How AI Cash Flow Automation Helps CFOs Reduce Late Payments and Improve Working Capital - Carlos Vega
In this episode of Future Finance, hosts Paul Barnhurst (aka The FP&A Guy) and Glenn Hopper welcome Carlos Vega, fintech founder, investor, and AI enthusiast. Carlos shares his unique path from investment banking to co-founding Tesorio, an AI-powered financial operations platform helping companies optimize cash flow and unlock real working capital. Carlos reveals how his frustration with traditional finance workflows led him to build tools that empower finance teams to move at AI speed - with smarter forecasts, automated collections, and deeper financial visibility.
Carlos Vega is the co-founder and CEO of Tesorio, an AI-powered financial operations platform. He has over a decade of experience in finance, including investment banking at Lazard and co-founding a factoring business. Carlos holds an MBA from Wharton, where he created a custom focus in “Innovation Through Analytics” He’s passionate about turning data into actionable insights and making cash flow more predictable. At Tesorio, he helps companies automate collections and optimize working capital using AI.
In this episode, you will discover:
How Carlos designed a custom MBA in “Innovation Through Analytics”
Why cash flow forecasting is more of a data problem than a finance one
The real reason so many businesses rely on factoring (and what it reveals)
How Tesorio uses AI and ML to clean and structure financial data
Why actionable insights matter more than predictive reports
Carlos Vega shared practical insights on why cash flow is one of the most overlooked-yet critical-parts of running a business. With his background in finance and experience building Tesorio, he brought a real-world perspective to the conversation. He emphasized the importance of understanding where your money is coming from and when, and how small changes can make a big difference. At the heart of it all: clear processes, good habits, and staying on top of the details.
Follow Carlos:
LinkedIn - https://www.linkedin.com/in/carlos-r-vega/
Website - https://www.tesorio.com/
Join hosts Glenn and Paul as they unravel the complexities of AI in finance:
Follow Glenn:
LinkedIn: https://www.linkedin.com/in/gbhopperiii
Follow Paul:
LinkedIn - https://www.linkedin.com/in/thefpandaguy
Follow QFlow.AI:
Website - https://bit.ly/4fYK9vY
Future Finance is sponsored by QFlow.ai, the strategic finance platform solving the toughest part of planning and analysis: B2B revenue. Align sales, marketing, and finance, speed up decision-making, and lock in accountability with QFlow.ai.
Stay tuned for a deeper understanding of how AI is shaping the future of finance and what it means for businesses and individuals alike.
In Today’s Episode:
[01:28] - Welcome to the Episode
[02:54] - Creating a Custom MBA
[09:46] - Lessons from a Factoring Business
[10:50] - Why People Ignore Cash Flow
[12:27] - Building in Panama vs. the U.S.
[14:12] - Cash Flow Is a Data Problem
[19:21] - Is Cash Flow Forecasting Overlooked?
[21:01] - Building Accurate Collection Models
[23:53] - Real AI Use Cases in Finance
[32:05] - Carlos’s Grounding Habits
[36:53] - Final Thoughts & Key Quote
Full Show Transcript:
[00:01:28] Host1: Paul Barnhurst: Welcome to another episode of Future Finance. I have here with me my co-host Glenn Hopper, and this week we're going to be interviewing Carlos Vega. Carlos, welcome to the show.
[00:01:40] Guest: Carlos Vega: Hey, thanks for having me. I'm really excited to be here.
[00:01:43] Host1: Paul Barnhurst: Yeah, excited to have you. So let me read a little bit about Carlos' background and then we'll jump into things. Carlos Vega is the co-founder and CEO of Tesorio, the world's first connected financial operations platform. Tesorio is an AI-powered financial operations platform that helps businesses turn revenue into real working capital by automating collections, optimizing cash flow, and enhancing financial visibility. It enables finance leaders to move at AI speed, reducing DSO, increasing investment velocity, and improving capital efficiency. Carlo's journey in finance spans over a decade, including time in investment banking at Lazard in Latin America, co-founding a factory company, and working closely with the CFO of GM's pension fund. Carlos holds an MBA from the Wharton School and a BA in economics from the College of Arts and Sciences, both from the University of Pennsylvania. Love the background, Carlos. We're really excited to chat with you. And Glenn is going to hit you up with his favorite question here.
[00:02:53] Host 2: Glenn : Yeah. So, Paul, I noticed this before I did, which is funny because this has been something that I can't believe more universities and more people aren't pursuing. So just by way of background, I got my MBA back in the.com era because I'm old as dirt. And then I went back later and got a master's in finance with a grad certificate in business analytics. And I thought at that time. So this was in, you know, around 2015 or so I thought at that time, how is every MBA program not going to include a major focus on analytics? Because we were seeing what's going on with data and more and more people using it and sort of the whole data-driven decision making. And to me, that was part and parcel with what an MBA should be. But you are one of the few people I've actually talked to who has the MBA, where you specialized in innovation through analytics, and I've I'd love to hear about the program, what you learned about it, how that was different from a traditional MBA, and how that kind of maybe changed your career trajectory? If so, and just what the focus has been and what that meant to you, to sort of get the business and analytics at the same time.
[00:04:01] Guest: Carlos Vega: Awesome. So that's a great question. And it might be a more colorful answer than what you signed up for. But long story short, whenever I get asked like, hey, was the MBA worth it? I always tell people, look, if you want to start a company and you're going to go through an MBA, you need to chart your own course, right? Like, you can't just, you know, if you're going to get an MBA and go into banking or strategy consulting, you know, you can go pick from the list, take a course, do the internship, and go where you have to go afterwards. But this title, innovation through analytics is something that I made up. So it's a customized degree. And now there is a degree at Wharton focused on data analytics. At the time, this was 2012 to 2014 when I was there. The term big data was the hot term of the moment. And I had been studying under a guy named Peter Fader, who's actually in the marketing department, but he teaches, you know, quantitative marketing and probability modeling, and I had also taken classes with a guy named Ethan Mollick, who is now actually very well known in the AI space. At the time, he was the entrepreneurship teacher.
[00:05:02] Host 2: Glenn : So I'm going to interject. Ethan is good friends with the professor in my program. They both went to the PhD program at MIT together with Karim Lakhani. So they're both in the program together. And Karim.
[00:05:13] Guest: Carlos Vega: So, you know, you know him?
[00:05:15] Host 2: Glenn : Yeah.
[00:05:16] Guest: Carlos Vega: So so basically here was my approach to the MBA. I would tell every professor at the beginning of the semester, like, look, I'm taking your class, but I'm actually starting a company. And the reason I signed up for your class is because I think your coursework can help me with my company. And so there's some homework I might not do, but, like, I will care more about your materials than anyone else. And so, long story short, there's a couple of things that happened. One, with Professor Mollick, I was able to get course credit for working for the university's comptroller, and so I actually worked for the comptroller and would have to prepare reports. And there were academic requirements for it. But my goal was to research and understand the space and how Controllership worked. In order to help inform my company with Professor Fader, I was like, look, the need for analytics and data and probability modeling and finance is massive, right? There isn't anything that, you know, I don't need to go take a finance class. I already worked in finance for a decade. I worked at Lazard doing M&A, for God's sake. You know, like, come on, what can I do? And he's like, well, guess what? I had actually prepared this curriculum for the dean, focused on data science for MBA students, but it got shot down because it was too heavy.
[00:06:26] Guest: Carlos Vega: And what we do here at Wharton is we create leaders of businesses, right? We don't create data nerds. And what ended up happening at the time is he said, take the curriculum, I will sign off on it, and you can create a customized degree path. And so I was doing all this innovation or entrepreneurship work with Professor Mallick. And these classes, quantitative classes from Fader's list combine them. And Pete Fader signed off and we made up a term on the sheet. We're like, what should we call it? Said innovation through analytics. And that was it. So now it is actually approved again. This was before. I know it sounds like a little over a decade ago. It sounds like ages ago. But now there is a focus on this at Wharton. It's one of the more exciting ones. And yeah, that's the story. Like I said, a little bit more than you asked for, but it was quite a roundabout way to get there.
[00:07:16] Host 2: Glenn : No. That's great. And it really, like you said, so many things that just were like making my brain fire. And this is what I've talked about all along. And, you know, the fact that you were, uh, working with the marketing professor because of quantitative marketing marketing jumped way out ahead of us because they, especially in e-commerce kind of businesses or SaaS or where they had more data like marketing, suddenly had all this data to play with. So I always think of financial analysts as the, you know, OG BI guys, but we just, you know, what was our data back? You know, back then it was well it's GL data and that's what we have. But I came up through Telecom where we had all kinds of data. And yeah so you know customer churn and you know and just in all the network outages and customer service calls and everything. So being able to use data in my financial forecast for me early on those were tied together. But um.
[00:08:08] Guest: Carlos Vega: I mean, the early, you know, you know, the early SaaS models were inspired by the telecom, like Zora's first customers, right? Zora was one of the first SaaS companies, uh, SaaS billing companies. Right. Like, their background in DNA has a lot to do with telecom, where you go look at utilities, these other industries that had that predictable like what is SaaS, right. Like the true innovation of SaaS is changing the way that Wall Street can value these companies because of its predictable cash flow model, right? Versus lumpy license fees that existed in telecom and utilities and all these other spaces way before. So it's funny you bring it up, but that's interesting. And that's where, you know, marketing had the frequency of monetary RFM modeling. Right. And that's existed since like the 70s. So it's like all these things that are now in tech and in SaaS and finance. Actually you're right. They come from these other, other areas.
[00:08:56] Host 2: Glenn : You caught the wave early on and you kind of read and saw where things were going. And I think at about the time you went back and got the MBA, that's probably when I remember the conversation. I don't. I know Elon Musk had a lot to say during that period, but there was a talk.
[00:09:10] Host1: Paul Barnhurst: Always has a lot to say.
[00:09:11] Host 2: Glenn : Yeah, but it was oh, you know, the MBA is a pointless degree. You don't need to get it. You need business experience and all that. So the fact that you knew going to get the MBA, it's like, well, okay, I'll get the key courses, but I'm going to chart this and craft it in a way that is going to propel your career forward. So that's a yeah, kudos on the foresight and ability to see and kind of read the tea leaves.
[00:09:34] Guest: Carlos Vega: I got lucky.
[00:09:35] Host 2: Glenn : Yeah.
[00:09:37] Host1: Paul Barnhurst: Well you know sometimes it's better to be lucky than good as they say, right.
[00:09:40] Guest: Carlos Vega: Yeah.
[00:09:41] Host1: Paul Barnhurst: There's a balance there. We want both. So you know, I know you have started a factoring company. And then after that, in 2015, you co-founded Tesorio. What prompted that company? How'd that come about?
[00:09:55] Guest: Carlos Vega: Yeah. So in factoring, I was working on that. I was at Lazard doing M&A on the side with a friend. And our whole thing was like, look, if you're familiar with factoring, it's when you're purchasing receivables at a discount, right. So someone's getting $0.95 on the dollar to get their money down. And our whole idea was like, look, we will help them solve their working capital hole, but help them understand their cashflow and create models for them so they can go get a proper line of credit. Everyone was like, just give me the money. I don't care which made it feel. Rather than helping businesses, it felt like payday lending for business. And so then when I went to get my MBA, I was like, look, I want to solve this problem. Like, clearly people have a hard time understanding cash flow. There's a whole bunch of reasons why I think even personally, it's kind of an enigma. But we wanted to help people understand their cash flow so they can make more sound financial decisions. Um, and that's, that's ultimately what led me down the path.
[00:10:50] Host1: Paul Barnhurst: Got it. So you found with factoring, even though you wanted to educate them and help get them off the factoring, it was almost kind of, so to speak, that drug. They just kept wanting to do it and do it and do it and didn't care about the education side. Why do you think that was? Do you think it was a lack of knowledge or just convenience? Or what drove most founders to just say no? And CFOs and companies say no. We're just good with the factoring approach.
[00:11:13] Guest: Carlos Vega: Yeah. Again, I think it's a fundamental thing where people don't understand cashflow like fundamentals well enough. Right? Like, I think a lot of people say things like, ah, you know, I just focus on growing my business and cash flow will take care of itself. Right? You go look at, you know, all the Harvard Business Review studies on Amazon versus Walmart, right. And why Amazon won because of their strong cash flow. Right. Or um, all these different things that exist for proof. Why optimized cash flow actually gives you a stronger business. And so, you know, we would hear stuff like that like, hey, I've got my margins are super high. I don't mind paying an extra 1 to 2% on, uh, you know, factoring versus what I'd pay on a line of credit. Like, it's fine, I don't care. Right? All these things that are just, like, nearsighted. And I think that was the gist of it. You know, folks were optimizing near term and not long term results.
[00:12:02] Host1: Paul Barnhurst: Yeah. Everything you said makes sense. And also, I'm going to guess that time off was probably early. Right. Right after the crash. Right. Early 20 tens.
[00:12:10] Guest: Carlos Vega: Yeah. It was uh, it was. That's about right. It was, uh oh nine 2010, 2011. It was in Panama. But still, that was a global financial crisis.
[00:12:18] Host1: Paul Barnhurst: So so you're on the company in in Panama. I was gonna say us you would also have really low interest rates. So almost free money.
[00:12:25] Guest: Carlos Vega: Mhm.
[00:12:26] Host1: Paul Barnhurst: Interesting. So how was that starting a company. You started the first one in Panama and then you did one here in the US. Is that kind of. That's right, that's right.
[00:12:33] Guest: Carlos Vega: Yeah.
[00:12:34] Host1: Paul Barnhurst: I'd love to know some of the difference. What were the biggest differences between founding a company in Panama versus here in the US.
[00:12:42] Guest: Carlos Vega: So I think the biggest difference was the nature of the company. Tesorio is a tech company. And that first company was a financing company. I feel like the biggest differences there are like at the time, the access to talent in the US was just phenomenal. The market's also a zillion times bigger. So there's a lot more potential. So any problem you find that might be you know, there's 4.5 million people in Panama. So you can imagine there aren't that many companies compared to the US, right? More people commute into New York City every day, um, than are in the entire country of Panama, to give you a sense. So, any small problem you find in the US is going to have more of a market than down here. I think that was a really big difference. Yeah. But as far as starting a company, uh, Panama was actually surprisingly straightforward to do so. But for this type of company, I don't think at the time I could have started it in Panama. Now things have improved. Uh, there's more access. Remote work is actually a thing now. Like, you know, I'm. I'm in Panama right now. Um, and most of my team is across 11 different countries in the Americas. I think now might be a different story.
[00:13:49] Host1: Paul Barnhurst: Yeah. Thank you. So question. I have just I know you started kind of cash flow forecasting. Did you ever think of doing more of the whole planning and budgeting or kind of how did you think about that as you're building a tool? Because I know there's a lot of similar integrations you're doing, right. You're connecting to a CRM or an ERP tool or a bank. And so what was the kind of reasoning to say, look, we're just going to stay focused really on 100% kind of optimizing that cash flow versus more general budgeting and forecasting.
[00:14:22] Guest: Carlos Vega: Yeah. So that's a that's a great question. And it's tied to a couple of things. Like if you take a step back, the way we view things is that cash flow is actually more of a data problem than a finance problem. Right. Like the data exists out there for every bit of cash that's going to move. Uh, as an example, like the, you know, just a little twist of thought here is like, you know, every dollar that's going out the door when it's going to go out the door, whether you're paying a vendor, you're paying an employee, you're paying rent, you know, you get to choose when that money goes out. But then you're sitting here trying to figure out your cash flow based on when you're going to get paid. But guess what you are is someone else's AP. So there is some wear out out there in some system, someone's head, some email like it exists somewhere. Exactly. When you're going to get paid. If you can get to the right person, the right system, you can know with exactitude, right? Like what your cash flow should be. And that's kind of the point, right? That's why that's why we say it's more of a data problem than a finance problem. And so then you start to think about that. It's like, all right, if I'm doing fpna, you know, forecasts and things like that, the execution of that forecast happens somewhere else across myriad teams and departments, right. If I'm working on cash flow and, you know, money goes out the door or it doesn't.
[00:15:37] Guest: Carlos Vega: It all comes back to your bank account. And the execution of that usually goes back to some system in the office of the CFO. And so the reason why cash flow was more interesting to us is because it's more of like a contained ecosystem. It's not saying that overnight you can map the whole thing, but like let's say you're looking at a are forecast of when my money's going to come in, I can quickly double click into that. Look at exactly what are the invoices that are predicted to come in, have that connected to my email and a dunning system and a team collaboration system, and instantly select those and say, hey, can you send reminders, send the reminders myself or basically the actionability of it, right? And so at the end of the day, cash flow is much more actionable. And we always talk about insights to action or keeping the data flows and the workflows interconnected. We've always talked about that. But now with AI it's even more important, right? Because now any action you take, you have correlated. You have perfect attribution, if you will. Right. So you have the action correlated to the Y and then also to the outcome. So now you can start to optimize that better over time. So it's actually if anything I'm even more excited to be focused on this space and not the other one, because I think there's even, you know, with greater connectivity potential and greater actionability, there's a lot to do. It's pretty exciting.
[00:16:55] Host1: Paul Barnhurst: Ever feel like your go to market teams and finance speak different languages? This misalignment is a breeding ground for failure in pairing the predictive power of forecasts and delaying decisions that drive efficient growth. It's not for lack of trying, but getting all the data in one place doesn't mean you've gotten everyone on the same page. Meet QFlow.ai, the strategic finance platform purpose built to solve the toughest part of planning and analysis B2B revenue. Q flow quickly integrates key data from your go to market stack and accounting platform, then handles all the data prep and normalization. Under the hood, it automatically assembles your go to market stats, makes segmented scenario planning a breeze, and closes the planning loop. Create air tight alignment, improve decision latency, and ensure accountability across the team.
[00:18:03] Host 2: Glenn : I was going to say you're I feel like preaching to the choir here, and I'm getting excited and jumping up and waving my Bible around here. But, um, so I'm winding up right now to just lob you a softball question. But as you go through this, I'm, I'm thinking, you know, put on your Verne Harnish cap and the Cassius King mentality. And I've worked with so many like time and materials billing companies that do not understand if you're managing your business through your income statement, but your DSO is 70 days or something. I mean, you're looking at data that doesn't tie to the reality of how you run your business. And, you know, and I work in the SMB space a lot. So it's a lot of businesses that they're not sitting on, you know, massive piles of cash. They're not exactly, you know, paycheck to paycheck, but it's very tight. So as I think about all that and the kind of problems I've had to solve, it seems like it's always difficult to shift people from that income statement mindset to the cash flow mindset. And all this is exposition. As I wind up, the softball question for, you know, is and I think from what you just said, that the biggest challenge that finance teams or the companies themselves are facing is sort of what you've chosen to focus on in cash flow forecasting. But do you? I guess what I would ask is, do you think cash flow forecasting is something that's maybe underlooked, and that this is the sweet spot where tesorio can really excel and that's why you feel good focusing there. Or do you have any anecdotal, you know, stories of working with, with your clients where you've kind of given them this insight?
[00:19:41] Guest: Carlos Vega: Yeah. So yeah, we definitely need to have an offline convo and nerd out about cash for sure. Um, but where I would say is, for us, like cash flow forecasting is just this outcome that occurs once you're like, running your business properly, it's more like predictability. I kind of joke around like, you know, forecasting is for the weatherman, right? Like, they can afford to be wrong. You can't with cash, or you can only run out of money once, right? So, so, like the thing that we're very focused on is like the building blocks up to a forecast. And that's why we started with the order to cash process, because that is a one part of cash that like I was saying earlier, you don't control, right? And that's the part where like you do have to stitch together more and more systems and more and more workflows or, you know, human actions that have to be the data has to be structured so that it can then inform the, you know, what is going to happen there. And so to your point like that, that's, you know, like to give an example, right. Like you can have a team that's trying to forecast cash flow. And they're going to look at typically let's say they're forecasting in your collections. They're typically going to go and look at your billings forecast, apply a DSO assumption and then look at your agings and say, okay, what percent from each bucket do I think I'm going to collect this quarter, sum those two different areas up and like here's my collections forecast. But what do I allow you to do.
[00:21:01] Guest: Carlos Vega: Right. Like we say there's like three main buckets. It allows you to clean your data, allows you to structure unstructured data and allows you to use granular data. Right. So on that bucket. Right. Like what if you could go and look at all of the data that exists in your system from order forms, signed contracts or you know, whatever. See that the data that's in, let's say, your NetSuite and your Salesforce or your QuickBooks and whatever CRM you're using is accurate, right? Like, the sales guy didn't screw up the billing and the shipping address or something like that. All the classic things that happen, right? Or the Po didn't get missed, right? Like they actually added all that stuff. So your data is clean. Great. Then there's structuring your data like you're exchanging emails with your customers and you're saying, hey, when are you going to pay me? When are you going to pay me? Like, I promise the money will get there by this date. Then you know the classic promise to pay dates, right? Or, you know, you're submitting it to Cooper or Reba, and they're sending you an email back that's saying, like, your invoice was rejected because you forgot to fill out this field. That's a classic. That's the worst, because the clock doesn't start on the net terms until they approve the invoice, right? So you have to jump on those ASAP right before because you know it's net 60, net 90 or whatever. And that clock isn't going to start until the invoice is perfectly submitted. So you have to make sure that's clear so you can structure that information right.
[00:22:17] Guest: Carlos Vega: And get that into a system that automatically creates a task for the person to go address it or I. That system has an agent that automatically goes and takes care of that for you. And then there's like the next step which is looking at granular data like, so you take all of that information I just mentioned and you want to come up with a collections forecast. It's like, let me look at every single invoice that every single customer has ever paid me. Train a model for that customer. And then for every open invoice for all of my customers, predict when they are likely to get paid, given how old they are today already, and then do a bottoms up granular analysis of what my collections are going to be from all of my open areas. That was impossible to do before, but you can do that now, right? And I mean, these are the types of things we do. Right. And so it's a back to back to your question. Like that's where I think it's exciting to think about cash flow and all these things add up. Right. Like today we're very focused on order to cash. We have forecasting on the AP side as well. Not as developed as this. Right. But like there are other data inputs that we'll be adding over time, you know, from say other procure to pay systems or from we already do bring in bank transaction data and things like that, but that actionability I think that's where today we get the biggest bang for our buck and our customers do as well.
[00:23:31] Host 2: Glenn : What I love about this. So first off, there's the, you know, as my CFO hat and just thinking about the problems I've gone through with collections and getting people to understand and getting and especially uploading into all the various different invoice systems that, you know, just horrible nightmares. But then from a tech side, as you walk through all this and I guess I haven't thought about it enough, but you have an opportunity. There's a lot of businesses out there that are kind of AI washing their, you know, they're just saying, oh, it's AI, but it's really it's some kind of rule based system or, you know, whatever, whatever it is in the background or that, you know, there's vaporware, that there's nothing there. But with what you guys are doing, you have the opportunity to do cool stuff with classic AI, that machine learning and doing predictive analytics using the data and, you know, helping forecast that way based on training on data. But because there's the communication piece and all that. I could see really practical benefits with generative AI in what you're doing too. And I'm wondering if you're leveraging both, you know, kind of the classical machine learning algorithms. And if you have found practical applications of generative AI yet.
[00:24:36] Guest: Carlos Vega: 100%. So I mean, we started the conversation with a concentration in business school, right? Innovation through analytics. What I was doing with Pete Fader there was using RFM technology to try and forecast cash flow by looking at exactly what I just highlighted. Like if you can look at every single payment across every single customer, it's not too different from what you know marketers do, which is look at the buying patterns of consumers on a website or things like that. Right. And so the idea that the switch that I was working on with Fader was like, can you apply like that's all based on how each human has a different buying pattern. You know, they talk about a Poisson distribution, like it's like a coin that you flip. Do I buy or do I not? Right. And so it was like, can you assume that of organizations because organizations have a culture too. And like as a group, they have a pattern that can be, you know, inferred as well. That was like the leap of faith that we were testing, right, that I was working on with Fader. Um, and it turns out there is. Right. And it's kind of interesting. And so like all along, I mean, before this AI stuff, I'm, I love that you separated machine learning from Gen I because we've been doing machine learning for for a decade. Right. Like at Tesorio. And we found that sort of pattern. Right. And just for an interesting point of view, we found that invoices have the similar type of modeling and behavior as like, uh, as humans and, uh, life insurance. It's a survival function. And what you're forecasting is like how long they will die. Dying means they get paid, right? And so, like how late something is already will determine how much longer it's going to live or how much later it's going to exist.
[00:26:13] Guest: Carlos Vega: Kind of like a human right. Like how old you already are is the biggest input on how long, how much longer you have to live. And so, um, it's a similar thing with invoices. So that was a fun insight. But then on the AI side, we found tons of applications. The one that we did within like over winter break of 2022, like, you know, Christmas break, my co-founder, the natural thing like let me have a gen AI like push a button, auto write my email for me. Like that was the basics, right? But the more complicated use cases which are more exciting are already alluded to in a few of them. Let me read through all of my communication with my customers and summarize, like every engagement that I've had with them, to understand if I'm a sales rep or a customer success rep or anything, I can now use all of my financial information about that customer to inform my decision on the next upsell or the next renewal, right? Or if I'm forecasting my collections, let me extract all useful information from all the communication structures and automatically update my forecast. Right. That's something else we're doing. So it's like the structuring of unstructured data I say, is one of the most powerful uses of AI, along with some other things we'll announce soon about like more like the operator from OpenAI recently. You know, that launch where there's like an agent that's actually not navigating a website for you and doing things? We'll have some pretty cool announcements coming up soon.
[00:27:34] Host 2: Glenn : There's one that I was thinking of, and this came from Professor Malik months ago in his One Useful Thing newsletter. I don't know if it was his original research or if he was quoting someone else's, but that AI is significantly more. These LLMS are significantly more persuasive than, um, than humans. And the study was people who believed conspiracy theories when they presented them to an AI. The AI could be more successful at talking them out of a conspiracy theory than, uh, than a human would be.
[00:28:07] Guest: Carlos Vega: That's amazing. I missed that.
[00:28:09] Host1: Paul Barnhurst: I missed that one as well. I'm gonna have to go look that one up.
[00:28:13] Host 2: Glenn : But my brain went too. I wonder if an AI would be more successful in convincing a slow paying someone, that slow paying an invoice to pay better because I used to. And one of one of the businesses where I was, uh, was the CFO. I wasn't just a consulting CFO. I was the person who was doing collections, would sit outside of my office, and I would hear her have the conversations with the payables person on the other end, and just hear those conversations. And how interesting would that be if you had, whether voice activated or via an email, a way to have an AI assistant trying to collect for you?
[00:28:46] Guest: Carlos Vega: Well, let me let me not freak you out too much. The other day, in about five minutes, I created some other player in the space who was working on AI collections agents. Right. And you've seen the AI and all this that are out there getting a ton of funding. I was like, this does not seem that hard because back to cash flow. Cash flow is deterministic, right? Like if you're trying to sell something to someone, there could be 1,000,001 excuses why they're not going to buy. But if someone owes you money, there's a contract that tells you exactly when they're supposed to pay you, why they're supposed to pay you, when they got notified that they had to pay like all the information is there. So I was like, hey, you know what? I think that can be injected into a prompt. So I took it, and it took me five minutes. I turned on advanced audio on my phone, and I was like, you are a collections agent, this and that. You're going to have a conversation with me to try and convince me to pay. Here's the information on the invoice.
[00:29:34] Guest: Carlos Vega: Remember to be polite, but you have to be very persuasive. That's all it did. I could play the recording for you. It was. It took me five minutes and I had an eye collector. And so, just for what it's worth, I had a similar hypothesis to yours, and I decided to test it, and it worked really well. It even offered me a payment plan when I started pushing back, because I was being very difficult on purpose to see what would happen, and offered me a payment plan. And then I said, like, wait a minute, how are you giving me a payment plan you're not authorized to do? I was like, no, no, no, you know, I will have someone from my team follow up with you in order to whatever. Wait, but you're going to receive payment. I'm not going to give you a payment. Like. No, I didn't say I would, I, you know, we have your email on file. Like it was. So. And I didn't train it to do all of this. And so to your point, it's, um, the capabilities are amazing.
[00:30:24] Host1: Paul Barnhurst: Exciting. Uh, I tried one today called Bawdsey, where they called, asked a bunch of questions about what your business needs are, and they go through LinkedIn and they basically say, do you need me to connect you with so and so and so I mentioned a podcast and they brought up some senior financial analyst. And do you want me to connect them to you as a potential podcast guest? I'll go ahead and put together the email to connect the two of you. And I said no at the time because I was like, I'll use this again later, but it's a three for about a 4 or 5 minute phone call, maybe a little longer. And at the end of it, I had an email with the person's LinkedIn profile and offered to connect me.
[00:31:02] Guest: Carlos Vega: That's incredible.
[00:31:04] Host1: Paul Barnhurst: Pretty crazy what you could do. I know one guy that's already got business from using it. Okay. That's awesome. All right. I think we're going to move on to our kind of fun section here at the end where we get to ask a couple personal questions. So how this works is we feed the questions we had for you and your profile into copilot, and hope to come up with some kind of fun, thought provoking, get to know you questions.
[00:31:25] Host 2: Glenn : Copilot did this.
[00:31:26] Host1: Paul Barnhurst: No, no. Sorry. Chatgpt not.
[00:31:28] Host 2: Glenn : Okay.
[00:31:30] Host1: Paul Barnhurst: And so we have 25 questions here.
[00:31:32] Guest: Carlos Vega: Oh my god.
[00:31:33] Host1: Paul Barnhurst: And we're not going to go. All of them. Don't worry. We're just going to do two. You get to either pick a number between 1 and 25 and I'll ask that question. Or I can use a random number generator to pick your question.
[00:31:46] Guest: Carlos Vega: Do random go for it. Let's go full AI.
[00:31:50] Host1: Paul Barnhurst: It's always fun to see what people say on that. We get a good mix. All right. You get question 13. And I have no idea what that question is.
[00:31:57] Guest: Carlos Vega: That's the, uh, the day I got married, June 13th.
[00:32:02] Host1: Paul Barnhurst: There we go. All right. Congratulations. But that is not the question. Many entrepreneurs have rituals or habits that keep them grounded. Do you have any personal routines that help you stay focused?
[00:32:13] Guest: Carlos Vega: I mean, it sounds lame to mention, but I really like playing tennis. Um, and sticking to, like, a sport like that, that's competitive, but it's enough good pressure. I find that to be incredibly helpful. But we are talking about AI. So something I am working on on the side, uh, right now, just for fun, is, uh, I'm trying to create an AI chief of staff that will be plugged into AI, use tools to, uh, take notes during all my meetings. Right. Where? Remote company. So they're all on zoom, so it's very easy to do that. You know, there's, you know, my calendar, there's, uh, different aspects, you know, different slack channels and things like that which, uh, which we can track. And basically having a, uh, a tool that can automatically keep track of, you know, key items, you know, help me create, like operating cadences and things like that. I think, uh, I think that should be pretty exciting. So that's, uh, one thing I'm working on on the side for that.
[00:33:04] Host1: Paul Barnhurst: Cool fun.
[00:33:06] Host 2: Glenn : Super cool. That's, uh. Yeah, I'm trying to figure out how I could have even. It didn't even have to be the level of a chief of staff. If I could just have some kind of, uh, executive assistant, um, to help with all that. I'm constantly, like, trying to figure out ways. And I've played around with operators. I did one kind of cool thing with operator. I had this list. It was just in a note file, just a text file of all these upcoming speaking engagements. And I was like, I kept thinking, oh, I've got to go put these in my calendar. I had an operator go through and this was a great use of operator. I had it go through, find the text, and then I opened up my Google calendar, you know, so it had Google Calendar open and then my list, and it went and created all these items in my calendar for me. And I thought, and it took forever, but what did I care about? I was off doing other things while I was doing work. While you sleep.
[00:33:53] Guest: Carlos Vega: All you care.
[00:33:54] Host 2: Glenn : Yeah, yeah. And then I realized, wow, this is what operator you know, it's still buggy and early, but that's going to be RPA for everyone when operator comes out or when it, you know, when it finally reaches maturity. So, um, all right, so I take a slightly different approach than Paul. We've tried we were doing a bake off for years, but for years in, in or in I years it's for a couple.
[00:34:19] Host1: Paul Barnhurst: Couple weeks I years.
[00:34:23] Host 2: Glenn : So you know we've done Claude, we've done Gemini and we and ChatGPT. And so while um, Paul does the random number generator, I just have the, uh, the GPT spit out and we'll have we'll see what it comes up with. Okay. If you could have. All right. If you could have lunch with any financial leader, past or present, who would it be and why? And I wonder at Wharton, you may have had you may have already had lunch with some pretty cool financial leaders, but, uh.
[00:34:50] Guest: Carlos Vega: Yeah, interesting financial leader.
[00:34:53] Host 2: Glenn : Is the default answer here, just Jamie Dimon. Is that what everybody says or. Yeah.
[00:34:58] Guest: Carlos Vega: So it's funny. Like I.
[00:34:59] Host 2: Glenn : Feel.
[00:34:59] Host1: Paul Barnhurst: Buffett's in that list too.
[00:35:01] Host 2: Glenn : Yeah yeah yeah yeah.
[00:35:02] Guest: Carlos Vega: I feel like I, I don't know the answer to the name, but like, I've always been intrigued by how early Amazon was on. You remember their letter to shareholders? I think it was either 040 6 or 0 eight. I don't know, it was like, I don't remember when it was, but when they announced to the world like, look, we don't care about profits. We're not going to be profitable anytime soon. Stop asking us to do it. Free cash flow is the only thing that matters. The textbook definition of the value of your equity is future cash, or the present value of future free cash flows. Right? And that's all we're going to care about. That's what we're going to optimize. But guts. Whoever was there in that room, I don't know if it was Bezos himself or his CFO at the time who, like, gave them the guts to put forth that letter and then actually execute that strategy. That's who I would want to meet, because now it seems obvious in hindsight. But back then, I mean, that was a different ball game and it changed, allowing them to do everything they did. That's always been super inspiring to me.
[00:36:05] Host 2: Glenn : Great. Great answers. Way stronger answer than I was expecting. Way better than just saying Jamie Dimon or Warren Buffett. So great. And actually, while you were saying it, I just googled it. I found the letter. It was zero four. So now I'm going to go.
[00:36:16] Guest: Carlos Vega: It was oh.
[00:36:16] Host 2: Glenn : Four because it really was. Yeah. So thinking back, I mean, it was you know, it wasn't until AWS really took off.
[00:36:22] Guest: Carlos Vega: Not the thing.
[00:36:23] Host1: Paul Barnhurst: Exactly. I mean, 21 years ago, which sometimes seems like forever, and other times it doesn't seem very long.
[00:36:29] Guest: Carlos Vega: Yeah, yeah, yeah. And not small. But how young Amazon was at the time. Right. Like now they can do whatever they want. They could dictate to the market. Anything back then took some guts.
[00:36:40] Host1: Paul Barnhurst: They say, hey we're not going to be profitable for the next few quarters. We're doing this. There was one. Okay.
[00:36:44] Guest: Carlos Vega: Yeah. Eps was the way everyone cared. Everything cared about is they were like, we're not going to give you any earnings per share, so stop asking. Like, I mean, think about it. That was amazing.
[00:36:53] Host1: Paul Barnhurst: Great example. I really like that one. Well Carlos, we've really enjoyed chatting with you. It's been a pleasure. It's been a lot of fun and best of luck with Tesorio and thanks again for coming on the show.
[00:37:04] Guest: Carlos Vega: Awesome. Thank you so much. I really appreciate it. I always enjoyed the conversation. And just to leave you with one, I feel like I need to repeat it every time I get a chance, but it came from one of our customers. He said, remember, revenue ain't real till you get paid. And that's basically the gist of a lot of what we do here. So I just wanted to leave you with that one. I know we're connecting on all the potential cash conversations we'll have later on, which we definitely should pick up on.
[00:37:29] Host1: Paul Barnhurst: Alrighty. Thanks. I love that one.
[00:37:31] Host 2: Glenn : Thanks, Carlos.
[00:37:32] Guest: Carlos Vega: All right. Take care.
[00:37:34] Host1: Paul Barnhurst:Thanks for listening to the Future Finance Show. And thanks to our sponsor, QFlow.ai. If you enjoyed this episode, please leave a rating and review on your podcast platform of choice and may your robot overlords be with you.