Building a unified API on the shoulders of OSS with Robin Guldener from Nango

[00:00:00] Robin: Hi.

[00:00:40]

[00:00:40] Robin: Awesome. Sounds good. Yeah, so I started coding when I was 12. I think like sort of, you know, got out of curiosity and trying to start up when I was 16. I think I've been sort of in the, you know, startup and B2B SaaS space in a way ever since. I was, you know, writing iOS apps for a few years.

[00:00:54] Like I. Like built a little agency around that and got bored with like the repetitiveness and, and knew wanted to build more [00:01:00] product stuff. So right out of college I started a company called a with a couple of friends and it was basically like a fleet management software for B2B, a B2B for fleet management.

[00:01:10] And as part of this, we had to build a ton of integrations. So kind of, I would say like, you know, a third maybe of the product in one sense or the other was like, kind of integrations basically. And so we had like about 50 different integrations or, or different providers that we had integrated with.

[00:01:25] And it took up, you know, sort of a good chunk, I would say like 15 to 20% of our engineering resources for a good of the life of the company. And I was the head of product there. And I always thought like, that's crazy, you know, that we're spending this much time on sort of. Building and maintaining what was essentially kind of data pipelines.

[00:01:39] And at the same time, my CEO, you know, my co-founder, CEO, he was very annoyed that we were not, did not have more integrations because he sort of felt like, you know, a large chunk of deals sort of would not go through because we were missing integrations. And that's kind of how I realized that like.

[00:01:53] Integrations are just like, you know, super critical function inside of the company. Super critical function inside of the product that you know, it's kinda like [00:02:00] never done. And it's super annoying to maintain. It's actually like, takes a lot more infrastructure than, than you first expect. And so when I left that company, I, you know, knew I wanted to start something again.

[00:02:09] And I ideas for about six months.

[00:02:12] On, on this integration idea and realized that it's actually still an, an unsolved problem for a large chunk of.

[00:02:27] Yeah, I mean, it was a side job, right? You know, I was in high school at that point in time. I kind of started working for a few hours a week next to school for, for, for the startup and like you know, didn't, I didn't pay me a whole lot of money, but like, you know, it was a great learning experience for me.

[00:02:39] And I. Yeah, like a, a year or so in you know, the, the iPhone came out. It was like the first iPhone you could program for, like, it was like, I think the iPhone 3G or three s or something like that. And I asked them like, you know, could, like, could you like maybe buy me an iPhone as part of like sort of my salary and birthday present in one?

[00:02:54] And that's how I, you know, got to have an iPhone as 17-year-old and I never could have afforded this myself. And that [00:03:00] got me into sort of like iOS development, which led to me doing some. Did that for a few years at about, you know, 20, 25 apps in the app store. Was making decent money for kinda like a high school student, you know, side project.

[00:03:13] But I just realized it's like not, I guess a repetitiveness of, of the way I was doing it was not something that I really appreciated. It was a good lesson sort of in, in business and, and some finance, but like, I realized that I think like what I really enjoyed at that startup was like building product, having an impact on people's lives and being able to grow something from zero to to one and beyond.

[00:03:34] And I think that like, really, I guess it's like kind of the thread that goes through the rest of my you know, career at this point is like, sort of starting things and, and building things from [00:04:00] scratch.

[00:04:03]

[00:04:03] Robin: Yeah, I mean at NGO we really build a single API for all your integrations. And I know that's, you know, there's a bunch of companies out there I think that. They're claiming to, to give you that. I think the thing that makes us different is that we, instead of trying to pre-built as many different integrations as possible, we've sort of built a really great developer experience to build integrations.

[00:04:24] And that, you know, has a, a bunch of interesting trade-offs. Like it gives you a lot more flexibility on the kind of integration and control over integration, so you can really build exactly the integration that your customers need and, and never be constrained by the platform. But at the same time, we aim to give you a lot of the sort of you know, ship integrations fast and, and low maintenance aspects that you get from sort of like more prebuilt solutions.

[00:04:44] And I think the secret sauce is sort of in like how we've chosen the abstractions and how we've made those trade offs to give you that flexibility while still providing a lot of support in the areas where it truly matters and, and you're not having to like, sort of learn a.

[00:04:58] Me and my co-founder started on this [00:05:00] about, you know, two years ago almost we, we basically, you know, an exec, so many, you know, teams. I guess we started with like, sort of an idea. And like started talking to like, so I saw this problem right at my last company and, and we started just talking to other, you know, engineering teams and like, so how do you guys build integrations?

[00:05:15] Right? It's like, you know, it's 2022. It was at that point in time and, and like, you know, sort of unified APIs that kind of like, you know, second you know, coming with like, you know, the rise of merch and, and fi and some of those other solutions. And so we talked to a lot of people and what we saw is this really, like the vast majority was still like building in-house.

[00:05:32] And like I. Was usually aware of sort of the unified API solutions, but it just didn't really feel like it fit, you know, with what they were doing. And on the other hand, you have like these embedded IPIS solutions that are more kind of like low-code, no-code tools similar to sier for kind of embedded integrations.

[00:05:47] But I'd also like, I think to a lot of developers just didn't feel like the right way forward. Like they. End up writing code sort of in a low code, no code tool to like run API requests and something that's like kind of a visual workflow builder [00:06:00] when really what we're they we're building right? Is like sort of a, a key feature of the product and the key infrastructure really.

[00:06:05] Yeah. And so we, we started iterating on like, how could we solve that problem? And honestly think it took us almost a year to sort of find the right approach and find the right angle. And there have been some, you know, interesting detours on along the way, but, you know, I'll save those for conversation.

[00:06:43]

[00:06:43] Robin: Yeah, it's a good question. I. We were following what we were hearing from developers and then we tried different approaches sort of to solve the problem. And so like I mentioned that like it took us, you know, almost a year to find the right entry point and like. It speaks to how complex the [00:07:00] problem is and I think like how, you know, nuance sort of the solution has to be to be truly helpful.

[00:07:04] So I think the part that we had to get right was like sort of find the right level of abstraction. Like we knew from the beginning that like what people truly want is control, right? Like they never wanna be constrained. By what they can build with their integrations. They like need full control over their product experience.

[00:07:19] This is a key feature your customers are asking for. Like if an enterprise customer wants and wants a specific thing and a specific Salesforce integration, like you can't be like, you know, I'm using this framework that makes things really easy for me. But unfortunately that means, you know, I can't like, you know, give you what you want in this integration, right?

[00:07:34] Like this is not really an option. And so I think what we had to really nail is like sort of these abstractions, and I think we started with like, eventually we figured out what we were gonna do, like sort of start really small. And so we ended up focusing on. For about a hundred APIs, it was kinda like, or 30 APIs I think was like the first version of this product.

[00:07:52] And so we, we took sort of everything that we had built and like threw everything out that was not o and just focused on making oof a really smooth and really [00:08:00] simple experience. That I think sort of got us, you know, the first, you know, foothold in, in foot in the door with, with a, with a number of you know, potential users.

[00:08:07] And then from there we kind of went back and we're starting to look at all the other things you have to do, like sort of two-way data, things at WaPo handling and, you know, data caching and, and deduplication and, and all those sorts of things that were sort of like added on back in.

[00:08:25] We could really start untangling the problem for people so that even with a sort of minimalist version, the product could already do things that were meaningfully better than if they were to build in-house,

[00:08:49] right?[00:09:00]

[00:09:08] Oh yeah. Yeah, I mean, you know, there's two of 2.1 coming up, right? At least or has been coming up for two years or something. And I think it's like, but think the hard part with all, with o what you're alluding to, it think is the problem that we see in all parts of APIs, right? It's like there's sort of semi standards and like sort of conventions on how you do things and that can give you the illusion of things being similar when actually they're not always similar and they behave differently in different edge cases.

[00:09:35] And I think that's kinda like what we're trying to. To do it. A platform of what we've done with oof is like sort, if we pre implemented oof 450 APIs, right? And still today, every time we implement a new off version, there's like a bunch of like little quirks that we, that we find that we sort of have to like, you know, work around.

[00:09:50] What we see is like, sort of, there's repeatability in those quirks. And so with the platform that we're building, really we hope to sort of give you the right tools so that it's kinda like, you know, you just need to find the right screwdriver inside of mango [00:10:00] and then like it will fit nicely and the screw will actually turn.

[00:10:03] But if you've gotta like build the screwdriver every time from scratch, it's just very annoying.

[00:10:14] Oh yeah, for sure. So today we're a full like, you know, solution for, for two-Way data syns, right? Like, that's what most of our customers use us for these days. So if you wanted to say like you know, integrate with A CRM, you wanna maybe create a contact there, but then if you want to create a contact, you typically have to know which contacts already exist.

[00:10:30] Robin: So we first like sort of sync out all the contexts. You can have a version of those inside of your own, you know, products database. And that's, you know, really useful because now you, you can do much, you know, quicker search. You can enrich the context sort of from the CRM with other data from your system.

[00:10:44] And then we let you have those full like sort of two-way sync experience. And then also webhook. So we support like, sort of webhooks on, on both sides of the equation. And so you can have like realtime data. There's two way data can also be real time basically with the external a p. And we have like, you know, sort of [00:11:00] support for that for 150 different APIs.

[00:11:02] And I think the, the part that we do differently is, as I said, you know, we sort of not, instead of like, we're not gonna go give you the full prebuilt, like how to create a contact in every CRM that we support. We have some, you know, templates that are prebuilt so that you can get started quickly. But then the goal is really that we have the building blocks.

[00:11:17] So you can have exactly the way to create a contact in A CRM that fits your app. And so you have full flexibility on the business logic and how exactly you integrate with the app. [00:12:00] I imagine.

[00:12:17] Mm-hmm.

[00:12:42] Yeah, I think it's a good question. I think it's you know, definitely B two, like we're exclusively doing this for sort of like, you know, integrations inside of B2B SaaS, right? So all our customers are, are B2B SaaS companies where integrations are a core part of their product typically. So, you know, with the, with the rise of AI use cases, you know, obviously almost any AI product, [00:13:00] right?

[00:13:00] Like needs to interact with other. You know, sort of services around it. And so, you know, that's a, a big sort of use case. But then I think another one is, is really sort of like any B2B SaaS that, you know, lives inside of an ecosystem. So like, almost no product today kind of is its own island, right?

[00:13:15] Like everybody, like your customers basically always have just like, you know. Bundle of other SaaS products that they use around you, and they expect you to sort of seamlessly integrated their workflows. So I think popular categories for us as sort of like, you know, CRMs, accounting systems also ticketing systems are pretty popular.

[00:13:31] Productivity. But then also like sort of, you know documentation, like kind of notion confluence file storage systems like Google Drive. We support 28, you know, different categories of APIs. And we see that that's really needed because. The way you basically build an integration and the way you integrate with those other APIs always depends heavily on your product.

[00:13:51] Like it's really, you know, goes to the core of like the value you provide to your customers. And that is gonna shape like what kind of data you need from the external system, how [00:14:00] the interaction should be with the external system and what the experience should be for the end customer. And so it's typically I think like, often people will actually sort of like, you know, use some part of CRMs, but then also, I don't know, have some, you know, task tracking systems, which at first you wouldn't think, like, sort of go together in the same, you know, product. But like it, it, it gets like pretty diverse.

[00:14:39] Right.

[00:14:44] Yeah, I think they ended up finding today they mostly find us. And I think initially we, we kind of, sort of went to. I think when with the first versions, basically we, we, we posted it in, you know, a couple of slack communities where they're like, sort of were were other startups [00:15:00] and, and then the entrepreneurs we wrote some blog posts about our journey on like, you know, how what we were doing and how we were building, like what we were learning.

[00:15:07] Those got some traction and we brought actually a blog post that was. Got a lot of attention, but like almost, you know, seemed like to, to get the wrong audience in a way. Like, we wrote a blog post about o and like, you know, why Oof is still hard in 2023, and this was published in that year. And it, you know, we came, was very successful I think in, in the sense of like, you know, it really resonated with a lot of people.

[00:15:28] What we realized is like almost everybody who read it apparently used oof, but used it mostly in sort of context like logging into an. So it didn't quite give us, you know, the users that we were hoping we would be getting out of it. But I think it did help sort of, you know, put the, the name on the, on, on, on the map.

[00:15:45] I think it's just honestly early on it's just like being creative and, and like sort of you know, I think the way we're thinking about, it's kinda like if, if even just 10% of what we do like ends up, you know, having some effect That's great. So like we just gotta do a lot and try different things and what [00:16:00] works.

[00:16:01] I think another interesting aspect to this is sort of how we. And NGO is fully open source. I think we never even talked about this so far. But like, you know, all the code is on on GitHub, right? Like you can, you can inspect everything. And that I think also helped sort of with, you know, early interest in adoption.

[00:16:16] But it's a funny backstory to this where we initially built like sort of a data syncing product first and that like, sort of didn't really resonate as much, but we needed off as part of the data syncing product to like basically, you know, be able to offer access token. And when we were looking into that, there was this company called Bearer that had built an oof library called Peasley.

[00:16:37] That was essentially a tool for, for doing oof with a lot of different APIs. And they had built this a couple of years earlier, had sort of like launched it and, and then unfortunately didn't have the resources to maintain it. So by the time we found Peasley was like, you know, a week or so after they had officially like archived the repository into maintenance mode.

[00:16:52] Onth, you know. A dozen or so users, I guess, and, and a couple hundred stars. And so we reached out to them and asked like, Hey, [00:17:00] you know, this specific thing, like is actually really what we've been thinking about building forgo. Like, you know, would you be okay if we like you know, maybe take this over and maintain it and, and they.

[00:17:09] Super grateful and, you know, sort of gave us the, the, the, the repo. And that I think has been a boost for us in terms of visibility, but also I hope that, you know, we could give something back to the community. And to this day, actually, we, we keep the WAF part entirely free, open source and, and, and like, you know, free for everybody to use like in, in part for the spirit of like, you know.

[00:17:29] Making sure that can live on its original mission. And, and a thank you for the gift of sort of free code that we were allowed to take over from.

[00:17:51] Super lucky. Yeah. Yeah, definitely. And I'm really glad also that, you know, the team had the you know, really like, was, was so [00:18:00] welcoming to us taking this over. Like, you know, they, they could have been like, no, this is something that we've built. And, you know, like but I think they really left the spirit of open source and, and made sure that, you know, the project can succeed.

[00:18:09] No matter how that, what, what format takes.

[00:18:25]

[00:18:25] Robin: I mean, I think we integrate with all the public APIs that are out there. So NGO is built to work with any, you know, public API that exists. I think the teams that are building public APIs have a lot of respect for them because I think it's a really hard task. Because you have, you know, such, if I look at like the use cases, right?

[00:18:40] That like people come towards you and are like, look, here's what I want. Build on top of your API. And it like, you know, ranges so widely. I think it's really hard to do. I think it's like the best you can do is Do your best job in a way, and, and, and like, you know, make sure that you know what you publish is accurate.

[00:18:54] I guess. Like it's the usual stuff, right? Like if the docs are up to date, that's amazing. Like, you make everybody's life easier. I [00:19:00] think one thing that we see is really helpful is like if there's, a P key off. And o that's usually seems to be like sort of a, a best off breach situation. Like O is great for like, you know, integrating when with third party APIs.

[00:19:12] API key makes it very easy for your customers to still use the API without going through a lot of hassle. Yeah. And then, beyond that, I think it's, you know, it like how you design the API, I guess like, you know, really comes down to your business and what you think is right for your company. And I think like most teams that we see there do a good job at this because they know their product really well.

[00:19:29] So I think that's you know good that they're doing that.

[00:19:46] Yeah, that's a good question. So we sort of have like two layers you can think of in, in ngo. One is like sort of the access layer to the API. So this is really about like implementing the authentication method, like o or API key. Along with that, we also do things like, you [00:20:00] know, figuring out what other, it's other like rate limits that we should respect.

[00:20:02] So like, you know, we, we pre-configure those in the platform. Then you're like, how should the retry handling work? Sometimes there's ation that's sort of globally applied to the API. All of these things we sort of like pre configuring. Angle. And then the second part is basically like the, what we call sort of the building blocks of your actual integration.

[00:20:18] So like the two-way data sinks in a way. Like which data do you wanna read from those external APIs and which data do you wanna write back? And those we don't pre-built. So we usually build a few examples, but then we don't really pre-built like sort of entire integrations there. Because that's precisely where we give our customers control integration.

[00:20:36] So what gives you basically is like this framework. Right, your actual business logic that interacts with the external API. And for this, actually we're using a lot of LLMs these days. So what we figured out is like, basically because NGO brings you, gives you all the infrastructure and you really just have to decide sort of, kind of like, which data do I want to fetch from the external API?

[00:20:58] What does that look like? And, you know, how do I [00:21:00] transform it into my worldview?

[00:21:04] That's something that like LLMs know really well once they know the documentation of the external API. So we either pointed at that. And then we have a, you know, A GPT for instance, that we, we fed with, with some examples from Mango Scripts. And we usually get like 80, 90% correct answers. It figures out the endpoints it should be calling.

[00:21:20] It usually figures out the parameters. It figures out a lot of the data mapping. There's still some editing required, but it makes us at least two to three times more productive when actually building integrations. And we're increasingly, you know, giving the same, you know, option obviously to our clients.

[00:21:34] And I think they're finding similar results that like they can have something that's custom basically at the speed almost of prebuilt, because thanks to all s it does all the heavy lifting for them of figuring out what to call and how.[00:22:00]

[00:22:05] Right.

[00:22:44] Yeah, I think like, you know, copilot came up pretty much with us starting mango or, or a little bit before. I don't remember the exact timeline, but I think it's like sort of, you know, I think we were using. Co-pilot and, and LMS to some degree to build, you know, NGO from, from day one. Right. And so I think it [00:23:00] naturally extended to us that, you know, when with the rise of Jet GPT, basically we would start to ask Jet G PT to like, you know, it's great at other coding tasks.

[00:23:06] Like why shouldn't it be great at like helping us write, you know, integrations. And I think it's like, you know, you mentioned like you, we don't, we don't see LMS as kinda like a black box solution, at least not at this stage for building integrations. And I think, you know, our, you know, one of the philosophical principles in a way of mango is that like we put you, the customer and the engineer who builds the integration into full control, right?

[00:23:29] Like you always have the last word on what goes out and you can always build whatever you want with the platform. We're is there to make your life easier. And I think, you know, that LMS in, in deploying them on Mango is a natural extension of that. So. Yeah, I think it was always clear to us that like it would be more of a productivity booster and then eventually if it can do like 99% of the job for you and you're kind of just like, oh yeah, that looks great, you know, awesome. But I think the other thing is also with LLMs and, you know, with prompts in general, right? It's like it's natural language and like we, even as humans, like when we're talking to each other on what we're [00:24:00] gonna build, have like, you know, misunderstandings of what we were communicating with like sort of natural language and describing what we wanna have built. And so I, you know, not sure that like sort of. Ever be able to kind of like read our minds? Because I think sometimes when we write out what we want, we don't even know yet entirely ourselves. Yeah. So I think it will be interesting to see how that, you know develops. But I definitely have been a big productivity.

[00:24:22] boost for everybody in our customer base that has been, you know, using them to, to build on top of

[00:24:27] ngo.

[00:24:27] Mike Bifulco-1: like you had really good timing for starting to build the company as well, and in as much as copilot, you know, sort of started to coexist with you at the same time there too. Can you tell me a little bit about how pricing works for Mango right now?

[00:24:39] Robin: Sure. We have a free tier, you know, to get started for like smaller you know, early stage startups. And then we have like sort of startup friendly, you know, sort of early category, like early pricing. So that, you know, you can really focus on getting to product market fit and like sort of, you know leverage language for that.

[00:24:53] And, and we see that people also there each trade a lot. And then, you know, we have basically a scale plan that's really meant for you to scale. [00:25:00] Thing that is very important for us is that like, you know, we try to align our, our pricing with like sort of the value integrations that driving for your business and, and sort of the usage that you have on the platform. So we're not big fans of like sort of you know. Prohibiting. I mean, we definitely wanna be prohibitively expensive, right? And like, sort of, we believe, basically we wanna power all your integrations, right? We understand that means that like we need to be on the same boat of like, if you are really successful building a successful business, having a lot of customers use those integrations, like, you know, we should get like a fair share of that success and, and, you know, we should be able to, to cover our costs.

[00:25:34] But we also believe that, you know, you should be able to have a very profitable business. I think that's, you know what, yeah, with integrations basically, I think that's not always easy because like they are so custom to the product and so I think a lot of our pricing packages end up reflecting that and end up sort of, you know, being really built in collaboration with the customer to figure out like what you know, makes sense for their business and for their specific use case of

[00:25:57] integrations.

[00:25:59] Mike Bifulco-1: to have some flexibility [00:26:00] there too, and hopefully. When everyone wins, happier. And, uh, you're all, you know, building together in the direction towards, uh, a future that makes not only their tools better in their company and product, but also your product can continue to support, you know, whatever's pipe.

[00:26:12] Um, Especially as new versions of the open API spec and I'll

[00:26:16] things, uh, start to be

[00:26:18] Robin: No, I think we, we definitely see it as a partnership, right? Like with, with our customers. Like it's, you know, integrations are never done. They always keep evolving. They always keep changing. And, you know, as you said, like, you know, the environment keeps changing because you've integrated with third party APIs.

[00:26:30] Those keep changing. And so it's like you know, we definitely see it as a, as a long term partnership, more than like, you know a quick, like build it once and then like you're done

[00:26:38] forever.

[00:26:40] Mike Bifulco-1: the right strategy to have. , Uh, everyone, what's the phrase? Um, uh, A rising tide raises all ships or whatever. Uh, The, the old trope is

[00:26:48] Robin: Yeah.

[00:26:48] Mike Bifulco-1: Uh, So we've talked a little bit about what, what Mango does and, um, how teams are adopting it, why teams are adopting it. I'm curious if you can maybe describe, what sort of like the typical first [00:27:00] experience with Mango is.

[00:27:01] So what's HelloWorld? Uh, When someone first jumps in to give the tools a try.

[00:27:05] Robin: Yeah, good, good question. So actually when you sign up for the, for the, for the cloud version of the product, like we immediately have getting started. That lets you connect to GitHub and sink in issues from GitHub. So you can see how that works. And then you're actually gonna be able to create an issue on a public repo as well.

[00:27:21] And so like we have like sort of, you know, test that, that we use for this. But you see basically like what. The end user experience flow is going to be, so if a customer goes and authenticate, you know, an integration of your product, what, what will they see? Like that experience is fully wide label.

[00:27:34] Like you can customize it entirely. Like we're literally just the JavaScript SDK that you call. It's nice, you know, to, to see that for people. And then you get to experience the Nway APIs as well. So you see like how the two day data syn work, you see how you know. How you're going to interact from your app with the platform. And like sort of, you know, the flexibility also of this. And I think we see that that's like a, a great way for people to imagine, okay, like, is this gonna like work for my use case? [00:28:00] And start to like sort of architect in their mind on like, how will I integrate my integrations here And like, how could that work for my product?

[00:28:07] Mike Bifulco-1: Yeah, I actually don't think we talked about this early on, but Mango, um, supports, uh, is it TypeScript at the moment? Are there other languages that you're supporting or planning to support?

[00:28:16] Robin: Yeah, so like the language itself is, is sort of language agnostic.

[00:28:19] So like, you know, any, any backend, you know, language that you use like is fine. Like we have a rest, API for you to interact with. We have SDKs for various different languages. And then there's a. Parts of the, the integrations yourself that you write, they're, they're written in code, they live inside of your gire, they're version controlled. You deploy them to NGO with a CLI and you can think of those as like, you know, very small sort of TypeScript Lambdas that just do some interaction with the external API. And they run in the context of sort of the NGO frameworks. So those are written in TypeScript. But as I mentioned, you know, there're usually sort of 20 to maybe 50 lines of code.

[00:28:50] Pretty simple types code. And, and that needs to be just because it runs on the platform itself, but is.

[00:28:56] Mike Bifulco-1: That makes a lot of sense. Yeah. A, a very interesting developer story. I [00:29:00] like that it's, um, uh, source controlled as well. Uh, The, the having that sort of history of how the integration plugs itself in is really helpful, I'd imagine when it comes to debugging and also environments and things like that as well.

[00:29:12] Robin: Yeah, exactly. So I think that's a big reason for us, right, is like, sort of a lot of our customers basically will deploy their NGO integrations as part of their ci slash cd process. So because it's just another like folder in your repo, basically it follows, you know, your regular sort of coding, deployment flow, whatever, you know, that might be. The only thing you gotta make sure is like when you deploy the production, you, you have the latest version on Mango, and that's as simple as like running a single CLI

[00:29:35] command.

[00:29:36] Mike Bifulco-1: Okay, a couple more questions for you. Uh, and then, then we will, uh, head on our ways here. what are you thinking about next? What are like the problems you're interested in solving or the, the big features you're looking forward to, uh, building next with Mango?

[00:29:46] Robin: Yeah, good question. We're building out more on the, on the webhook side. Like we have support for webhooks for a lot of the APIs already. But, you know, definitely building out more there. I'm very excited about our initiatives to leverage LMS more, to build integrations more quickly on the platform. [00:30:00] That I think is something that I'm, I'm, I'm very excited about and honestly, but I still see that we still have a lot of potential. And then the other parts is sort of like, I think you know, one thing that we. We already let you customize integrations for individual customers. So what this means is basically like if you, if you have a customer that has Salesforce, right?

[00:30:16] Like, I mean, almost every customer customizes their Salesforce. And so we see that like, sort of those differences in Salesforce can make it tricky to build integrations against those kind of you know, more complex systems. We have some helpers already there for you today, but like we have a couple of good ideas on like how we can make that a lot less painful in the future. And that's something I'm really excited about because I think the world of SaaS is in a way moving more and more towards customization and more complex systems as like everybody has all of their data and software. They want their software to be more flexible, to really fit their needs. And so I think we're gonna see much more customizability of different, you know, systems and APIs in the future. And so I think supporting our customers in supporting their customers you know, with this, I think is a, is an important part [00:31:00] of our mission.

[00:31:01] Mike Bifulco-1: Yeah. And again, that's another great place where you can grow together and, and sort of hopefully support features that help your, your customers do better cool. Robin, what about, uh, your now? Are you currently hiring,

[00:31:11] Robin: Yeah, we're always growing. We have just had a, you know, good, growth spurt, but we are always looking for, you know, talented people. Especially on the backend engineering side. If you are, you know, excited about APIs maybe have experience building integrations, would love to build an open source dev tool for other developers.

[00:31:25] We'd, you know, love to hear from you.

[00:31:27] Mike Bifulco-1: what's the URL? Where can people go to find mango?

[00:31:30] Robin: Sure that's www ngo. Or just Google Mango. That's like mango?

[00:31:35] with an N and you're

[00:31:36] gonna find us

[00:31:38] Mike Bifulco-1: I'll make sure I have some links in the show notes as well, both to your website and your, uh, repos on GitHub.

[00:31:43] and one last question for you, Robin. Where's the best place to find you online? If people wanna shout about integrations and, uh, APIs for, for integrating and things like that, where are you most active?

[00:31:52] I.

[00:31:53] Robin: You can always find me on LinkedIn. If you look for Robin Mango, you'll find me there. You know, when you sent me [00:32:00] a connection request, maybe, you know, call out the show here. That I.

[00:32:03] know you know where you're coming from and I'll be happy to accept and you know, we can message there. Otherwise, I'm also active on the NGO Slack community.

[00:32:10] If you are looking into ngo, you're probably gonna end up there and I'm always

[00:32:13] active there.

[00:32:15] Mike Bifulco-1: Well, Robin, thanks so much for joining today. It's been, uh, really interesting to talk to you and, and to hear about Mango and how you got to where you're at. Um, Feel free to come back anytime. Uh, If you've, you've got launches in the works and things like that, we'd love to hear from you again.

[00:32:26] Thanks so much for joining. It's been a pleasure.

[00:32:28] Robin: Thanks a lot for having me,

[00:32:29] Mike.

[00:32:30] Mike Bifulco-1: Take care.