Skip to content
Sei
All Posts
Pranay Shetty

Pranay Shetty

CEO & Co-Founder

74 articles

Mortgage

When Rates Drop, Your Servicing Portfolio Gets Called by Someone: Refinance Recapture With an AI Voice Agent, and the Tests That Decide Who It Calls

A rate move puts thousands of loans in the money in the same week, and the servicer has the data to reach those borrowers first. Before an AI agent dials any of them, it has to know which loans can actually close: VA's seasoning and recoupment tests under 38 U.S.C. 3709, the early payoff window on loans you sold, which consent covers a marketing call, and what the agent may and may not say about savings. What we learned running a voice agent on a lender's funnel, applied to the recapture campaign.

6 min read
Read more
Servicing

Your Subservicer's Calls Become Your Exam Findings: Overseeing a Mortgage Subservicer When You Can Review Every Call and Letter

A bank or IMB that owns servicing rights and outsources the work keeps the regulatory liability, and most oversight programs still run on the subservicer's own scorecard, an annual audit and a sampled call file. What changes when the owner can score every call and every letter against Regulation X, Regulation F, UDAAP and its own procedures, what the contract has to say first, and what we learned codifying hundreds of SOP documents for a lender's call review.

5 min read
Read more
Servicing

Curtailment Is Written Against Dates: Preparing FHA and Mortgage Insurance Claims With an AI Agent in Post-Default Servicing

By the time a defaulted loan reaches the claim, every mistake made since the first missed payment shows up as a number someone deducts. A claim package is hundreds of pages of servicing history, invoices and legal fees that have to agree with each other, and FHA's curtailment rules punish the dates that do not. What we learned running a claims agent in production for a post-default claims outsourcer, where it helps, and why a specialist still files.

6 min read
Read more
Mortgage

The Mortgage Lead That Arrives at 9pm on a Sunday: Speed-to-Lead With an AI Voice Agent, and the Lines It Does Not Cross

Calling a web lead back in under a minute is the easy part. Doing it inside federal and state calling hours, on consent you can prove, without the agent drifting into loan-originator activity or qualifying the 9pm lead differently from the 9am one, is where speed-to-lead programs get into trouble. What we learned putting a voice agent on both ends of a top-25 IMB's funnel, including the 9.2x jump in booked appointments and what it did to the loan officers' calendars.

6 min read
Read more
Servicing

When a Disaster Hits the Whole Portfolio on the Same Day: Running Natural-Disaster Forbearance With an AI Servicing Agent

A hurricane does not deliver its delinquencies one at a time. A FEMA major-disaster declaration can put thousands of loans into distress in the same week, and the servicer still has to offer each borrower the right relief, hold every Reg X protection, and keep the collateral and the escrow straight while the call volume triples. How we build a servicing agent for the event that breaks a queue-based operation by arriving all at once.

6 min read
Read more
Mortgage

The Loan That Missed Payment One Was Clean When You Shipped It: Early Payment Default, Early Payoff, and the Feedback Loop Your QC Program Is Missing

A loan that defaults on its first payment, or pays off in month four, is a loan your pre-funding QC already passed. Both are post-funding signals that something upstream was wrong, and most QC programs never route them back to the decision that let the file through. A production view of how we turn early payment default and early payoff into a closed loop that corrects the agent, not just the file.

7 min read
Read more
Servicing

The Lien Release Clock: Why the Last Step in Servicing Is the One That Gets Servicers Sued

A paid-off loan is a closed file to the servicing floor and an open statutory deadline to fifty state legislatures. Most states give the servicer a fixed window, often 30 to 90 days, to record the satisfaction or reconveyance, and many attach a per-violation penalty plus the borrower's damages and attorney fees. The deadline runs per loan, per state, per recording jurisdiction, which is exactly the kind of obligation that falls through a servicing operation's cracks. How an AI agent tracks the release clock to the county and proves it met it.

5 min read
Read more
Mortgage

When an AI-Generated Pre-Approval Letter Becomes an Application: The Reg B Line Digital Lenders Keep Crossing

A pre-approval letter is a sales asset to the growth team and a legal event to the compliance team, and an AI agent that issues them at scale sits exactly on the seam. Whether the letter is a prequalification, a preapproval, or an application changes what the lender owes the borrower: an adverse action notice on a decline, a Loan Estimate in three days, a fair-lending record of the criteria applied. How we build the agent to know which of the three it just produced, and to never issue the strong letter off the weak file.

5 min read
Read more
Compliance

The Exam Starts With a Document Request, Not a Conversation: Preparing an AI-Assisted Mortgage Operation for Supervision

When a CFPB or state examiner opens a supervisory exam, the first thing you get is an information request, and an AI-assisted operation has to answer questions the exam procedures were written before anyone was asking. What examiners actually request, how the modules in the CFPB manual map onto an automated file, and the evidence that lets you answer "how did the machine decide this" in a sitting instead of a scramble.

6 min read
Read more
Mortgage

You Staffed a Desk, Not a Tool: What Happens to the Underwriter When the Agent Does the File

The hard part of a mortgage AI deployment is not the model. It is the org chart. When an agent clears most of the routine work, the processor and underwriter roles change shape, the exception queue becomes the real job, and someone has to own the confidence floor as a policy decision. What we have seen change on the floor, and why the oversight role is a control an examiner expects to see named.

5 min read
Read more
Servicing

Most Servicing Calls Are the Same Five Questions: Answering Them With an AI Agent Without Turning an Answer Into a Reg X Notice or a UDAAP Problem

The bulk of a servicing call center's volume is a handful of repeatable questions: my payment, my escrow, my payoff, my statement, my options. They are automatable, and each one has a trapdoor, because a casual answer can misstate what a borrower owes or walk past a sentence that was legally a notice of error. How we let an AI agent handle the repeatable volume while recognizing the moment a question becomes an obligation with a clock.

5 min read
Read more
Mortgage

A Mortgage Takes Forty-Odd Days and Spends Most of Them Waiting: Where the Time Actually Goes, and the Days an AI Agent Gives Back

The industry still measures time to close in the mid-forties of days, and very little of that is work. It is a loan sitting in a queue between two people who are each busy with a different file. A production leader's map of where the dead time hides, which regulatory clocks keep running while the loan waits, and which gaps an AI agent closes without touching a decision anyone is accountable for.

6 min read
Read more

You Ain't Seen Nothin' Yet

Book a Demo

Pack up some of your complex historical files — any loan type, any investor. We run them through intake, income and condition clearing, and in 30 minutes you see every condition we created and cleared efficiently for your own team, and why.

  • Any loan type, any agency guideline or custom investor overlays.
  • Every finding cited to the guideline or document it came from

Case study highlight

Loan review fell from 4 hours 50 minutes to 47 minutes per file, and 78% of items now clear at high confidence — so QC touches only the exceptions.

Pre-Close QC Agent · Better Mortgage— read the case study

Please provide your full name so we know how to address you.

Tell us which company you represent so we can personalise our response.

Use your work email. One of our experts will get back to you.

Which desks would you like to discuss?*

Choose the desks you’d like us to cover. Pick “Not sure yet” if you’d rather we worked it out on the call.

Roughly, so we bring the right person to the call.

Complete the verification to submit the form.