faster in the wrong direction.
Something to keep in mind: AI and automation aren’t events; they are conditioning. The lenders that win are the ones that treat AI and automation adoption as part of their operating rhythm, not a one-off project.
HW: Agentic AI is the newest buzzword. What role do you see it playing in mortgage?
Nair: Agentic AI is pretty powerful and has real potential inside guardrails. Imagine a virtual assistant that tracks conditions or proactively flags missing verifications. That’s a productivity gain. But we’re not ready for agents making credit or compliance decisions. For now, think of agentic AI as an assistant, not an approver. The human remains the decision-maker; the AI just brings them better information, faster.
HW: Looking five years out, what’s your biggest prediction for AI in mortgage banking?
Nair: We’re heading toward componentized lending. Every stage, from application through closing, will operate as an interconnected API service. GenAI will run quietly in the background, summarizing, classifying, and reasoning across those services. The winners won’t be the loudest “AI-first” companies; they’ll be the lenders who built resilient, data-driven workflows that can evolve with technology.
HW: Final thoughts for executives who are still hesitant to take the leap?
Nair: Inertia is the real risk. Waiting for perfection means missing the window. Start small, measure everything, and design for transparency. AI won’t replace people but people who know how to use AI effectively will replace those who don’t. This is the beginning of a brave new world for mortgage banking, and it’s powered by data, trust, and accountability.
