In this episode of Lykken on Lending, I sit down with Adam Carmel, Founder and CEO of Polly, to explore how his team is transforming one of the most overlooked yet critical areas of mortgage banking—capital markets technology. We dive into why legacy pricing engines have held lenders back, how Polly is using AI and agile infrastructure to cut origination costs, and what the future holds for a more transparent, configurable, and profitable mortgage ecosystem. If you’ve ever wondered how technology can solve the $13,000-per-loan challenge and bring real innovation to the secondary market, you won’t want to miss this conversation.
[David] Listeners, we have joining us today on the podcast, Adam Carmel, he is with Polly. When you look at technology, you look at the markets, it’s so exciting to see what is coming on in the way of innovation. What Adam has created with Polly is just that very innovative, and I’m really excited to share his vision for the capital markets, where they’re at, where they’ve been, where they’re at, and where they’re more importantly, where they’re going.What can we anticipate? So Adam, I’m really thrilled to have you here.
[Adam] Thanks very much, David, for having me. Really looking forward to it.
[David] Real honor. Especially with the subject matter recovery today because it is one with the markets being as crazy as they’ve been and what’s going on. But let’s first start by letting our audience get to know you. Now we have a lot of people in the production side of the business and we have all the realtors, builders, and all kinds of folks, even people on the Hill staffers that are listening to this podcast. So let’s let them get to know you a little bit. How do, tell us about your journey into mortgage lending, specifically into the capital market side.
[Adam] Sure. So I started my career in the mortgage industry in 2007. I went to work for an independent mortgage company and entered right there, right before the credit crisis, which was quite an experience at such a young age to learn and so anyways, that’s really where I sucked my teeth into the industry. I learned it.
[David] Did you start on the origination side of it or in the opts side, or did you get right into the secondary side?
[Adam] Originally I was hired at a REIT and then I transitioned over into the mortgage side of the business and suck my teeth into the servicing side of the business, actually.
[David] Oh, interesting.
[Adam] Yeah. And then I was involved in a couple other things within that business and, and then I ended up leaving and in 2013 I started my own mortgage company. Way too young of an age.
[David] Great experience though.
[Adam] Great experience. So I have a real deep appreciation and admiration and recognition for how hard the business is, but also how exhilarating and fun and and great it could be. And but that’s really where I learned all about capital markets and the impact that technology could have on the industry. And so we grew pretty quickly, actually and we were using, all the same legacy kind of systems that everyone still uses but my thesis was that over time the mortgage business would become more commoditized. There’d be less product, the aperture of product available would be more limited as a result of Dodd-Frank and so forth. Also, I thought the digitization of the mortgage would, create price transparency. And so I had this thought, and plus we had a bunch of pain, like a ton of pain with the, the legacy pricing engine that we were using. And I said why don’t we go build our own pricing engine and kind of solve all the pain that we were having with that legacy system, but then as well try to build it for the future and what we’re looking to achieve.
[David] That’s awesome. What year was it that you launched Polly then?
[Adam] So Polly wasn’t launched until 2019, so there was a whole, the mortgage company kept on running all the way until 2019, sold the mortgage company, and then that’s when I started Polly.
[David] Good. I love what you’re doing. You brought a real innovative approach, Adam, to the whole capital market, especially risk management and part of this, and. I have real everyone who uses a system, you have a almost a cult following of how many people. So you’re pulling people outta legacy systems. Saw that opportunity as a result of being in legacy systems. And so let’s talk a little bit about the things that you wanted to achieve. One of, I know you had talked about was agility. You wanted to make sure that you had a system that could be agile.
[Adam] Foundationally we thought that when the company was started. We built the whole backend. We spent a couple years doing it. We didn’t even focus in on really, frankly, anything else. And so that’s why this, the actual pricing engine wasn’t available into the market until 2022 or 2023. But we had a couple core principles that we were very focused on. One, was to deliver a system that had unlimited flexibility and configurability and granularity across all parts of the system. And so that ex, that starts with, for example, margin management and pricing eligibility and so forth, all the way to all of the workflow automation and all of that. So that was tenant number one. And then 10 number two was, like you mentioned, having really strong agility, flexibility. Basically a system that is entirely self-service, no code deploy. And lenders, banks, credit unions and so forth are able to build a custom parameter, build a new product, build really anything they want within the system on their own without any involvement from us. Now, of course, we’re more than happy to help. Very often it’s the case in the industry where you’ve gotta submit a ticket or, you get some SOW and it takes months and or it’s not even possible. And so we really wanted to build something that like you mentioned, was really fast really dynamic and can get you to market as quickly as possible, which, after with how difficult the market’s been here in the last couple years, that’s been really important in helping our customer partners increase their revenue per loan and drive their cost down and so forth.
[David] Great point. A lot of people would say, why do you need agility? Isn’t capital markets been the same for decades and decades? In one sense that’s true, but another sense. There’s been a lot of need for the. Explain why that is, if you could dive into that part of it a little bit more, Adam.
[Adam] Yeah I think that therein lies some of the opportunity, like you said, that’s been done the same way for decades and decades and that’s just no longer good enough. If you compare what the available software tooling has been in the mortgage industry relative to what we’re used to as consumers, whether it’s name, a software tool or name, anything. It really is quite stark how bad the available tooling is or has been in the industry versus what we’re used to as a consumer, and our mission has been always like, how do we help get these products to market faster? How do we get how do we help lenders and banks drive profitability? How do we help drive an entirely new loan officer or broker experience? How are we redoing and rethinking the mortgage kind of capital market, supply chain from start to finish? And doing it in a way in which it allows for them to go to market at their pace. Some want to go very fast. Others might want to go slower and they shouldn’t be. Here’s the thing, the industry has been conditioned to operate a certain way for decades. Yeah. And so the, if you look at what’s happened here as a result, the cost origination’s gone from 2,500 to 13,000 in the last 20 years. In refi waves, the industry hires a bunch of people and then when it’s tough, they unfortunately have to lay off.
[David] Boom bust. We’ve been stuck in boom bust cycles for decades.
[Adam] Technology should have solved this.
[David] Should have.
[Adam] But it has not. and quite the contrary. In fact all the fact…
[David] It’s have been a contributing factor to the cost rising from 2,500 to 13,000.
[Adam] That’s exactly right. All that’s happened is legacy companies have sucked out more out of their customers, but they’ve not delivered any value. If you charge more, then you need an ROI on that it needs to be three x five x, 10 x, 20 x, whatever it is, you need to see an ROI on that additional spend. But what’s happened here is there’s been more cost from software companies and less value, and we’ve just been stuck in the stone ages no, it’s so true. When I when this thing got started I said, boy I did a whole bunch of market research and, it was more like boy, if we could just be a little bit better than that’s not the hard thing to do necessarily, but what if we can be exceptional? What if we can really revolutionize the way things are done and help to drive that cost down from 13,000 all the way down, hopefully over time, back to the 2,500 or even less? or how can we create a system that scales in good markets and, you don’t have to be moving up and down your head count based on market conditions and so forth.
[David] Talk about, I wanna get into margin management because you talk a lot about the importance of being able to do that in a flexible way. Get into a little bit how you handle the margin management.
[Adam] The way the industry’s worked historically, legacy systems, all of them, they use a tree folder, like a structured tree folder, and it goes from one to the next. And, And as a result that makes things very complicated. You can’t really configure multiple dimensions. You really have no idea what your profitability is at the time of the lock you, when you wanna make a change, this goes back to your question on agility. When you wanna make a change, you’ve gotta change the rule across, however many products you have. It could be 200 products, it could be 500 products, it could be 2000, whatever. But you gotta go one by one. It’s really cumbersome. It’s really a problem. And this goes back to that backend. We completely reimagined how it should operate and we took best practices from stuff that we’ve seen. And, I’m from the Bay Area in Silicon Valley. We just did a bunch of research on how to re-architect and actually we’ve got a system that’s patent pending now on related to margin management and so forth and so now what our system allows us for the lender or the bank to use an unlimited number of parameters in real time, you can do anything you want to get down basically to a loan level configuration if you so choose. And so that’s really driving profitability higher or volume higher, and it just depends on what lenders or banks choose to do related to what their margin is. It’s gotta help retain good talent that you have as well as attract top talent. Is that a true them?
[Adam] That’s the number one thing. Nothing else matters. At the end of the day, we’re here to serve. Our customer partners, and we can’t do that without the very, very best talent. And I think that we’ve been, I, if you were to ask me the thing that I’m most proud of related to Polly it’s just that, that we’ve been able to attract and work with and partner with the most talented individuals I’ve ever worked, had the opportunity to work with in my life and quite frankly, like people that the mortgage industry has historically not had in it. And so we are I think that’s our greatest differentiation, actually. The product can only be built as well as the talent that is building it. We’re fortunate to have world class talent and we’ve really expanded now we’re, we’ve got, 150 folks on the team and growing.
[David] That’s, congratulations on that. Yeah. Doesn’t this also your system open up some new niches that are out there?
[Adam] The system’s evolved to we, the apertures expanded quite a bit. We’ve got a lot of products now that are driving a lot of value in ROI they’re really changing the way in which a loan officer or a mortgage broker interact. With the pricing engine, our AI product is really something special and it’s we’ve seen exponential growth on that. We’ve got a lot to offer now and then, of course our, the anchor product, the pricing engine, we just keep going deeper and deeper into it and we continue to listen to our customer partners and others on all of their pain points, and they’ve got a long list of them over the years from legacy systems all the way to partnering with us on things that historically have never been thought about on what a pricing engine is capable of doing and it’s been really miraculous to see. And so it’s been really cool.
[David] Yeah, it is. When you look at what Polly is today I wanna get into understanding more of your vision. It started out as a pricing engine, if I’m not correct. That’s right. And it started there, but you would continue to expand. This may be a good point, seeing as you’re bringing that up, how you, what you were and what you’re becoming to expand that. What, so it start out as a pricing engine, where have you grown to and where are you planning to go? to the extent that you could talk about it. Yeah.
[Adam] There’s always more to announce.
[David] Always more to announce. Yeah.
[Adam] Six months or a year, yeah. We did start off as a pricing engine. And, we remain, that’s our core product. It always will be. But we’re starting to look at things more holistically and understand, and just really understand all the pain in the mortgage industry and see what’s interesting about all that’s possible related to Gen AI is what’s been conventionally thought of, and it, I’m not even talking about necessarily just the mortgage industry, but what’s been conventionally thought of in terms of what’s possible, how to create value in ROI at an for, as an enterprise to it’s partner, it’s customer partners and so forth. It’s unbelievable what we’re able to do versus what you would historically think of as a pricing engine or some capital markets tool and as we are digging deep into what’s possible, I think we’re uncovering that everything is so for us it’s really. For us, what’s really exciting right now is picking the right spots and then understanding where the most amount of value is we can create. And what I mean by value is where can we help drive the most cost out of the system? How do we help that 13,000 come down to 2,500 as best we possibly can? And so my general view is that the amount of software companies in all of these major industries, mortgage, insurance, whatever, there’ll be less of them, because no one needs or wants seven AI things running around using maybe different data. That’s not really a holistic solution and so I think we’re uniquely positioned where we sit in terms of market share and where we will be in a year and a half too. The amount of our balance sheet and so forth. Our team and our continued execution on our vision. Those are just hard in our culture. Those are just hard things to assemble and so I’m really excited about what’s to come. And so what is that? I think it’s, rather than going through product, by product, it’s more, we’re really hyperfocused on driving the cost of origination down, one. Two driving revenue up as best we possibly can. And I think there are some non-obvious ways in which we’re gonna be able to do this that haven’t conventionally been considered. It’s been more very basic, frankly, historically and so you put those two things together. We’re trying to bring up as much economics as we can into that transaction between the loan officer and the consumer. And then it’s up for the mortgage company or the bank to determine how much they’re gonna pass through or not. In terms of profitability or just going and getting more volume.
[David] Right? Cost reduction is certainly a focus of everything, but it really comes down to industry adoption. How to what rate you’re getting your product adopted in the market. It sounds like you’re up to 150 employees and you’re profitable. You’re doing well. What has been the key to that happening? I think just our customer centricity and our innovation, those two things. If you were to boil it all down, I just don’t think that you’re gonna find a company or a group of people who are more passionate and more committed to the success and happiness and value, hopeful value creation that we’re driving into our customer partners. And then number two, our pace of innovation and agility and execution is unmatched. We’re really, we just have a phenomenal team and executing on both those core things. And that’s not so easy to create and it’s impossible to duplicate. And so we’re really excited about that.
[David] That’s right. Talk about the importance, capital markets is one of those topics that’s oftentimes referred to as, it’s the black box. I don’t understand what goes into it. And goes on in it, but it’s really critical to my success. That’s how a lot of loan officers look at this. How important, how much do you help with communication and collaborating with the loan originators to understand that, to the degree that they need to?
[Adam] Yeah. Quite a bit. We’re part of our part of what we’re, what we’ve done is to take a, what is generally a complicated and like you said, just more opaque situation and within capital markets, and we’ve made it super transparent. We’ve particularly within our AI product it’s pretty sensational what it achieves for loan officers and ultimately consumers and whether that’s it’s small stuff, right? Like loan officers or mortgage brokers have the ability with our system to be, to be fully mobile. The system has AI embedded throughout, so it’s making recommendations and optimizing products and explaining in eligibility and making recommendations and so forth. And doing…
[David] When you say it talks about making recommendations, get a little more specific on that are you telling when the lock, we know we have Barry out there that’s putting out these Barry Habib that puts out the market alerts. We have MBS live.net. That does a great job of publishing. There’s some great services out there. To what extent do you provide intelligence out to your customers and out to your, to the banks, mortgage bankers that you’re working with and to their LOs
[Adam] Quite a bit. We don’t try to predict or time the market or make suggestions on when to lock or when not to lock. We’ll leave that to Barry and others that, I have a better view on that than I suppose me. We do now we, we do have tooling in the system that would make a recommendation based on the current price of the loan or what the lender has said that they would like to be triggered in terms of notification and so forth. So it our ability to recommend. Near miss pricing, near miss eligibility, optimizing pricing on a rebate on a particular loan and frankly now it extends beyond just like pricing and eligibility. It’s pretty extensive what to do with the loan, how to do it, and so forth. It’s become a system on its own, which is really cool to see. The inference and the training has been really cool. So is it is probably more of a system that the capital markets department use the secondary people? or is it actually being rolled out to the mortgage loan originators?
[Adam] Both.
[David] So it is okay.
[Adam] We are, yeah.
[David] More capital markets works is a mobile lock desk to a certain degree then?
[Adam] Yeah, that’s right. So capital markets folks use it. Lock desk folks use it. Loan officers, mortgage brokers. We’re starting to see others use it as well. Process, certainly processors, loan officer assistance, in some cases, even underwriters. We’re seeing finance accounting and so forth using it, particularly our data and analytics and other tooling we have around that. So it’s become a bit bigger.
[David] Yeah, quite a bit. And the APIs that you guys have, I understand when we had a conversation a while back, you were talking about your APIs, they’re really pretty astounding.
[Adam] Yeah, it’s, it, the whole system’s been the whole system’s available through an API, through API.
[David] Which just speaks to the collaboration that you’re trying to get, working with other vendors in the collaboration to make sure that…
[Adam] We take an approach of, we have no, we don’t want, we don’t take a view on anything, this goes back to that core tenant of being hyper, having unlimited flexibility. Our APIs allow for lenders and banks to choose, do they want their experience? Would they want our experience? Do they want a partner’s experience in terms of the Ui? Do they, do they want to use our automation? Some other automation like we’re just here to serve, right? So all we do is we understand as part, once we start working with a prospective customer, we really take a partnership approach to this prior to someone signing up with us, we really wanna understand their pain points. We really wanna understand what they’re looking to achieve, what their strategy is. We’re really, so we have a deep scoping session with them in multiple sessions, in some cases, and we are constantly formulating and dialing the knobs, if you will, to optimize every single component of the system for them, and that could be done through API or otherwise.
[David] Wow. It’s impressive. When you look at AI, you’ve touched on it several times, what can we anticipate from ai? It’s, it seems like it in so many aspects of the industry is taking out the busyness and the real technical aspects and allows us as humans to get back to the most important part, which is the relationship side of it.. What’s your vision for AI within Poly?
[Adam] We’re very we’ve been fortunate that we’ve got like this cool advantage again that we’re, we have this interesting nexus of domain knowledge, institutional knowledge, and technical knowledge that is, is really unique and rare and special because here we’re based in San Francisco, where that’s the heart of AI in the world and all that’s being built. So we always, and our investors are in invested in name, the AI foundational model company. Anthropic, OpenAI, you name it, X AI, whatever. And so we are always understanding and learning what the very, very cutting edge is out there related to AI. So that’s like a real advantage. And we’re able to learn and understand best practices. We’re able to learn and understand where those companies are going because you gotta really understand that as you’re building things. And so we’ve been very and then of course we have a real deep understanding of all the pain points in this industry. I, myself have experienced them as we mentioned, or as I mentioned. I experienced them for, whatever, almost 15 years or so, maybe a little less. And so we so you, and then our team, the same thing. We’ve got a lot of people also that have experience in the mortgage industry or with deep technical background. So putting all that together has been a huge advantage. And so where do we think AI is going? For the enterprise, at the enterprise level, it’s gonna be largely, we think agentic based and really we think the there will be a, it’ll, it will augment and supplement labor, if you will where you agree with, where individuals can do a lot more with a lot less mind share and a lot less work. You know what used to take eight hours, might take an hour or two. What was, whatever conversion. For ca call centers or even distributed retail will be a lot higher. What was, whatever the cost was to do any number of things in the mortgage, it you in that mortgage supply chain, they’re gonna massively go down. They better massively go down.
[David] Yeah. They’ve got to or otherwise we’re. Let’s put it this way. He who gets that done first is gonna have a decided advantage in the industry.
[Adam] That’s exactly right. That’s exactly right. And it’s gonna happen. I, it’s happening, I don’t know who said it, but I’ll quote it, where people, the human beings overestimate what can be achieved in one year, but underestimate what can be achieved in five years.
[David] Yeah, that’s a great quote. That’s a great quote. Yeah. Gotta go look that one up. That’s a, that is really, it is really true. That’s really true. I’ve been recommending to our audience for the longest time the coming Way by Mustafa Suleyman. Mustafa Suleyman now is the president of AI at Microsoft. And a lot of people are, they’re, everyone’s trying to figure out how to monetize this. And I understand correctly, you’re closer to it than I am, but there’s not a lot of money flowing into new AI They’re trying to figure out how to monetize what’s there.
[Adam] That’s exactly right.
[David] So it sounds like you’ve got a great application for that to happen in.
[Adam] I think if you’re not, if you’re a software company, a mortgage company, a insurance company, a I don’t, a hotdog stand,
[David] Mortgage loan originator even. I don’t care what level you are.
[Adam] You really need to understand what’s happening here and be really well positioned or it, it could be it could make things very challenging.
[David] Yeah, I agree with you, Adam. What a joy to get to gather with you. I really love the vision. How could people reach you and learn more about the company and yourself?
[Adam] I’m available 24/7. For those who know me.
[David] And are we talking to an avatar here? that doesn’t imply an avatar.
[Adam] No, I, I legitimately you can ask anyone in our within poly, they can get they and our partner customer partners. They can get hold me anywhere I am in the world at any time. With, I need to sleep, but with some exception. Yeah. Yeah. So feel free to email me, shoot me a note on LinkedIn. Gimme a call My cell, you name it. We’re here. We wanna listen, we wanna learn, we wanna partner and we wanna just create a lot of value and drive ROI and do the very best we possibly can.
[David] Make sure you get your email and cell phone number emailed over to us, and we’ll make sure we put it up on the screen so people can reach out to you. Adam. Again, kudos to you for what you’ve built, the thoughtfulness that has gone into it, the intentionality and the innovation quite frankly, you’re bringing to this otherwise very uninnovative space part of our industry.
[Adam] Thank you very much. We really appreciate you having me and look forward to chatting hopefully sometime soon.
[David] I can’t wait to see at the next conference, that’s for sure.
[Adam]Thank you. For sure. Likewise.
Important links:
About the guest:
Adam Carmel has been ingrained in the mortgage capital markets space throughout his professional career. Upon firsthand experience of the fundamental gaps that exist within legacy technology solutions, Adam shifted his industry acumen and deep domain expertise into fintech. Launched in 2019, Polly has pioneered the next generation of mortgage capital markets technology with its cutting-edge, data-driven platform. Its enterprise-grade solutions, including the industry’s only cloud-native, commercially scalable product, pricing, and eligibility (PPE) engine and first-of-its-kind Polly/™ AI platform, empower the nation’s top banks, credit unions, and mortgage lenders to automate and optimize all areas of the capital markets and secondary marketing functions, accelerate success, and ultimately—deliver a better borrower experience.