Demos: EarnUp | Kastle | Flatworld Solutions | Tavant

Immerse yourself in an engaging series of technology-focused presentations, unveiling an array of innovative products and services specifically tailored to enhance your workflows.


Transcription:

Panel Member 1 (00:09):

Well, first of all, thank you everyone. This has been so fun. So this is the final round. We're not going to play the Europe song and the final countdown, but everyone has been watching these eight minute countdowns. These folks are up there, they're working so hard to do that. This is the last round. There's going to be a round of four this afternoon to round us out for a total of 20 over two days. And then just as a reminder to everybody, it's 5.30. So when we finish this, there's a break, but out in the exhibit hall, there's going to be a riser out there and that's where the awards segment is going to take place. We're going to announce position three, position two, and of course the winner and award the winner with the $10,000 prize from LendingTree. So with that will, yes, absolutely.

(01:05):

And most importantly I will say what I said in the beginning too is that everyone that's up here is a winner because the main stage is the place. This is this one intimate forum in the entire industry where you get a bunch of tech on the main stage. If you're here, you get a hit on everything. You get caught up, see what everyone's been doing for the last year, and it's by design that way and it's just been really fun to see and hear the feedback over the last couple of days. So with that, I don't think there's any further ado and we can turn it over to Will and kick us off.

Panel Member 2 (01:38):

Absolutely. Yeah. First company to kick us off for the final round, Manish Garg, and Erik Garcia with EarnUP. Thanks.

Manish Garg (01:49):

Hi everyone, I'm Manish. And this is Erik and we are EarnUp. We are here today to talk about how lenders can maximize recapture and also help the borrowers who are very close to getting qualified but not quite there. Earn up is an AI driven payments platform that helps lenders create brand equity, brand connection with consumers and really engage with the consumers along the way. In this particular demo today, we are going to talk about Rosie, who's a consumer who's trying to seek a loan but was recently denied a loan by her bank Acme because she didn't quite meet all of the qualifications. By the way, we are a white label solution. So you can imagine that to be Acme Bank where it says up there on the top right. So Erik take it away. She comes in and says, Hey, I was recently denied a loan, what should I do next?

(02:40):

And that's kind of how a lot of consumers are. They don't understand the technicalities, they don't really understand what DTI is and what credit score is and why they get denied. They just know they got denied. So she comes in and we are able to pick up information about her and we'll talk about how we get some of this information, but right off the bat she's informed that hey look, your credit score is not quite there. You're at six 20. If you are at six 40, your chances will go up and your DT is a little bit higher, which means you have either too much debt or not enough income and you need to bring that into balance so she gets some general overview around that. What do you want to do next, Erik?

Erik Garcia (03:14):

So now that Rosie has identified, she's got two areas that she needs to focus on to Manisha's Point credit score as well as DTI. So as of now, she wants to see a full report of her personal finances, which she'll ask the advisor here.

Manish Garg (03:28):

Alright, let's do that. So she wants to dive a little bit deeper into what's going on with her finances and what she could do around there. And in the background we are connecting with multiple data sources and then we bring in all of that information in a secure private way and present that to the ai. So here you can see that just with one simple prompt and just like that she's able to get all of our debt assets, expenses, expense categories. And if you scroll up a little bit, you see these three debts, student auto credit card, let's hold onto that, put it in the box, we'll come back to that in a minute because when she wants to pay all these loans, we will refer back to these. But just I want you all to think about this for a minute, how powerful this can be for all of the consumers. Imagine doing this today manually or trying to aggregate this information from different places and trying to put this together. By the way, this UI is also rendered dynamically by the ai. This is not a predefined UI that someone's quoted. This is all created by AI on the fly. So here she gets the entire dump of her finances. So she now starts to appreciate, okay, this is my credit score and how much money do I have left over after all of my expenses? How can we help her more Erik?

Erik Garcia (04:44):

So one of the areas that I mentioned that she wants to focus on is improving that credit score. So she's going to want to ask the advisor, how do I go about doing that? Create a path for me.

Manish Garg (04:57):

Great. She's going to obviously zoom in and try to figure out how she can improve her credit score because that's the crux of what she's trying to solve for here. She does that. And remember I told you a lot of consumers don't really know what credit score is and how that works. So not only do we tell her how to improve credit score, but also what goes into it. Make your payments on time, reduce some of your debt and just some simple things can help you boost your credit score by tens of points and establish a credit history. And then it goes on to give her a little bit of an action plan as well. Directing her, Hey, check your credit report, make your timely payments and make sure you're doing all of these things. And then maybe she doesn't know what to do next. It also prompts her like, here's a question. What do you want to do next? Which of these steps do you want to take? What do you think Erik, what should she do?

Erik Garcia (05:51):

So now that we've identified an action plan, she needs to know exactly which debts should she focus on. So she's going to go ahead and ask the advisor of either the three debts, student loan, credit card, auto loan, which one should she tackle first?

Manish Garg (06:06):

Alright, so let's figure out, she doesn't know which loan should she pay and where she's going to get maximum bang for the buck and how she should deploy the capital that she might have. So she's just going to ask away and solve, let somebody else try and solve the problem for her. So she says, which should I tackle first? And by the way, this is all happening in real time. I do not know what's going to come up on the screen exactly because it's all being generated by the ai. So it says, okay, you should try to work on your credit score and you should work on your credit card and can you expand this why this is section thanks. I want to emphasize on this, we've talked a lot about explainability of models, transparency of ai. As we generate these recommendations or these responses for consumers, we are always trying to also explain why the machine is generating these responses and what's going in the line of thinking to create transparency around the responses.

(07:03):

So anyone who's consuming this information knows why it came to these conclusions. So there's a little bit of explainability built into each of the responses that is provided to the whoever's using it. So let's move forward and let's try to help her by getting her to a place where she can drill a little bit more into her credit cut scenario and learn a little bit more. And we also keep telling the consumer, Hey, please make sure before you make any critical financial decisions, you do talk to a human being if you need to talk to a human being.

(07:39):

So she wants to go in into credit card and says, you know what? You can also make a little bit of an extra payment if you want to accelerate that even more because you have the money to do that because we are able to detect it from her balances. So that's what the system's telling her, Hey, go make a little bit extra payment. You have a budget to do that and we can show you a little bit more on and the system can show her a little bit more on how to do that and that's what she wants to ask. I want to allocate additional $500, can I do it? Would it make sense?

(08:16):

So she doesn't have to put in a lot of thinking and she's trying to explore. Obviously she can try to do all of this on spreadsheet later on by herself, but now the system goes in, looks at her cash flow from the past few months, looks at how much money she has, how much extra money she has, and can she actually afford to put in the $500 And it says, yes you could and you should probably put in that extra amount to help accelerate even more. And then she says, okay, I'm ready. I want to pay this credit card before you. Okay, you hit it,

Erik Garcia (08:46):

Sorry,

Manish Garg (08:46):

But

Erik Garcia (08:48):

Time and each time,

Manish Garg (08:50):

Okay, yeah, we are running against time. So remember the three loans that we were talking about? Those three loans are automatically set up in our system. We know where to route the money from, where to make the payments to so she has student loan and credit card loan and all of that. So let's go ahead and so she can look at that and she can look at all of the details of her loan and she can onboard it, but just like any good cooking show, let's fast forward three months now she's been paying her loans and she's sort of made good on being able to pay off her debt. Three months later or four months later, the lender Acme banquet originally denied her, gets a notification, Hey look, remember Rosie, she's coming back and they click on her and see, okay, great, looks like we recommended her to go pay off her credit card and decrease her DTI. And she did that and her DTI sitting at 42, her credit score has improved and she's doing great. They call her and guess who she's going to be closing the loan with? I don't think she's going to be talking to any other lender at this point and that's how it's done. Thank you for your time and attention. Thank you.

Robin Clayton (10:01):

The payment side was amazing. I wish we had more time on that. I know that was sort of at the back of the demo. This has come up a lot today about understanding the language that consumers speak, the recurrent non-discretionary expenses. I'd be really impressed the average first time home buyer that understands what that means and maybe that's sort of tweaking just the AI prompts to put it more in language would be curious if it's actually tracking the questions that they're asking and there's a conversation log that's in the backend so the loan officer or the human advisor can sort of see where they're concerned. I think this is right for visualization instead of, oh, here's your categories of debt or your DTI, if it just had more charts and things that are a little bit more fun and interactive and clickable. I think there's some potential there.

Manish Garg (10:53):

Thank you.

Panel Member 1 (10:54):

From a branding perspective, AI advisor is spot on. And I will also just echo, I wanted to spend maybe two to three minutes on that and then the rest of the time on what it does because if I saw what I think I saw, you've got purchase, you've got refi recapture in there, you're focused on fine tuning of the DTI, which is the whole point in actually creating a loan out of it, doing debt consolidation, I might have some equity. These are the kinds of things that I think I saw on that screen, but I only got to see it for a minute, but it looked very sharp from what I did see. So that's more of a comment on demo quality and presentation, less so on the product.

Erik Garcia (11:35):

Thank you.

Panel Member 1 (11:36):

Thank you.

Christina Randolph (11:37):

Yep. Similarly, from a demo or a pitch perspective, I think initially I was confused about whether it was a POS or payments platform or debt counseling, but I think if you are wrapping all of those things into one, maybe just make it more clear about what it is in addition to tying it to the problem you're trying to solve. I liked the action plan, but I would like to see maybe more links or interaction for the borrower or the consumer to go and say, click this instead of this long scrolling like Robin said, something more digestible instead of having them scroll on endlessly. But that said, I think again, you have a good use case for LOS to help borrowers that may not be purchase ready right then and there and then kind of a loop for them to go work on things and then come back, right? So I think overall it's a lot of potential there.

Manish Garg (12:33):

Thank you.

Erik Garcia (12:34):

Thank you.

Panel Member 2 (12:34):

Thank you EarnUP. Great job.

Manish Garg (12:36):

Thanks everyone.

Panel Member 2 (12:38):

Okay, next we have Rishi Choudhary and Nitish Poddar at Kastle.

Rishi Choudhary (12:44):

Thank you. Good afternoon, we Kastle and we are building AI voice agents for mortgage servicing. Services use Kastle to collect payments, qualify new loan accounts over the phone. My name is Rishi and I'm the Founder and CEO.

Nitish Poddar (12:59):

I'm Nitish, I'm the co-founder and CTO.

Rishi Choudhary (13:04):

So how many of, I just want to pull the audience here and get a sense of who is here, how many of you are services in the room, how many are lenders, and how many are tech vendors core? So not all services. So servicing, if you're a servicer, call centers is your biggest cost, and that's a problem because call centers are expensive and they don't scale. If you're a large servicer, you're spending up to $300,000 a day employing a large team of call center agents, which could be costing you millions. And if you acquire a new loan portfolio, you need to have hire more call center agents to service the new loans that you just bought. But with Kastle, you can reduce the cost of your call centers by 90% and also give borrowers a zero hold time experience when they're interacting, when they're trying to make payments. Let's do a live demo.

(14:13):

So let's assume I'm a borrower who with Mortgage Mac as my servicer and Mortgage Mac doesn't support making partial payments online and I need to call in to make a partial payment. Let's try that again. Again. Should we try another? Yeah, do you want to try it? Yeah.

Jerome Powell, AI assistant  (15:09):

Thank you for calling Mortgage Bank. This is Jerome Powell, an AI assistant on a recorded line. Who do I have the pleasure of speaking with?

Rishi Choudhary (15:17):

Hey Jerome, this is Rishi. How are you doing today?

Jerome Powell, AI assistant  (15:22):

For verification purposes, could you please provide the last four digits of your social security number, Mr. Rishi?

Rishi Choudhary (15:28):

Yeah. 4 4 3 0.

Jerome Powell, AI assistant  (15:32):

One sec. Let me check the system. Thank you for providing that. Mr. Rishi and I got to state that Mortgage Bank gives a debt collector. This is an attempt to collect a debt and any information obtained will be used for that purpose. I see that you have a payment due of $5,000. Would you like to take care of that today?

Rishi Choudhary (15:51):

Actually my paycheck didn't come through. Can I make a payment of $3,000 instead?

Jerome Powell, AI assistant  (15:57):

Got it. Lemme set that up for you. Great. I see that you have a Chase account on file ending in 5 6 6. Would you like us to draft that for the payment today?

Rishi Choudhary (16:07):

Yeah, that'd be awesome. Thank you.

Jerome Powell, AI assistant  (16:12):

Thank you. Mr. Rishi. Your payment of $3,000 has been processed. I will also text you your payment confirmation number. Is there anything else I can help you with?

Rishi Choudhary (16:22):

No, that's all. Thank you so much.

Jerome Powell, AI assistant  (16:28):

Before I let you go, I see that due to a drop in rates over the last month, you might be able to reduce your monthly payment amount with a refinance. I would like to bring a loan officer on the line for you to discuss your options. Would that be okay with you?

Rishi Choudhary (16:43):

What are your rates like?

Jerome Powell, AI assistant  (16:48):

I'm not able to quote rates, but I can definitely connect you with a loan officer who can discuss your options and provide you with the most up-to-date information. They'll be able to go over the current rates and help you determine if refinancing is a good fit for you. Would you like me to transfer you to a loan officer?

Rishi Choudhary (17:07):

Yeah, let's do it. That sounds exciting.

Jerome Powell, AI assistant  (17:11):

Sounds good. Please hold for a second as I transfer you to one of our specialists transferring you a call.

Rishi Choudhary (17:19):

So that's what the user experience looks like. If my phone worked, Nisha's phone number would've been called by the AI agent to show that live transfer. So as you can see, the call felt really natural because we have invested a lot of time in making the AI feel very human-like, but the benefits of AI don't just stop there because AI is so observable, you can actually go in and perform quality checks. If you are a servicer, you're performing quality checks on 1% of your calls, but with AI you can perform quality checks on a hundred percent of your calls. So let's see an example of what that looks like.

(18:03):

So here we are going in to the call logs and we see that Kastle performs an audit log of every single scorecard, every single audit that the servicer has already being performed and we see that this call was a 10 on 10. You can also see, you can look inside the brain of the AI and see how it's actually making decisions. So you have a lot of control and visibility in how the AI is actually interacting with the borrower. Now let's go in to see how we actually control the workflow. If you go into the workflow tab, this is the decision tree that is followed by the AI and this is how the AI interacts with the borrower and flows through the conversation. So here we see it, recite the mini Miranda collected bank account details, process the payment and perform the upsell. And this is very configurable. If I have a new loan product that comes out that I want to advertise to my portfolio, I can quickly go into the workflow here and make that edit.

(19:15):

Another big advantage is reporting with Kastle, you get a lot of data from total calls, payment received, QA scoring, and then also why are people calling in? This was never possible. This is never possible with your human agents because it's hard to keep track of every single call. Here you can see why is a borrower calling in, what does the disposition look like and why was the call transferred? So this is the future of customer communication between financial services and borrowers where borrowers can access financial services 24 7 with zero whole time and with a trusted advisor. So when I call Mortgage Mac, the next time I don't get hit with press one to speak to an agent, I get hit with, Hey Rishi, great to meet you again. What can I help you with today? Thank you. That's Kastle.

Robin Clayton (20:25):

Wow, that was really cool. There's so many interesting things. Really appreciate how you set up the problem with the cost of the call centers as zero hold time. I'm also thinking time zones, if I want to call someone at 2:00 AM would be interesting if you support multiple languages, we do. So if someone wants to speak in the language of their choice, the transcription I think opens up a lot of potential for partnership with marketing automation, potentially a point of sale an LOS, of course, as someone who's tried to train loan officers, please disposition your calls and ask for intent. And of course they just put other on every single one That was really interesting to see. This is what we come here for innovation. Very cool.

Panel Member 1 (21:11):

Second, that I do like the natural language. I mean we talk a lot about natural language in text so far with ai, gen AI specifically, but it is fun to, during rehearsal we were just doing sound checks and so I now got my second listen of that and I said something just so everyone knows about like, Hey, I just lost my job and it was a completely different experience. So that generative experience I now have seen twice. Just lucky this afternoon. So it was very interesting and I'll echo Robin's comments about intent and disposition. Also just the quality checks in terms of overall quality, those comments that the AI is making about being on a recorded line and all these things. It's like the precision is built in which I also really like. Good job, Kastle

Christina Randolph (22:03):

Echo, those same sentiments. I love the live demo. That was really cool to see and how natural the agent sounded. I think it did miss that agility and the exchange of the pleasantries, but I mean I think that's as you train the model more, that'll get more sophisticated. Also, love the quality control and the scorecard aspect. I really loved the thought process and decision tree transparency. I mean there's so many applications for that. I would love to use that on my kids or my boss or my employees. Just what were you thinking, right? That would be an awesome application. I would've loved to hear more about, again, anything with ai, gen, AI risk and guardrails, right? I think if you know about the DPD use case and the risk of the consumer asking a bunch of different questions and what that response is going to be. I mean when you have a human in the loop, you can train them to say that's not what you're supposed to say. So that's always a concern. But I think if you set out to show effectively the future of customer service and that AI agent, I think you accomplish that really well. It's very optimistic for the future of an application like this, so great job. Thank you Kastle.

Panel Member 2 (23:29):

Next we have David Antony and Vish from Flatworld Solutions.

David Antony (23:35):

Thank you. Just two more to go. We are very excited to be here today and present to you M Suite, which is our AI powered mortgage automation platform. M Suite is built. By Flatworld and Flatworld has been for over two decades, we've been providing services to customers across different areas, and this is across 104 different countries and we provided services to customers, both technology and back office services, and this is across mortgage, insurance, healthcare, and customs brokerage to name a few of them. Now, this intimate knowledge and understanding of complex processes combined with our understanding of AI and expertise in AI has enabled us to build solutions that are very, very practical. So on one hand it enables you to really harness the power of GAI. At the same time, we also ensure that in all of these solutions that there is the guardrails of human oversight. So if you look at M Suite, the key skills or the basic skills that it has includes indexing, versioning, the data extraction piece of it, and checklist review and also a rules engine. So my colleague Vish is now going to walk us through some of the different modules that leverage these key skills to provide solutions in the mortgage space.

(25:02):

Vish.

Vish YR (25:04):

Hi, I'm Vish. I'm the VP of AI transformation at Flatworld Solutions. I'm 40 years old. I have to keep saying that out loud because in my head I'm still 28 and that's kind of a case with M Suite as well, even though we are introducing M Suite now as a platform. The document indexing engine has been around for about four years, and we've processed over 200 million documents across millions of loans using M Suite's core building blocks. We have built various automation for processes in mortgage industry, like debt monitoring, pre-purchase, audit, and bank statements income. Given the time crunch, I'm going to focus on specific features of M Suite to demonstrate how it combines technology, including AI with thoughtful user experience design to save time and deliver measurable ROI. Now M Suite can integrate with any of your systems, be it your L-O-S-C-R-M, front-end systems using shared database, API or even RPA bots if there are no other options. And M Suite has its own API that you can use to enhance your existing web application and mobile apps. Let's look at the government insuring app.

(26:30):

M Suite automatically assigns the next loan available in the queue for you. This is your standard Stare and compare checklist review process that's most common in most of the mortgage processes. This one has over 200 checklist items and most of it is automatically processed by M Suite and it presents only a few of those for manual review. This could be because M Suite's AI data mine. That manual review is necessary for those conditions, or it could be because it's pre-configured for a specific customer or a process that specific conditions should manly be manually reviewed. Let's look at this example, which talks about does the loan term on note match with final 10 0 3. Typically you would log into LOS, navigate to the respective folders, download the note document, do the same for 10 0 3, search for the data that you want to look at, stare and compare and tick the checkbox with M Suite.

(27:35):

You can do that with a single click of a button. It brings the 10 0 3 document navigates to the page and zooms into the section and highlights the long term. It does the same for the note document as well. It highlights the payment start date and just it highlights the payment start date and also the maturity date calculates the difference between that, which is 30 years in this case and says that it matches with the 360 months on the 10 0 3 and then it marks it as yes, it meets the condition. Of course, you can override the M Suites recommendations with just a click of a button.

(28:20):

M Suite also has timers at different levels. You can see how much time I've been spending on a particular loan or on a particular document and even on a particular condition. And no, it does not time out at eight minutes. We use these timers to identify bottlenecks and eliminate those bottlenecks. We introduced the government insuring solution to a customer and we saw 35% time saving in week one of rolling it out and it went up to 65% in just two months. Now what about processes where users consume the output of M Suite without any user logging in? We are careful about the user experience even there as well. This M Suite's appraisal review solution M Suite reviews the appraisal report, whether it's form textual content and even the images, it images, it analyzes the images using its latest AI and produces an output report, which is a PDF document.

(29:27):

That's what the customer asked and presents its findings that you should look at right on top of the PDF. And if you want to look at the data backing M Suite's finding, you can do so with just a click off a button within the PDF document and come back up to the findings. Again, M Suite also has a standard data model that you can integrate with your data warehouse for various reporting and dashboarding needs. But what if you could speak to your data warehouse to get the insight that you need instantly? Now you can do that with Claire built into M Suite, which are my top performing branches. Claire connects with your data warehouse to bring the insight that you need when you need it. It also brings in certain recommendations based on the question that you have asked. You can just click on it and then get the answers Claire conforms to your data security setup on your data warehouse. For example, if the loan officers will only see the data for their loans while your CXOs can see data, which is organization wide. So here's David. I'm Vish from Flat Solutions and this is M Suite delivering Measurable ROI. Thank you.

Robin Clayton (30:59):

I think there's an incredible potential to scale this. You can see this working with a really, really large company or someone who's spread out over a large geographic area. Really interesting stuff there. It would be interesting in that view that you have there. If I wanted to drill down for Mount Laurel just to see analysis, it was a little bit more interactive. I loved on the review with the multiple panel style, it almost reminded me how a typical operations personnel would work with multiple monitors. They're used to saying multiple things at the same time, and that was pretty cool. The timers were scary. I don't know if I'd like to see just a timer on every single thing that I'm supposed to do. Sort of a running clock on that, but very interesting. Thank you.

Panel Member 1 (31:43):

Thank you. I'll counter the timers comment. I like the timers because if we're going about operational efficiency and units per fulfillment employee, I think it's actually, I actually quite like that. There were three things, ratio calculations, decisioning clearing conditions that were part of your promo for the show. And I did want to see more of that because for me, I'm looking for how you getting the loans done. So that's just a demo quality thing. I wish I would've seen that on the Claire AI part. I will say that I quite like the clickable bit. I haven't quite seen that type of functionality in generative AI responses where you can actually then just click down into the responses and start getting more data. I thought that was clever and I liked it. Good job. Thank you. Thank you.

Christina Randolph (32:28):

I'll also counter on the timer. I like that from an efficiency standpoint, I think we enter into conversations a lot with lenders where they have no idea how long a task takes, and in order to figure out how much you can save, you need a baseline. So I think from a tracking and monitoring perspective, that was huge for me to see. I love the quick access to the data. I mean, we have to wait a long time for reports and they have to go into releases. So being able to just ad hoc, ask it a question and it pull back all that data is also a huge benefit. I would've liked to see more, I think of the system and more of the versatility of it as you did describe it. But other than that, I think the use case and just some of those features are huge. I think you just need to pull on those a little bit more and there's good potential there. Sure.

Vish YR (33:19):

Thank you.

David Antony (33:19):

Thank you,

Panel Member 2 (33:20):

Thank you. Flatworld Solutions, M Suite. And next, closing us out. We have Hemanth and Brad Subvert with Tavantt.

Brad  Subvert (33:38):

All right, so we are the only thing between you and what's on the other side of that wall, but that's fine. We're here to do this. So I think most people know Tavant, most people know what we do. Tavant in 2023, processed, 37.3% of all loans in the United States went through Tavant platform. So we have scale on the flip side, when you're sitting at home in bed with your significant other and you're on Hulu and you're wondering why Hulu is serving you this thing, that's also something Tavant does. Hulu, Disney, ESPN, on that platform. So those are the two aspects of what Tavant really does today. Hemanth will be here to talk to us about Tavant's home equity box solution and honestly why it is a solution.

Hemanthkumar Jambulingam (34:25):

Absolutely, absolutely. So Tavant's home equity in a box, there are multiple solutions in the market, which provides piecemeal automation, but the home security in the box provides an end-to-end transformation starting from your assets or even from your leads all the way till your closing and funding. So this is one of a kind platform which is available in the market, which will give you complete digital transformation, provide a seamless, faster efficient journey for the borrowers and lenders. So it's not about just automating it, it's about how you transform business from one state to another state and you scale it up. So that's home equity in a box.

Brad  Subvert (35:18):

But why should anyone believe that Tavant can do it faster or better than any of the other amazing companies here?

Hemanthkumar Jambulingam (35:24):

Absolutely. Very good question. So with Tavant month, we have been here in the industry for more than two decades. We have seen multiple different systems how process automations have been come and went. So what we have done is we have identified the task, which can be completely automated, and the process areas where you need human in the loop. And then we are introducing AI into the mix to bring automation into the workflow automation into the data capture so that we can bring a seamless, faster, and accurate decisioning process. So that's the secret behind it.

Brad  Subvert (36:08):

Alright, so that's the secret. So we have multiple clients in this room actually,

Hemanthkumar Jambulingam (36:12):

Yep.

Brad  Subvert (36:12):

We're not going to share the AI aspect of it. So why should anyone believe the AI aspect of it?

Hemanthkumar Jambulingam (36:18):

Absolutely. So for ante AI is not a buzzword. So you pointed out, I mean we work with other industries also, right? I mean we learn what is happening in the manufacturing side on the retail side. So we bring the best from the AI learnings and then we plug that into our lending space. So our AI not only helps you to streamline the process, it also assists you in your compliance related check. So that's where we are plugging in to ensure that we are within the boundaries we are playing and then we are able to make efficient loan processing and decision.

Brad  Subvert (36:56):

So are you saying that Tavant AI solution is the compliance solution or it's abiding by the lenders who owns the compliance and security in this situation?

Hemanthkumar Jambulingam (37:09):

Yeah, so the whole platform, the way you guys are seeing the workload which has been orchestrated is any lender, this is not just a box solution. A lender can take this, they can see how you can fine tune. Of course, we give the industry best recommendation, learnings and recommendation. So when we define the compliance standard, we come up in the box saying that these are all the list of regulatory standards, what you need for the home in the box, or it could be any product. And then the lender, the compliance team can review and then they can add in their specific rules, which has to be included. So that's the power of the platform when it comes to configuration. You can define the workflow for home equity versus any other close nine seconds.

Brad  Subvert (37:59):

So your marketing guy, he is wearing a pink jacket right now, publish something saying that Tavant had a 64 NPS score on this. So how does that actually translate and why would anyone in this room believe us or you?

Hemanthkumar Jambulingam (38:13):

Yeah, that's a very good question. So what we have done is when we define the workflows, specifically when it comes to underwriting and automation, what we have done is we have built the system specifically for our underwriters and then we extended this onto the origination space on the POS side. So the moment the information has been captured, everything has been validated upfront with the verified source. We work with multiple different partners in the industry to get verified validated data so that before anyone who's picking up this loan in the processing, underwriting compliance or anywhere, we make the informed decision and we present the recommendation. So that gets verified, validated upfront, it's all black and white,

Brad  Subvert (39:04):

So everything's black and white, which is great. We're talking about data, verified data. For Lenders the value is actually in the data. If a lender can go from a one and done transaction to a one in many transaction, how does this platform provide the data to go from one and done to one in many

Hemanthkumar Jambulingam (39:21):

Very good. I mean, data is the key. In fact, for the past two days, everyone is focusing on data. With the data, what you can do, how you can churn your pipeline or look for the insights on analytics, on the data. So with Tavant, we do have a 360 degree view of analysis, we call it as data beats. Data beats focuses on specific areas in your process. It understands what's happening with your data, what is the workflow which has been streamlined, how much time the process it is taking, what is the bottlenecks? And then provides insights, recommendations and action so that you can see through this, configure the workflow and see how the process get measured or increase.

Brad  Subvert (40:09):

So with all that being said, anyone in the audience can maybe look to your right or left and probably find a to client, but that statement is not a believable statement in itself. How does someone actually get to the point where they're going to believe anything that you're saying on stage?

Hemanthkumar Jambulingam (40:23):

Yeah, one good point is we have been doing this. We are transforming the whole digital experience. When FinTech boom started in 2014, 2013, we were the pioneers to take a mortgage journey into a complete digital capability. I mean, let it be the borrower experience, let it be the broker experience, our loan officer experience. We are transforming tech into the mortgage space so that the lenders can gain efficiency, increase productivity, reduce the operational cost. So those are the key things any lender look for in terms of operation.

Brad  Subvert (41:07):

So most of our clients, honestly, we sign double NDAs with,

Hemanthkumar Jambulingam (41:11):

 Yeah.

Brad  Subvert (41:11):

We don't talk about who our clients are, that's the business that we're in. But we're happy to showcase anything that we do with our clients.

Hemanthkumar Jambulingam (41:20):

Yeah.

Brad  Subvert (41:20):

And most of the clients that are in this room, I can see half of you looking at me right now, will either say good things or bad things about us, but they're going to say something about us. But the truth is we're very transparent in what we talk about here. Final words.

Hemanthkumar Jambulingam (41:35):

I think the equity in the box, homemade the boxes, we are disrupting this industry in terms of the lender doesn't have to think about, okay, am I going to miss out in the game? We can instantly transform, bring all the integrations together and you can go live within a day or a week. Anything else?

Brad  Subvert (41:59):

No.

Hemanthkumar Jambulingam (42:00):

Oh, we are on time. Okay, awesome.

Robin Clayton (42:03):

Thank you. You mentioned several times, why would anyone believe us? I would say seeing is believing, would've loved to see it or some kind of something. And it sounds like the idea that there's just a lot of folks that have already seen it and love it, but there may be some folks who haven't seen it and would love to. I love the idea of touchless lending. That's a great name and that's what we want to see. We want to eliminate touchless really insightful insights and recommendations and actions that are tied to that are really important. Data without actions is not really helpful, but if I know exactly what I need to do next to be successful, that can be really powerful. So yeah, you left me hungry for sure.

Panel Member 1 (42:56):

Yeah, I'm scratching my head on why we didn't get to see the product and especially on home equity. I've seen your product suites, so I think I understand what you have up on the screen. I will comment on some things that I did want to see, which is most specifically when it comes to home equity, the hangup, as we all know is collateral analysis

Hemanthkumar Jambulingam (43:17):

Yep.

Panel Member 1 (43:18):

And I want to see how you're doing collateral analysis in 15 plus all the rest of the underwriting. So that was the big, that was probably my biggest remark. But if it's because you don't want to present in front of competitors, I will just say that everybody, anyone can watch a Bruce Lee movie. It doesn't mean they know karate. So I show your product.

Brad  Subvert (43:40):

Fair.

Christina Randolph (43:41):

I'm going to agree. I think Mont, I've been a long time fan of you guys and I know that you have so many different products and solutions, so I'm sure it was a challenge for you just to come here today to talk about just one of the many things that you're doing, but also was disappointed that we didn't get to see anything. I will say one thing that I didn't hear you mention, but I wonder if it's a capability, is the giving with home equity potential, giving the lender an ability to size up the opportunity within their existing pipeline. So if that is at all a feature, I think that

Hemanthkumar Jambulingam (44:16):

Yep.

Christina Randolph (44:16):

Would really help bring more value.

Hemanthkumar Jambulingam (44:18):

Absolute, absolutely. Thank you.

Christina Randolph (44:21):

Nice act

Panel Member 2 (44:25):

 Thank you Tavant.