APIs You Won't Hate

Show notes* Sandeep Dinesh + LinkedIn + GitHub + Twitter * Mercoa + GitHub - mercoa-finance + LinkedIn

Transcript [00:00:00]

[00:00:00] Mike Bifulco: Hello friends. Welcome back to APIs you won't Hate. My name is Mike Biko, your co-host and APIs you won't hate. Co-founder leading you on this journey through API Tech in the world. I am super happy today to get to sit down with Sandeep Dineh from Meco to talk about the product he's building. And I think this actually may continue a bit of a weird streak I have where I have some lots in common with the folks who have been interviewing on the show.

[00:00:22] Maybe that's a bit of self-selection, but in this case, I think Sandeep is working on something that is. At least tangential to a world I used to live in . And he also happens to be a YC co-founder. Sandeep, thank you so much for joining me. How are you doing today?

[00:00:33] Sandeep Dinesh: I am great, Mike. Thanks for having me on.

[00:00:35] Mike Bifulco: Of course. Yeah. It's, my privilege and pleasure to be able to chat with you.

[00:00:38] Why don't we start here? So you, you are founder of mea. Tell me about Meco. What's the elevator pitch?

[00:00:43] Sandeep Dinesh: Yeah, so Meco is an API for accounts payable. And so a lot of folks don't know what accounts payable is, but when you're a business, you got bills to pay and usually when you're at a certain scale, I. You have approvals and invoices and vendors to manage and this whole process that your [00:01:00] accounting team is probably really familiar with.

[00:01:02] And so we help our customers launch accounts payable products to their customers.

[00:01:08] Mike Bifulco: Got it. So a bit of a, whether that's a B two, B2B, or B two B2C or how, how does

[00:01:12] Sandeep Dinesh: B two. B2B. Yeah. So I think the best way of looking at that is Stripe has Stripe Connect really popular product for marketplaces. So let's say you have like Uber drivers they need to get paid. So you pay, I. Uber and then Uber pays out to the drivers. That helps 'em get paid. We help make payments out.

[00:01:32] So you have a bunch of customers that have their own vendors and they're managing their whole back office on your platform. We help you launch bill pay and accounts payable so your customers can start paying their bills on your platform.

[00:01:45] Mike Bifulco: Sure. Okay. That makes a ton of sense. And I think especially for someone my, like myself with a background primarily in engineering and less so on the finance and business side of things, I'm always fascinated to hear a bit about the history of how you got there and like what, what caused this to [00:02:00] happen.

[00:02:00] And I also think we just hit a world record or maybe a, a show record for quickest time to me needing to give a disclaimer that in a past life I worked for Stripe on the developer advocacy team. Had a great time there. It was a good job. I'm no longer at Stripe. But it's probably just worth calling that out to begin with since it'll likely come up here and there during the conversation.

[00:02:17] So, Sandeep, with that being said, tell me about how you got here. What, what caused you to you know, take a path that had you build, has you building an accounts payable product?

[00:02:25] Sandeep Dinesh: Yeah. So in order to answer that correctly, I think we gotta go back around 10 years. So 10 years ago I pretty much just graduated college. I was working at my first job and I was really bored. And so I actually met my current co-founder at that company. And we did a few hackathons together and we did our first startup way back in like 2014. That crashed and failed real hard. But it was a great learning experience. And we've been working together for the past 10 years. But after that I actually joined Google on the developer advocacy team for Google Cloud. And so I was our [00:03:00] early hire on the cloud advocacy team there. And yeah, that was a great place to start my career as an engineer.

[00:03:07] You know, I've been coding for a long time, building websites since I was a kid and all that kind of stuff. But I think just learning how some of the best developer advocates go ahead and talk about APIs and developer tools. And kind of really seeing Google Cloud grow from Yeah. They have a cloud platform, I think, to being in the same conversation as like Azure and AWS.

[00:03:29] Was a really cool experience. And so I did it about for about five years. And right before the pandemic me and my co-founder launched a fully bootstrap startup. So beginning of 2020. So we built that for about a year. And then I joined a small byc company as a lead engineer. And so I learned a lot about. Kind of the, the venture world and shipping fast and kind of. Doing that kind of stuff. And then I joined Stripe as well. So yeah, I think a [00:04:00] lot of career similarities with you, Mike. And yeah, so at Stripe I was a engineer on the revenue recognition team, and so we were building some cool products for accountants.

[00:04:10] And so while I was there at Stripe, my co-founder was actually a product manager@bill.com. And so he was working for a company called Invoice to Go that did invoicing for small businesses. They got bought by Bill and one of his projects@bill.com was trying to build Bill dot com's account payable into invoice to go's platform.

[00:04:32] So basically trying to turn bill.com into an API into an embedded, uh, API, and then embed it into like an internal project for that. And the thing that he really realized was other companies were coming to bill.com and asking for this kind of functionality. And bill.com really couldn't support them because their organization wasn't designed to sell APIs and developer tools.

[00:04:59] And so we [00:05:00] started MEA because we saw this opportunity of like, if we are API first, if we're developer first, can we go ahead and help people launch their own accounts payable products? In a, in a, in a way that kind of competes with all these different AP systems out there.

[00:05:14] Mike Bifulco: I'm really interested in all of this because it seems like there's a, there's a real confluence of things happening where you're both working at companies where, you know, opportunities are ripe in, in sort of the payments world, especially around this time period. Right? The past few years has been pretty crazy and, and all this stuff.

[00:05:27] But also to be to have you at Stripe where developer journey is everything. And your co-founder at a company where they, they're desperately in need of a solid developer journey, feels like the, the birth of your product is a sensible outcome. From there I'm curious for yourself and your co-founder did you either of you have sort of like business education background or were you both sort of traditional engineers going into this?

[00:05:48] Sandeep Dinesh: So my co-founder has an economics background. He's not an engineer. Lot of product experience, sales support, customer service. So he's, he's had the whole go to market kind of [00:06:00] experience from a career standpoint. For myself I technically have a business background. I graduated with a degree in marketing and computer science.

[00:06:08] I don't know how useful that degree in marketing was.

[00:06:11] Mike Bifulco: You a job in developer advocacy, I think it probably went a a really long way. Yeah.

[00:06:15] Sandeep Dinesh: Yeah. So I think that really helped getting that job as a dev in Dere. But yeah, I think so to your point about kind of things just kind of converging together, I think learning about developer tools and dere at Google, learning about the accounting side of things at Stripe having my co-founder also super interested in FinTech.

[00:06:35] Having the startup experience prior getting into yc, all these things kind of just combined in a way that made starting meco pretty, pretty obvious to do. Yeah.

[00:06:46] Mike Bifulco: So you've, now come together with the idea you have people with, interesting backgrounds that sort of knit together to build something that makes a whole lot of sense. What was your story for chasing down your first actually, maybe let's talk about it this way. Your first use case, like what was the first thing you built and then [00:07:00] how did you convince your first users to jump on?

[00:07:02] Sandeep Dinesh: Yeah, so we started talking to customers before we had anything built. I think. We've gone down the road of like building a bunch of stuff and then figuring out you built the wrong stuff plenty of times. So we started having phone calls with folks through, in our network, through LinkedIn.

[00:07:18] We've been very fortunate that we've done a lot of people coming in talking to us through LinkedIn. So yeah, we had a lot of folks, we talked to them, try to really understand what problem they're trying to solve what's important what's difficult about this. And we kind of sprinted to create a demo.

[00:07:35] And so we had like a nice happy path of like, you can embed this into your, into your product and now your customers can start paying bills. And that first demo was basically an I frame that you drop in and the word I frame kind of like burns the ears of many folks out there. They hear that and they're like, I don't like that.

[00:07:55] But it was kind of the fastest way for us to validate that. Hey, drop this in, see if it [00:08:00] works. We had an API, it was hand documented. Hand rolled pretty bad. But, you know, we just wanna validate does this, does anyone even care? And so we started getting folks who were interested. We started signing some pilots.

[00:08:16] And then we got started getting really serious about how do we deliver this product. So that was kind of the beginning. And I can go into more, but yeah.

[00:08:23] Mike Bifulco: Yeah. Okay. I'm, I'm already hearing things that make me think that maybe having a marketing background was probably pretty helpful there. But, but maybe also having built and been through the ringer with. PR past companies, especially bootstrapped ones is really interesting there. I talk about this a lot and I almost feel a little I don't know.

[00:08:38] I, I always feel like it's hard to really land this with people until you've actually experienced it in some way. But many first time founders will go and build a product with an idea that is, I. Whether good or not not told to enough people and then they don't go and chase down enough feedback before building the thing.

[00:08:53] And

[00:08:53] Sandeep Dinesh: Yeah,

[00:08:53] Mike Bifulco: you know, the, the trope for me is someone who talks about like, well, it's not ready yet. I haven't shared it with anyone 'cause it's not ready.

[00:08:59] Sandeep Dinesh: Never [00:09:00] ready.

[00:09:00] Mike Bifulco: just now Yeah, it's never ready. Nothing is ever done, ever. That's the whole whole thing with software. What you've just described for me is it.

[00:09:06] Entirely the inverse of that, right? Like we didn't build anything. We went and talked to a bunch of people first, and then built something that was no, nobody's favorite idea, but at least better than what existed before. That, that's a skill that takes some time and is definitely a, a nuanced approach, especially as compared to, I don't know what I would call maybe the traditional, like first time founder thing.

[00:09:24] And so. When you do that, what are the signals that you were looking for to even pursue further? So I'd imagine you got some early signals that people were using it and liking it. Is it, Hey, I want more, or, Hey, we think we could make this better for you, or, or find a thousand more of you, or what was, what was what you were looking for at that point?

[00:09:40] Sandeep Dinesh: Yeah, so it's actually really funny. I think we went about Meco in a way. Completely different from our other products. So I think one thing is we are building financial infrastructure. So that takes quite a lot of time to build and it takes quite a lot of time to build trust in that product, right? So like, we're moving millions of dollars.

[00:09:57] You're not gonna wanna just [00:10:00] trust who, random guys who build a product in a weekend to do that, right? So I think for us, we were really trying to figure out. Is there market demand for something like this? First? Can we find those people in a, in a repeatable process? And then what are those people's biggest like need?

[00:10:19] Right? And so at the end of the day, our product is pretty complicated. We call it like the eight components, and each one of those components can be used a la carte. By itself together. So it's a pretty like complicated platform. So the first thing that we really wanted to do was validate, Hey, are there people out there?

[00:10:38] This is like a market that even exists. And I think we did that by just trying to talk to as many people as possible in the industry. Are you building this yourself? We got a lot of, yep. This is on my roadmap for next year. For next year. I don't have any resources against it, but I wanna do it. And we're like, okay, that sounds really interested.

[00:10:55] It's like, do you wanna partner with someone? It's like, yeah, absolutely. If someone would build [00:11:00] this for me, I would definitely wanna talk to that person. We're like, okay, that's great. And then the last challenge was how do we find these people? And I think for that it was really a struggle. 'cause there's no like, list of vertical SaaS companies and that's who we primarily sell to is vertical SaaS.

[00:11:15] So it's really hard to find those folks 'cause there's a bunch of founders starting them all the time. They have a lot of deep expertise in like, let's say the construction space or dental or home services, right? And so they maybe they've worked in that industry for a while. Maybe they've run Airbnbs and they wanna start a business to help other folks in that industry.

[00:11:35] Like how do you find these folks? Was really difficult. And for us, we really try to solve that by being the most helpful folks. Out there. So a lot of people starting these companies don't have FinTech or finance backgrounds. And so we started writing content on how do you monetize this kind of like platform.

[00:11:51] What are your customers looking for when it comes to like payments and just trying to be really helpful. And so we actually wrote a 28 page guide [00:12:00] on how to build Meco from scratch. And so it's like. If you wanna just build this yourself here is like the PRD for you. And yeah, we got a lot of hits on that.

[00:12:09] I think people were like, this is really cool. It was just a Google doc. And so that was, that was our first, I think, real sign of, okay, there's something here. Let's invest more.

[00:12:17] Mike Bifulco: I think that sends a strong signal too, that it's like, well, we don't have secrets. This is just a hard thing to do. Like you're, you're welcome if you want to go and, you know, go on this journey. But it's probably the sort of thing that you need a few people and many, many person hours of engineering time to get through.

[00:12:31] And even then your industry background probably helped quite a bit there

[00:12:35] Sandeep Dinesh: Yep.

[00:12:36] Mike Bifulco: Right on., I'm super into that as a path to go and very, very much a show of strength there too. Like, cool. Here, have all our secrets. Good luck. So where, where does Y Combinator come into the picture?

[00:12:46] Sandeep Dinesh: Yeah, so I was at Stripe. I. Basically just joined. I was about almost like a year in I felt like I was fully onboarded at this point. And my co-founder was, Hey, let's apply to yc. And we would apply to YC plenty of times for other [00:13:00] ideas and like, Hey, let's throw in an application. So we threw one in very last minute.

[00:13:04] And then we got in, which is a very happy surprise. We took the weekend and we're like, yep, this is, we've been waiting for this opportunity for a long time. I don't wanna live my life thinking, what if I think this is a huge opportunity. Like, let's go. They're gonna give us some money, let's go try it out.

[00:13:21] And we just hit the ground running really hard. So we quit our jobs and started building, started talking to customers. Basically like as my co-founder was lining up these meetings trying to validate that idea like I was talking about. I was like, what is the smallest demo we can throw in front of people to kind of give them an idea?

[00:13:38] So we did Figma to start and then very quickly moved to like a clickable demo. And that was all before we actually started the, started yc. So we were winner 23 and so we were doing this in like November and December. And so YC started in January and, I think within the first month we signed our first contract.

[00:13:58] And so that was like a big validation for [00:14:00] us.

[00:14:00] Mike Bifulco: yeah, of course. Again, to, it is funny. I don't know this off the top of my head, but to clarify, winner 23 is. Starting January, 2023 ish, or was that into, okay.

[00:14:10] Sandeep Dinesh: yeah, yeah. First three months of, yep, yep, yep. Mm-Hmm.

[00:14:14] Mike Bifulco: And so you've, you've obviously come a long way since then. One of the things that I found valuable when Kraftwerk went through Y Combinator was we were really focused for a few months during YC on set, set an ambitious and, and hard goal that is meaningful and do everything you can to build a narrative towards that.

[00:14:30] It sounds like you, you headed into YC with a working clickable demo and were maybe chasing down customers. What was sort of your North Star at that point?

[00:14:37] Sandeep Dinesh: Yeah, so we wanted to close 10 customers, very aggressive. We did not get there. We got pretty close. I. I think again, our goal through that three months was can we kill this idea of like, what do we do to lose conviction that this is the business we wanna build? I think both of us are, you know, kind of mid-career.

[00:14:55] We both are married and all that kind of stuff, and so it was like. [00:15:00] You know, it's not our first rodeo doing a startup. We know how difficult it can be and how easy it is to lose motivation. And so it's like, let's do that fast rather than slow. So I guess like fail fast. I don't know if you wanna call it that. But yeah, I think that was kind of the whole goal was like, I. Can we just lose conviction on this idea and kind of just shut it down? And everything that we did went the other direction of like, oh wow, okay. We've kind of gained conviction. And I think at the end of that, we were like, yeah, this is what we wanna spend a long time building because we think it's gonna be really big.

[00:15:35] And I think we have the right team to do it. And so that was kind of our internal goal at yc, obviously. I think they talk about this a lot. You know, you have the two week check-ins with the group partners. And so we had pretty aggressive goals on trying to close new customers and leads. And so yeah, that was kind of like the external goal, but I think really internally it was, can we kill this idea?

[00:15:54] And not even launch it.

[00:15:56] Mike Bifulco: Sure.

Debate me!

[00:15:57] Mike Bifulco: You're absolutely the first founder I've talked to who is, [00:16:00] is going about this by way of like the debate me route. Like, you know, convince me this is a bad idea. I, I really like that, especially in the case where like you're, you're being forces you to be maybe intellectually honest with yourself which can be really hard to do.

[00:16:12] A lot of, a lot of people get almost delusional about their idea and we'll go a really long way before they decide that like. Oh, you know, it turns out the thing that I'm building is just not a big enough molehill to be scalable at, at the scale that we need it to be. Especially once you start taking on investment, that becomes a much, much bigger molehill to tackle too.

[00:16:30] Okay. So you finished YCS program at the end of last winter, so March-ish of last year. What, what's happened since then?

[00:16:37] Sandeep Dinesh: Yeah, so I think we've learned a lot about what is kind of the, what is an MVP versus an MLPA minimum lovable product. So I was talking about the iframe. Yeah. No one likes that. Our API has gotten a lot more complicated. So I think you've had Danny from Fern on on this podcast before. So we are big users of Fern.

[00:16:57] We use 'em for server side, [00:17:00] client side and docs. And so we basically moved our whole. API over to Fern during the batch. And so we were batch mates together, so very glad I met them early in the business. So I think one thing that we've learned is just how important developer experience and API documentation is for our sales process.

[00:17:20] And so I think a lot of folks in FinTech the bar is very low for good APIs in FinTech. I think Stripe is like way up there. And everyone kind of assumes that everyone also has like stripe level documentation, SDKs, all this kind of stuff. And no, it is like pretty bad. And so I think like the bar is low, but we wanna go really above and beyond to kind of provide that really good experience. And I think the big reason why is, well twofold, one. We sell to product teams, but a technical co-founder or VP of engineering or lead engineer, whoever it might be, depending on company [00:18:00] sides, does have a very big say in the go, no go. Right? If they look at our SDKs and APIs and they're like, this looks like my team is gonna struggle to implement, I don't think they're gonna be serious about this.

[00:18:13] We are not gonna win that deal. Right? So like the features and all the check boxes might look really cool, but at the end of the day, because we are a embedded product, we are API first, we don't make money until you start using us, right? And so like that implementation's all about the technical kind of coupling of our service to yours.

[00:18:33] So the API SDKs are a huge part of our Go-to market movement. So that was a big learning. I think we. We're trying to build as fast as possible and the docs went outta sync and they didn't look good, and our guides weren't great and we invested a lot of time into like trying to make that onboarding experience really nice.

[00:18:53] And so that was a big thing. I think Fern was a big help there. So for folks who don't know what Fern is, it [00:19:00] basically lets us define our APIs in a Fern yaml. I think they also support open API specs and it actually generates our server side routes and client SDKs for a bunch of different languages.

[00:19:12] As well as our docs. So every time we make a change, all three of them are always in sync. And so we can get basically guarantee to our customers that like, everything is well documented, always up to date. And it's like what you, what you see is what you get and people really like that. So that was a big part of our investment.

[00:19:29] The other side on the dev side was our front end. So we talked about the iframe, iframes. I, I love iframes. They let you move really fast. Uh. With things like JWTs where you can pass in custom props, you can actually customize iframes pretty easily. I think a lot of companies that deliver their solution with iframes do that, but I think a lot of her customers were like, this is even with, this is not good.

[00:19:52] I don't like, it's a black box. My, like, observability tools don't show me what's going on. If a customer has a question, [00:20:00] I'm just kind of stuck on you guys to, to figure that out. And so we had built that iframe with react components underneath the covers. And so moving that from internal React components to an external library was a really big deal.

[00:20:15] And, you know, I think like react components as a delivery mechanism. I think like Algolia does a really good job. It's a company called Quill. They do embedded analytics. They does a really good job, but it's kind of like a, a new. Way of delivering an embedded product. And so.

[00:20:31] Mike Bifulco: think so. Yeah.

[00:20:33] Sandeep Dinesh: There's really not that much like best practices out there.

[00:20:35] So we've had to like learn a lot. So we're using storybook for a long time to do kind of a documentation, but that's really great for like UI libraries and that's not really what we're doing. And so we're actually migrating to a fully manual documentation for that where we put our components, we talk about how you can use them.

[00:20:52] All of our components take can take in a child component. Then we just pass you a bunch of hooks, right? Like that's not really a [00:21:00] common use case for like UI libraries. So I think like that's been a big learning of like, how do we help our customers move fast? And that's the reason why they wanna partner with someone like us and not build it in-house is time to market, right?

[00:21:13] And so I think the backend is really complicated. That's, you know, kinda expected. We have a lot of good documentation tooling for APIs. But then the front end, the react component side, I think that's kind of like a new world that we're exploring and trying to build a lot of tool that kind of tooling ourselves. And yeah, so that's been a big learning for us.

[00:21:32] Mike Bifulco: Yeah, that's a really fair point. I don't think I've really heard it put that way, but it's definitely something I've felt maybe as a beneficiary of it, as an end user of these sorts of things. That many of the SaaS products that have come to love are ones where I essentially consume it through NPM install.

[00:21:47] And become an end user that way. And I can think of a few off the top of my head. Clerk does authentication very much in that way where they ship a bunch of React components that you just toss it into your React app and suddenly have authentication. Post hog does a really good [00:22:00] job. They have some really nice React libraries.

[00:22:01] They'll let you add. Product analytics and do a whole bunch of tracking and ab tests and all this stuff that you know, in prior days would've been a much deeper build out with loads and loads of code and docs that you could imagine, and certainly a few others that I've used. And there's something to be said for that too, that, that is scaling your developer experience by way of taking advantage of the massive amount of time and love and care that folks from these other companies are putting into their.

[00:22:24] A deeply focused slice of the thing. You know, like authentication is something that I never wanna become an expert on. And I'm sure I've said that on the, the podcast a billion times by now. But being able to effectively hire out smarter people than me for a price that scales with my success is like, sign me up, let's do it.

[00:22:39] That's great.

[00:22:40] Sandeep Dinesh: The interesting thing with these like embedded, so clerk embedded authentication like the thing there is, it's the same for everyone, right? Like almost 90% of what you're trying to build with these embedded providers, it's the same as everyone else.

[00:22:53] And I think that's just kinda the same thing for us. We're say like, you know, 80% of bill pay is like the same. It's an accounting process. Like there are gap [00:23:00] principles that everyone follows to like pay their bills.

[00:23:03] Mike Bifulco: Right.

[00:23:03] Sandeep Dinesh: Just add your secret sauce on top that's like in specific for your industry or your vertical or like your, your customer base.

[00:23:09] But why are you trying to reinvent the wheel? And I think that's kind of like the, the bread and butter of embedded SaaS, right? We're fully white labeled. You don't know that we exist, but we let you move fast and like not worry about like the things that you don't wanna worry about.

[00:23:22] Mike Bifulco: Yeah. And again, to that point, you, you meco is successful only when your customers are, and there's an alignment of incentives there. That's really interesting too.

[00:23:30] Sandeep Dinesh: I'll try.

[00:23:30] Mike Bifulco: Another case where maybe solving the boring problem, so to say is great for everyone involved. You can stay focused on your thing much the the way that, you know, say a clerk would be focused on authentication.

[00:23:41] So I wanna talk a little bit more about how Meco is built. So Fern is, is super interesting because they give you so much of that sort of love and care in the world of SDK Cogen. So now Meco is available sort of. Pseudo natively in a bunch of languages, I'd imagine as, as installable packages, SDKs, whatever you wanna call 'em, and you get documentation out of it and [00:24:00] all that.

[00:24:00] What, what is the underlying system that you're building then that that speaks to Fern? Like, what's your architecture look like, maybe what's the size of your team? Those sorts of things.

[00:24:07] Sandeep Dinesh: Yeah, so bit of background on me At Google Cloud. I was on the Kubernetes team, so microservices were like my bread and butter. And so very specifically not doing that with Meco. So team is very small. It's basically on the engineering side. Right now. It's just me. So we've kept the team pretty small on purpose to move very fast, kind of find product market fit, and then start scaling it out.

[00:24:29] And so it's pretty simple. We are running on Google Cloud Run, so fully serverless. And our database is on CockroachDB, which is also fully serverless. So that's for folks who don't know what CockroachDB is. It is a Postgres compatible database kind of based on Google Cloud. Google Spanner and Spanner is a.

[00:24:48] Horizontally scalable asset compliant database that uses like crazy atomic clocks and all this kind of stuff to keep transactions always in sync as you scale out. And so we've kind of [00:25:00] chosen technologies that let us scale by default because I don't wanna deal with that. So really simple. It's a monolithic TypeScript, no js backend. We use kind of. The, the normal tools, GitHub and Sentry and all that stuff for monitoring. But yeah, we kept it very simple just so we can move fast. I've worked at teams where, you know, everything was split out into its own microservice and it just kind of, as a small team, very hard to deal with that.

[00:25:24] So yeah, we just kinda, how can we move really fast and how it deliver a product that, you know, customers really want. And then we can think about breaking it down in the future.

[00:25:34] Mike Bifulco: Right, and you mentioned you, you're the only engineer, is it just you and your co-founder currently as well?

[00:25:39] Sandeep Dinesh: Yeah, full time. It's just me and my co-founder. Yep.

[00:25:42] Mike Bifulco: Amazing. Definitely very small and lean and mean at that scale. And what I, if you're listening to the podcast, one of the things I would encourage you to do, which I think is maybe a good data point to like put side by side with that is me's site. Does not convey a two person team even slightly. It looks pretty, pretty intense.

[00:25:59] There's a lot going [00:26:00] on there. So you know, a, a applause is due to you and your co-founder. There's loads of stuff there, and I'm sure you're standing on the shoulders of giants in many ways by using lots of interesting tech there. But it speaks to what you can get done with, with some focus and maybe a bit of a marketing polish on things too.

[00:26:15] That's, that's super cool.

[00:26:16] Sandeep Dinesh: Thank you for the compliment. I think, yeah, definitely building on the shoulders of giants. I think like we try to use as much tooling as possible. We try to partner with vendors that can move us really quickly. But I think at a small, small team size, you know, our secret sauce is just customer support.

[00:26:32] And I think YC says this also where it's like, what can you do that like Stripe or Google can't? And that is just like. Really, really focusing on the customer and really delivering solutions to them. You know, we don't have a lot of red tape. If someone really needs something, we can ship it tomorrow. We wanna move very fast.

[00:26:51] We wanna be very thoughtful and, back to the original point of no secrets. I think like even if you don't end up working with us, we still want to make you successful. 'cause maybe in the future [00:27:00] you will, maybe you won't, but at the end of the day, we wanna move the whole industry forward. It, it's really painful when you see folks still using like pen and paper or checks or spreadsheets to do their finances.

[00:27:10] Like can we just move the world away from that is kind of like the goal, right? Like maybe it's us, maybe it's not, but let's move the world forward. Is kind of like the big picture. Yeah.

[00:27:20] Mike Bifulco: Yeah, you, you've painted a very cool story as well around like having a focused idea and building something that addresses a direct problem that you understand really well. Often on this show we talk a little bit about, like, talk about a problem, a project, a product. Fern was one, Danny's been on the, the show where we've talked to him before about what they're doing and sort of the pitch for API developers at that point is something along the lines of like, how do you know if you are someone who should be interested in using Fern?

[00:27:45] I would imagine there's probably a very direct story there for Meco too. So for, for a, for devs who are used to integrating with APIs and building products what's a sign that Meco might be solving problems that they should be interested in?

[00:27:56] Sandeep Dinesh: Yeah, so I think the big one with MEA is if you're trying to [00:28:00] build the financial back office for your customer. I think a lot of folks start with invoicing to start, right? So helping your customers get paid. Then kind of the next natural thing is banking. So helping them store those funds, manage those funds and then you can go into payroll, you can go into spend and expense management, and you can go into accounts payable.

[00:28:18] And so basically we wanna help you create that financial one-stop shop for your customer. And so that's money in money management and then money out. And so we plug into that money out piece. And I think the reason why a lot of folks don't really understand the opportunity in the money out side of things is I think when you're thinking about money in, it's like, Hey, Stripe charges me 2.7% plus whatever.

[00:28:44] Okay, cool. I can charge like a fee for invoice processing, right? Like. That's pretty like cut and dry. Like I don't understand how that works. When I'm storing funds, you know, I can make some float interchange interest maybe I can like give you some like credit cards or debit cards to spend and I can make something called [00:29:00] interchange on those cards.

[00:29:01] But then when you get a invoice for a thousand dollars, you gotta pay a thousand dollars, right? Like there's no middleman to take a fee there. And so I like think a lot of people are like, there's no opportunity here to like, monetize the payments. Maybe I can charge like. Some SaaS fees, like 20 bucks a month or something to my customer to like have them manage their bills through my system.

[00:29:22] But kind of the secret here is that all these big account payable companies have found is your customer who's using your product, really wants to keep cash with them as long as possible, right? But the folks they're paying wanna get paid as fast as possible. So we basically have a mismatch in, in incentives here where.

[00:29:44] The payer wants to hold onto money as long as possible, and the vendor wants to get that money as fast as possible. And so you can offer financing on both sides of that equation either accelerating the payment to the vendor or delaying payment from the payer to help both sides manage their cash flow.

[00:29:58] And that's really [00:30:00] where the unlock comes in on monetization.

[00:30:02] Mike Bifulco: Oh, super interesting. So you're, you're running financial mechanisms underneath this then that people are keying into on both sides. Really cool. Wow. And so, man, I, I that's a part of the world that I know enough about to be scared of. Does this mean you also then had to go chase down, I'd imagine compliance and certification and all sorts of things to be able to offer those.

[00:30:20] What's that journey look like?

[00:30:21] Sandeep Dinesh: Yeah, so I think we have a lot of flavors on how we do this. So right now we're using a lot of partners to kind of do those services. And so our big value prop is. one is gonna take your fancy financing product until they have a workflow that matches what their accounting and finance teams really want to use.

[00:30:38] Right? So we spent a long time kind of building those workflows, those react components, those backend APIs so that you can deliver a product that'll actually be used, right? Because if you just have like a. Pick a financing button, like it's not really, it doesn't fit into that workflow. So our first step is kind of building something that your customers will actually use.

[00:30:57] And then we work with a bunch of vendors to provide [00:31:00] that financing, to provide those payment rails. And eventually we'll bring some of that in-house. But I think our big value prop is we will go ahead and integrate with all of these platforms. And so then you just have one API to implement, drop it in, and then you get best in class payment rails out of the box.

[00:31:15] Mike Bifulco: Yeah. Nice, clear story there too. That's fantastic. Okay. I do wanna ask, maybe the inverse of the question I asked before is you know, what, why would an API developer recognize that meco is something they should be interested in? I also think there's a good number of people that listened to this show that may be sitting in shoes similar to, to yours from a year or two ago, where they may be getting signals that they're.

[00:31:36] Experiencing things or learning things at work or seeing problems that might make their way into a good company at some point. Do you have any advice for people like that? For how to test an idea how to, how to prove something out as a business

[00:31:48] Sandeep Dinesh: That's a great, great question. I think experiencing that pain yourself at a big company definitely a good sign. You know, at Google and Stripe, we have so many internal [00:32:00] tools that like I wish existed outside. Good, great example of that is, you know, both companies have like a launch process that you had to go through to launch anything public. And it's pretty, it's, it's, it's one's process, but it's like really well thought out. Like on the Stripe side, like if you do anything that touches an API, someone from the developer relations team will come review, make sure it's consistent across the board. So like that kind of stuff, like is there a product that you could build externally?

[00:32:25] There's also really simple things like, I think there's a company called Go Links. We use 'em at mea I think like Google had that originally. And it's a whole company now where you just like go slash and you put a link name in and then it redirects you to a page and it's so useful and you're like, I had this at Google now I don't like, let's go build a company about this.

[00:32:41] And so I think like, how do you validate? That was kind of the question that you asked. I think having it yourself is great. And then going to, talking to your friends, going and talking to your people in your network, right? Like if you're an engineer and you're building a developer tool, I'm. I think you probably know some other engineers, like either work at your company, [00:33:00] work, went to school with you, you worked previously with Met at a meetup.

[00:33:04] I don't know. Right? Like, but you probably know somebody, so just go talk to them. People like talking about their problems. I think that's something that especially first time founders are really afraid of. They're afraid of getting, being rejected, being like, I have this great idea. I don't wanna talk to anyone because what if they say, say it's bad? And for me it's like, that's the best thing then I don't know how to waste my time building this thing anymore. Right. Because you know, it's so easy to burn out building something that no one uses. Like you can spend years and years and years. Building something and then you launch it and then you, it just crickets and then it just feels super bad.

[00:33:37] And then maybe your next idea is gonna be the billion dollar idea, but you're just so tired from working on something that no one used that you don't even try it out. Right. And so my advice to anyone kinda listening who wants to dip their toes in startup is just talk to people. Validate it. Get your idea in front of as many people as possible.

[00:33:55] No one's gonna steal it. In fact, they might give you ideas on how to make it better. Right. And [00:34:00] honestly, being part of yc something like that I didn't realize before I joined was, you know, it's a filtering function for really great founders and we're, we are fully in person in San Francisco.

[00:34:09] There's a lot of other YC companies here too. There's a lot of events where we meet up and I think talking to other folks, they give us ideas that we're like, oh my God, this is gonna. I didn't think of this. This is like a easy thing, low hanging fruit that just unlocked like some like big win for us, right?

[00:34:26] And yeah, just sitting in your, in, in your room coating is not going to to help you. You do have to do that. A lot of coating, but you also had to get out there. Yeah.

[00:34:37] Mike Bifulco: Yeah, that's, I, I will pile on and add to that, that at some point it's not about coming up with good ideas, it's about being able to execute on them and identify the problem space that matters. And one thing that I wish, like I get a lot of folks who ask me questions about starting a company or they wanna talk about an idea they have.

[00:34:54] And I, I always wish I could talk to people like a year or two or three or four before they, they start with [00:35:00] pursuing a business to, to plant a seed with, with you, whoever you are, wherever you are, that one of the most important things you can do is learn how to find. People find the network, find the problem space, get embedded in the conversation.

[00:35:11] And for a lot of developers who are maybe a little more introverted or you know, reticent to communicate openly online that creates a problem. If, if you woke up tomorrow and said, I wanna start a startup, and you have. No you know, no following on LinkedIn. You don't have a public GitHub, you don't have a website, you don't have anyone who you talk to about these things.

[00:35:28] You're starting at a massive disadvantage. So I'm not saying you need to go and be an influencer, like, I don't even think that's even remotely the answer, but I do think you should be able to participate in open source, right? You should be comfortable with jumping into GitHub conversations, looking to Reddit for answers, helping people out diving into things and, and learning.

[00:35:45] And you'll start to knock out the bad ideas just as well as meeting people who have really interesting, good ideas and. Don't have time to pursue 'em, you know?

[00:35:52] Sandeep Dinesh: Yeah. I think just to add on to that one, if, if you're doing a startup because you were like, I wanna try a startup, may, maybe not the best reason. [00:36:00] I think like

[00:36:01] Mike Bifulco: Yeah, yeah,

[00:36:02] Sandeep Dinesh: the, to your point of like being in these communities, like you probably wanna care about this problem enough that you're already there or like.

[00:36:10] You want to, you like, you naturally want to go and like help those folks. And then like the startup comes after, right? Where it's like, oh, I see there's a bunch of people to help. I can probably build a business to help these people. Then it kind of makes a lot more sense versus like, I'm gonna take this and shove it down their throats and make them buy for me.

[00:36:28] Maybe not the best way to do it.

[00:36:31] Mike Bifulco: For all but the most extreme outliers, your startup idea at best will take 10. 10 years, right? You'll be working on a problem and it has to be something you can be interested in for that long. I. Odds are, it's probably already something you're interested in or at least related to a problem you've had that would, would make life better.

[00:36:46] Otherwise you're you know, punching from the back foot or whatever the phrase is there. Sandeep be, before I let you go, I want to hear a little bit about what's coming next for Meco. So what are you working on now?

[00:36:56] Sandeep Dinesh: Yeah. So we are relaunching our React documentation. [00:37:00] So stay tuned for that one. We're pretty excited. We're. Always trying to improve our developer experience. So that's a big one. We have a few cool payment partnerships coming out too, so hopefully we'll be announcing that. But yeah, aside from that, you know, just come check it out.

[00:37:13] If you are at all interested in FinTech APIs developer experience send me a note on LinkedIn or Twitter. Happy to chat. However I can be helpful, I'm happy to help.

[00:37:25] Mike Bifulco: I appreciate that. I will make sure I put both your LinkedIn and your Twitter profile in the show notes here for folks to find you. And where's the best place to find Meco online?

[00:37:34] Sandeep Dinesh: Mea.com. This is a, it's, it's exactly how it sounds. It's spelled M-E-R-C-O-A. And then we do a lot of content on LinkedIn. So we're gonna start doubling down on that as well. So if you want any of our content, you can probably find it there. But yeah, our website is probably the best place.

[00:37:50] Mike Bifulco: Amazing. That will obviously also be in the show notes Sandeep Dinesh from mea. Thank you so much for chatting with me. I really appreciate it. It's been fantastic to learn from you and really cool to hear about your story. Feel [00:38:00] free to join us anytime to give us updates on where things go between now and then.

[00:38:04] Sandeep Dinesh: Thank you. Yeah, Mike, this was great. I had a great time, so I really appreciate you having me on.

[00:38:08] Mike Bifulco: Of course. Take care.

[00:38:10] Sandeep Dinesh: Thanks.

APIs You Won't Hate: Recent Episodes

APIs You Won't Hate

Debate me!

Introduction and Guest Introduction

The Future of Voice AI

Wapi's Elevator Pitch

Challenges and Opportunities in Voice AI

Building Voice AI Solutions

The Genesis of Wapi

API Design and Developer Experience

Docker vs Kubernetes: A Developer's Perspective

Hello World with Wapi: A Hands-On Experience

The Developer Experience: Making APIs Accessible

The Future of Dev Tooling with AI

Use Cases and Real-World Applications

Technical Deep Dive: Building with Node and Kubernetes

Team and Future Plans

Conclusion and Final Thoughts