Transforming Financial Operations How AI is Changing the Game with Laurent Charpentier

Episode Overview

Laurent Charpentier is the CEO of Yooz, a pioneering company specializing in cloud-based accounts payable automation solutions. With a robust educational background in electrical engineering and computer science from the prestigious Massachusetts Institute of Technology (MIT), Laurent has leveraged his technical expertise to drive innovation in the fintech space. His leadership is marked by a commitment to integrating AI and machine learning into financial processes, transforming the way businesses manage their accounts payable operations.

Laurent’s journey began in the south of France, where his early love for swimming led him to compete at a high level, eventually joining the varsity swim team at MIT. This experience not only shaped his discipline and work ethic but also instilled in him a deep appreciation for teamwork and leadership. After graduating from MIT, Laurent began his professional career at Accenture in Paris, where he gained invaluable experience in IT transformation and ERP implementations, setting the stage for his future endeavors in the fintech industry.

Under Laurent’s leadership, Yooz has grown into a leading provider of innovative financial solutions, helping businesses streamline their accounts payable processes through the power of AI and automation. His vision for the future of fintech is rooted in a deep understanding of both technology and business, making him a respected figure in the industry. Laurent’s ability to blend technical knowledge with strategic leadership has been instrumental in driving Yooz’s success, positioning the company as a leader in the rapidly evolving world of financial technology

Episode Transcript

Kevin Rosenquist: Hey, welcome to Pay Pod, where we bring you conversations with the trailblazers shaping the future of payments and fintech. My name is Kevin Rosenquist. Thanks for listening. Laurent Charpentier is an MIT grad, a competitive swimmer, and now the CEO of Yooz, a leading provider of cloud based accounts payable automation solutions. They use AI and machine learning to automate accounting tasks, freeing up accounting teams and improving efficiency. We discuss AI and automation, where fintech is heading and how to be a great leader. Please welcome Laurent Charpentier. You have a degree from MIT in electrical engineering and computer science. You’re also on the varsity swim team there. Come on. Laurent. You’re not supposed to be super smart and athletic. You need to save something for the rest of us.

Laurent Charpentier : Thanks. All I can say is. Yeah, MIT was a great time and an amazing experience. And, yeah, one of my wishes was really to combine, you know, the best academic experience I could ever dream of. Really. and, and while practicing, you know, my sport. So it gave me a great way to do both. And that was an amazing experience and one that really made me even more attracted to the US and the culture, the business aspects as well. So it did play a role in the, in the next steps of my life.

 Kevin Rosenquist: Did you were you a swimmer your whole life? Was that always been a big thing for you?

Laurent Charpentier : Yeah, it started pretty early on. mainly because we were living, pretty close to, some kind of a reverse. My parents were, like, afraid we could go play by and drown. So. And, you know, I love the competitive aspect to it. And, and, yeah, it’s, it’s been probably, 20 years, 20 years worth of swimming 10 to 15 competitively at different levels, including MIT, NCAA. So a lot of fun.

 Kevin Rosenquist: Good for you. That’s awesome. Yeah. I have a three year old and I was quick to get him in swim lessons too, for the same reason. He’s got he did not drown the earlier the better. Yeah, exactly. That’s kind of how I look at it too. Yeah. and you grew up in France, correct?

Laurent Charpentier : Yeah, in the south of France. Born and raised there. And life took me after high school to Lyon, for engineering school. and from there I transferred to MIT to get my bachelor there. So that was the university career, I would say. And then after that started my professional life at Accenture in Paris, did, some consulting in the what I like to call the IT transformation world, a lot of ERP implementations, a lot around SAP, some Salesforce, etc. and that really shaped me in terms of methodology of work, etc. Accenture was a great first experience. And from there it led me to use any particular.

 Kevin Rosenquist: Reason for MIT or just because it was prestigious in a great school. Was there any particular thing that brought you there?

Laurent Charpentier : Well, yeah, we I try to have, when I apply to US universities, I try to, you know, maximize my chances. So I did apply to the traditional exchange programs of my engineering school in France. But I also, you know, thought maybe, maybe I can get even more than that. So I applied to schools that were extremely strong in swimming, you know, kind of D1 schools to strong that I had very strong academics. And MIT was, you know, kind of the dream for an engineer. And the swimming program was also actually amazing. It was a D3 program, but made a lot of great friends. We went to nationals both years actually. I think the team just got better and better over the years. So they were in the top three several, several years in a row in the past few years, I believe. So that’s, that’s become a great program for D3.

 Kevin Rosenquist: Yeah. That the Summer Olympics are coming up too. And they’re, they’re in Paris. Too bad you’re not back there. You could go.

Laurent Charpentier : I actually will be back. Oh, you will in July. I didn’t I didn’t score tickets for the Olympics. So, I’ll be watching from my TV, but, yeah, it’s going to be exciting.

 Kevin Rosenquist: All right, well, if anyone listening to this podcast has some pull, get get some swimming tickets for the Paris Games, please, please. Yeah.

Laurent Charpentier : And we have, we have a strong, strong French talent who actually, you know, trains in the US, right? Under, former Michael Phelps coach. So I’m excited to see what this young French talent is going to do in swimming. I think he’s going to crush it. So let’s root for him.

 Kevin Rosenquist: Yeah, I love the Olympics. I’m excited. I always get excited about them. I’m always up late watching sports I never really knew I liked, you know? Yeah, it’s just fun. It’s fun.

Laurent Charpentier : It is, it is.

 Kevin Rosenquist: All right. So when you graduated from MIT, did you ever imagine you’d be CEO of a fintech company?

Laurent Charpentier : No, not at all. I mean, you know, I knew a lot of business, but, I also love technology. And, you know, I think I think this is my strength or my value. So, you know, I didn’t imagine that, but I did love Accenture. And, you know, the fast progress that I made, within Accenture, you know, you have a clear path to, to progress and, and get toward management and then even even more, after that. And when I left Accenture to join the EU’s adventure, he was use was launching its first branch, international branch. So that was the US. And I thought, man, that’s a great adventure to be part of. There is really something to build on the US market. and this is, this is what drove me more, more the challenge of building something new than than the potential outcome of leading as a CEO. One day I just just kind of happened over the course of the user journey.

 Kevin Rosenquist: Do you still get to be techie or are you too busy running the show?

Laurent Charpentier : A little less than before I try, I try, I definitely try to stay on, you know, dangerous on the product and then make sure that, you know, I know, I know more than than most for sure, but a little less than before. But I’m, you know, I’m very passionate about the product, so I stay very involved on in terms of roadmap, innovation, what we can do around AI, etc. and we can we can expand on that during our discussion.

 Kevin Rosenquist: Yeah, I’ve talked to people on the show before that start in the tech world or start as, as programmers or engineers and work their way up to leadership positions. And, you know, it seems like it’s a real advantage being able to talk the tech talk while being in your position is, is not always the case for CEOs. Do you find that makes your job a lot easier?

Laurent Charpentier : Yeah, absolutely. Because I’ve spent in my first years at Hughes, I spent a lot of time more on the tech side first. So, you know, pre-sales, understanding our clients needs, why they would buy our product, then implementing them, supporting them. And then this helps, you know, drive you the product roadmap based on what you hear from the clients. But also, you know, kind of the vision you have for the company. And they’re both very important. And from there, my background being tech condition for me to start leading the US side of the business and then later on taking the CEO role was, you know, let’s create a great team. And this is where I hired marketing talent, sales talent to compliment myself and really create that scalable machine or engine in terms of sales. And yeah, this has helped a lot because then you can really understand your audience and help people.

 Kevin Rosenquist: Have you always had a leadership mentality?

Laurent Charpentier : I think so probably comes from the, you know, the family entrepreneurial mindset a little bit. So I, I’ve always heard, you know, discussions about leading a company, leading a team, etc.. I had common topics around, family dinners, etc. because of the family. But so I think it played a role in that. And I think swimming also helped because it, you know, helps you get organized, helps you be part of a team. And I think that also gave me some foundations for that.

 Kevin Rosenquist: Yeah. It’s an underrated thing with sports because I agree on sports. The teamwork, you know, pressure, the adversity. Like a lot of the stuff that comes with playing competitive level of sports can translate and does translate to to the business world.

Laurent Charpentier : Yeah, yeah. And it plays a big role into into leadership even even though swimming technically is not. You know, I want to say it’s not a team sport per se because there is not a team play. But there are, you know, all the relays and all the swim meets where everyone has to score points. So there is a huge team spirit. Even though the sport itself is, is you being in the lane and swimming laps. But but the this plays a huge part.

 Kevin Rosenquist: Absolutely. And it’s still a team sport in the sense that you guys are a team and you’re backing each other, you’re helping each other, you’re motivating each other. So yeah, I think it still applies. I know what you mean, though. It’s not. It’s like a soccer team. It’s like a football team. But yeah, it’s still a team.

Laurent Charpentier : Yeah, there is no game, but. Right, right. It’s really. Yeah, it’s really the speed. But yeah.

 Kevin Rosenquist: What makes a great leader.

Laurent Charpentier : That’s that’s a great question. I mean obviously one key part I think is having a vision of where you want to go and being able to, you know, bring people to buy into that vision. And so this means, you know, the second part to that is obviously your team and your people because you need to, you know, bring them along with you. Make them believe into the project that they’re a part of. So this means listening to them, learning how to motivate them, etc.. So I think I would say vision and make sure you drive the vision with integrity, with accountability, and then people to build that, that culture that is key to success, in my opinion, has been something that that we’ve been very good at. I choose to, you know, create, create a place for people love to come to work, have fun, you know, kind of work hard, play hard mindset, trust each other, help each other. So you have a, you know, you foster innovation, you foster collaboration. And this creates a high performance team.

 Kevin Rosenquist: Do you have a strategy? when building an effective team, do you do you have any things you usually gravitate towards or is it just taking the best of the best?

Laurent Charpentier : To me it’s it’s a mix of both. I mean, taking the best or the best in terms of, you know, from a hiring standpoint, when you want to build your team is obviously important. We want to have a very high standard into our recruitment process, but we’re also very much about choosing to promote from within, right? Someone who has worked his way up and knows the business, knows the product is going to be someone we want to keep on board because, you know, it takes a lot of time to train people. And when you have good people on board, you want to retain them. So a promotion from within is equally, if not even more important than than just finding the best or the best. Obviously, when you grow and when you reach a scale like like ours, it’s a mix of both. And especially as you know, bringing someone from the outside can sometimes give you a fresh eye or a fresh perspective on some, you know, issues or things that you want to solve that maybe are new to you because, hey, you were 50 before we were 100, and now we’re almost 500 globally. So it’s we’re running into issues that we we never had before. Right. So but but they’re good problems.

 Kevin Rosenquist: Yeah. Yeah that is a good problem. So it’s like fresh perspective versus knowing what you’ve got. Like you know you’re always kind of you know, sure, you’re rolling the dice a little bit when you hire someone new, whereas if you hire someone from within, you at least know their tendencies and stuff. But you’re right about the perspective. Someone who’s been there a long time might just be more status quo than fresh ideas.

Laurent Charpentier : Yeah, absolutely. You know, it gives you that fresh eye, that fresh perspective that, you know, been there, done that, which is maybe what we want to accomplish next. And they’ve done it somewhere else versus, you know, promoting from within and creating that culture of where people belong and people grow with their, you know, work, family, and they call it this way.

 Kevin Rosenquist: So let’s talk about you. You guys are all about automating, streamlining the accounts payable process using AI, machine learning, optical character recognition. Talk about how you use these technologies to automate AP processes.

Laurent Charpentier : Yeah. The very, first way we use that and that, that’s really part of our DNA and, and what what made us different from, you know, very early on, both both in Europe and in the, in the US as well, is that we’ve applied AI, machine learning approaches to document capture and data extraction. And we’ve done that even way before. I was actually a buzzword almost when you didn’t want to tell it. I because people were like almost a bit afraid of it or what it was all I knew was Terminator two. Yeah, exactly. But yeah, we were pioneers in in applying AI in that, you know, smart data capture and smart data extraction. And our, our goal was when we built use was let’s, let’s make it a cloud based solution where every invoice captured and every optimization we’re making in the system will help all its, all of its customers, because we’ve taken that kind of global approach or multi-tenant approach thanks to the cloud. So it’s it was one of the very first application. And our differentiator was we are doing a real time data extraction with the highest level of accuracy, accuracy on the on our market segment without any human outsourcing behind the scenes. So you get a high level of accuracy, the highest speed, because you don’t have to wait 24, 48, 72 hours to get your invoice back for processing.

Laurent Charpentier : You get it in about 10s. And really we try to make the system read an invoice like a human person would. So it’s based on keyword and it’s not based on what used to be old school CR, where you have to define templates for each vendor. And it’s a nightmare in terms of maintenance, etc. and you have a lot of setup to do upfront. So our approach is to make it work out of the box and make it deliver the highest level of automation on that front. So that was the DNA. Obviously this has evolved where from there we’ve also added in more intelligence on the email capture channels on the way to split document batches on the way to classify documents, invoice credit notes, purchase order invoices, match those to POS and other types of documents, which is also growing. Growing flow of documents of other types of documents that also are being uploaded in use. And there are more applications that are more recent around fraud detection, outlier detection, etc. that are also very, very interesting for the future.

 Kevin Rosenquist: Because you’re using AI does do you have like a learn and adapt kind of, aspect of, of use? Does it learn kind of like some AI models do?

Laurent Charpentier : Yes. So there are different AI is a very vast, I guess, domain. So there are areas where, yes, there is, you know, self-learning based on, just a user feedback loop. And there are areas where we supervise the learning. So we don’t we don’t let that sit back happen completely automatically. We want to know what the system will learn. So, just to give you two example, we, we, we take a deep learning approach when it comes to document identification classification. So, what’s what’s coming in? Is that an invoice? Is that, a credit note? Is that a bank statement? Is that a contract? Because use can capture pretty much any type of document. And on this is a deep learning approach in terms of AI with a feedback loop, which is the more we ingest and the more users tell us, okay, you got it right or they make a change, then this feeds back into our AI model. On the other hand though, from an invoice capture and a data extraction standpoint, the self-learning approach, we keep it under control because we realize sometimes what the user inputs is wrong, even though the machine got it right, and we don’t want to learn that, because then essentially you’re creating a bad loop into into your system. And we see this, for example, sometimes where the accountant is going to add a prefix or a suffix or on an invoice number, and they may have a good reason to do that, but maybe they do that. And the ten other or 100 other companies that are using that same vendor or getting invoices from those same vendors, we don’t want, you know, to learn that across the board. Right? So we’re very careful on the learning capability and how and supervise it on, for example, the smart data extraction. So it depends what what I and what it’s supplied. What is supplied to.

 Kevin Rosenquist: Yeah that makes sense to from a from a compliance angle to I suppose it needs to be looked at, you know, any sort of learning because you’re in the financial world. So you want to make sure that you’re not doing anything. You’re not that it’s not doing anything. It’s not supposed to do or learn. Yeah.

Laurent Charpentier : Yeah, absolutely. And when we train our systems, the thing we try to minimize the most are the errors obviously. So we always track three things. The success rate which is what the machine got. Rate Error rate, which is what the machine got wrong. And we want to make that as close as zero as we can. And then the, you know the not available or nothing was nothing was automated or nothing was identified or and this is the part where we work on optimizing on daily basis.

 Kevin Rosenquist: Pretty much automating tasks is definitely one of the use cases for AI. That seems to be something that people are very interested in, that companies are very interested in a good positive use case. Why? Why is fintech such a great fit for AI automation?

Laurent Charpentier : I think a natural way, a natural response to that is the amount of data that there is around around finance, because obviously there is data coming from all the incoming documents that an accounting department or finance organization will will receive. And again I was invoices the natural the number one. But there are other types of documents that are part of that organization that are also, good candidates for, for automation. And second, all the data you’ve got in the books, you know, from a GL standpoint, are, I mean, are these key data where a lot of application can be can be done around that? From an AI standpoint? We were talking earlier about fraud or outlier detection. What the system the AI could tell you. Hey, usually every month you get an invoice from this vendor. This month you haven’t gotten it. And for an organization that doesn’t create POS, that’s key because then essentially it gives you your accrual or it tells you, hey, you may have something missing in your books. So things like this that are very interesting applications for fintech And all of that comes from obviously the data you you have and all the historical data that your accounts that your accounting system holds.

 Kevin Rosenquist: Optical character recognition or OCR. You mentioned a little bit earlier. It’s pretty incredible technology. And you made a comment about that. It’s gotten way, way better in recent years.

Laurent Charpentier : Yeah. So actually it’s not really the necessarily the OCR part that got better. It’s the AI part on top of it. So in our world, we’re pretty much telling my sales people or my tech people, hey, don’t don’t use OCR anymore because it’s not really something we need as much as we used to. Ocr is just optical character recognition. So it’s just a way to convert an image into text. Right? So you need that for invoices that are incoming in as images to make to make them text. And from there what’s what’s key and where our value is, is the AI layer on top of that. That will make sense of these texts that will understand, you know, it’s you know, there is an invoice date here. So this must means that the invoice data around that. And let’s go find that to automate that data entry and not have the user enter that date. So the AI part is really what matters today because more and more documents are coming in as text already. You get your invoices, your PDF directly via email, and it’s already text. So there is no need for that image conversion anymore. And this is really what OCR was or or is, but the need for it decreases. But the need for AI, which is technology of use is, is what’s going to be increasingly interesting over the next few years.

 Kevin Rosenquist: Well, yeah. What do you what do you see changing over the next few years? I mean, well, besides a lot, um, you know, obviously it’s just going to keep it’s going to keep getting more and more interesting out there in the world of AI. But we’re as far as, you know, fintech specifically. Where else do you see room for AI to come in and big changes.

Laurent Charpentier : I think with, everything becoming more digital, fraud and trust are going to be key applications of around AI. So I can do a lot of things which automate a lot of things, which is great, but let’s also leverage that to make sure that what we automate is as safe as we can and obviously still under human supervision. I mean, I really think AI is here to augment us and not completely to replace us. And I think someone in my network and I don’t remember who said that. So hopefully I don’t want to plagiarize the quote but said, hey, you’re not going to be replaced by AI. You’re going to be replaced by someone who uses AI better than you. Yeah, I.

 Kevin Rosenquist: Heard that too.

Laurent Charpentier : Yeah. And it’s, you know, if we think about it 40 years ago when Excel was released for the first, the first version of Excel was released. The a lot of the accountants were maybe thinking, oh man, Excel is going to replace me, because now you can do a lot of math or, you know, computation in just, you know, seconds versus doing them through a paper ledger. Just made the world change on how accounting works. And I think, I think this is going to be maybe with a bigger impact than Excel four years ago. But this is going to be something like that where it’s about learning how to use it to augment ourselves and make ourselves more strategic.

 Kevin Rosenquist: Yeah, you can say that about a lot of things, right? The the internet, computers, calculators, like any technology that ever came out, I’m sure there there’s in the Industrial revolution, people were, you know, were always thinking of like, oh no, well, these machines are going to replace me because they can do it faster and more productive and cost less and all that. But we always seem to adapt.

Laurent Charpentier : Yeah, absolutely.

 Kevin Rosenquist: Automation a lot of times is more associated with larger businesses, but you guys work with a lot of small and medium sized businesses as well. How can small businesses automate tasks efficiently? Like I think maybe a lot of small businesses might say, oh, we’re not big enough to worry about automation. Would you say that’s false? And why? Why is that?

Laurent Charpentier : Yeah, I think that false. It’s just the ROI may be a little different depending on if you’re a very small business doing, you know, 115 voices a month from a very large business doing ten, 20 or tens of thousands of miles. The business that does only a few may just have one person dedicated our time to accounts payable, but that doesn’t mean this person cannot save time on some tasks like data entry. Because we take we eliminate all of that we use. Or as the company is growing, do more with less, you know, so you’re going to double your small business size. You may not want to hire a second person for doing the accounting. So automating gives you that that way to, you know, without adding headcount. So I think beyond the pure automation and the pure the pure do more with less. There is also the data and access to real time data. That is key. Knowing who did what when. So you may be a small business with just, you know, 1500 employees in it, but having an approval process so you can make your audits easier at the end of the year, can be a big can make a big impact on that small business style. Obviously, for big companies that are doing tens of thousands may be a bigger impact on the bottom line itself in terms of absolute value, but also all of the benefits that I just mentioned in terms of real time data and all of that.

 Kevin Rosenquist: Oh, yeah. I mean, you think of all the mundane tasks associated with any finance department in any company. I mean, to be able to automate that. I mean, I’m sure it’s got to be got to be desirable for those people.

Laurent Charpentier : Yeah. No, absolutely. And for us, we really built a product that can address a broad spectrum of, you know, complexity of complex use cases. So we can address a very simple workflow for a small business. We can address a very complex workflow for a very large company, Multi-location, etc. so it’s all about understanding the client’s pain points. What are you trying? What are your pain points today? What are you trying to solve? And let’s focus on that. And part of the product in use that’s going to help you solve that. And the product is very flexible for that and very easy to customize. It doesn’t require a coding degree or long, you know, project with hundreds of service days. It’s a matter of a few weeks.

 Kevin Rosenquist: So as a computer scientist at heart, in the larger scale of AI, are you confident in the way it’s being developed? Are you do you feel like the big companies we all read about, are they doing it right? Are they being ethical?

Laurent Charpentier : I think so there is it’s hard to know. Obviously there are so many players and there are new players that pop up every day and some of them are getting large quickly. But overall I think there are some leaders out there that are putting an emphasis on, hey, let’s be ethical, let’s, you know, let’s be transparent on what is being done with the data, etc.. I think it’s key and definitely for someone listening to this and looking at, hey, what can I do with AI in my company? Or if I use products that have embedded AI in the in the use cases, I, you know, are they’re following those rules. Is it being developed internally? Is my data staying internal or is it going on the internet. Things questions like this are very important for anyone buying software to or to think about.

 Kevin Rosenquist: I think that just comes down to reading the fine print or how can people stay safe in this?

Laurent Charpentier : Yeah. Reading the fine print is a start. In some cases, it’s taking asking for, you know, a little more information about the security policy or, you know, having an audit report around, you know, the SOC, one SOC, two reports, things like this that are starting to be very commonly asked, at least in our in our prospect base or client base.

 Kevin Rosenquist: Well, Laurent, thanks for being here, man. I really appreciate your time and I really enjoyed talking to you.

Laurent Charpentier : Yeah. Thank you for having me. It was great. And, thanks for the opportunity. Absolutely.