Ep. 118 The MongoDB World Series - Nick Gamble From Unqork

Shane McAllister: Welcome once again to the MongoDB podcast. We are as always glad to have you here. I'm Shane McAllister. I'm a developer advocate here at MongoDB. And for those that have started recently listening, once again, I am also joined by Cedric, who is our intern here at MongoDB. Cedric, how are you?

Cedric Clyburn: Doing good, doing good. This week, I'm working on some stuff. Two upcoming articles and perhaps video on the DevHub as we talked about before.

Shane McAllister: Excellent.

Cedric Clyburn: Yep. Yep.

Shane McAllister: Okay. And what are they on? What are you writing about? What are these articles we look forward to seeing?

Cedric Clyburn: One of them is going to be on the Kubernetes Atlas operator, which I guess just became GA during MongoDB World. And the other on MongoDB and the Tableau connector. So as always, they're keeping me busy here for sure.

Shane McAllister: We've got to keep the interns busy. You're one of 120 plus this summer, and for us, it's always amazing to bring young interns in from college to see their take on what we do and to share our knowledge. But most importantly, obviously, to get your take on what we're building and also projects. We've seen some superb intern projects in the past, some of them which have gone on to live beyond just intern projects and have been subsumed into other parts of the MongoDB ecosystem, so it's super valuable. I know it's a short time that we have you for, but we hope that might turn into something more because most of our interns will also, all things permitting and all things being good and equal, have an opportunity to return to MongoDB once their studies are finished, so it's a really, really good path to a potential role and career here at MongoDB. Always good to have you, Cedric. Good to have you.

Cedric Clyburn: Thank you.

Shane McAllister: I know at world, and this is episode three of the world series of podcasts, interviews that we did at MongoDB World back in early June. And I know before we get into the main clip that we want to have for you today, that we sent you out with your microphone along the show floor, or the partners pavilion, or promenade I think that we called it at the time, Cedric. We played some in the last episode, but we've got some more, right?

Cedric Clyburn: Yeah. We've got two more here to show. One is a MongoDB employee who got hired during the COVID period, so she's sharing her experience at her first ever world.

Shane McAllister: Inaudible.

Cedric Clyburn: And it was the first world for a lot of people, as it turns out, as I was interviewing.

Shane McAllister: It was. It was myself included. I'm here two and a half years, but obviously COVID stopped both the previous worlds, so it was superb to be back in 2022 and in person, and actually put a face to a lot of colleagues that we only see on Zoom and meet them in person, and also to share some time with our customers and our partners as well.

Cedric Clyburn: Right, right. Yeah. I mean, you nailed that on the head right there. I mean, during the day before the conference at the office, people were coming in, and I guess a lot of people in these teams had just never met each other. So there was a lot of handshakes, there were a lot of hugs. It was, honestly, a good experience to see that and to see the family environment that we have here at Mongo. But the other was also a panelist at the conference.

Shane McAllister: Okay.

Cedric Clyburn: She was sharing her relationship between the company she works for, Apollo, and Mongo DB, and also how excited she was to be on this panel at the conference. Two really cool clips. Let's go ahead and take a listen. This is Cedric Clyburn. I'm here with-

Isabella: Isabella inaudible.

Cedric Clyburn: And today we're talking about what you're most excited about for MongoDB World.

Isabella: Well, this is my first time to attend a MongoDB World. I got hired during COVID. I'm excited seeing my customers. Today, I got to attend a session with Coinbase, which was really exciting. I've met with a lot of other MongoDB employees that I've been seeing virtually, but I saw them in person, so it's really great.

Cedric Clyburn: Well, absolutely. Hi, I'm Cedric Clyburn and I'm with-

Peggy Rayzis: Peggy Rayzis, I'm the senior director developer experience at Apollo GraphQL.

Cedric Clyburn: And we're talking today about what are you most excited for for MongoDB World?

Peggy Rayzis: I'm really excited about the partnership Apollo and Mongo have together. What's really great is it allows some of our customers like American Airlines, for example, the ability to use their legacy relational databases alongside some of their new, exciting tech like MongoDB Atlas. I'm also really excited for my panel tomorrow. It's at 3: 15. It's with some other leaders from Vercel and Prisma, and together we're going to talk about how our companies are revolutionizing the developer experience along with MongoDB. So check it out.

Cedric Clyburn: Well, thank you so much.

Shane McAllister: Wow. That was great. It was brilliant to capture that flavor of MongoDB World and obviously that's an annual event for us. But we do have local events in different countries around the world, so keep an eye on mongodb.com for what we call a dot local event coming to somewhere near you in the future as well, too. Without further ado, we'll move on to this interview. Again, another interview Mike Lynn did on the show floor at MongoDB World. And this one is a super hot topic. And Cedric, this will be interesting for you too, because obviously what you're studying in college applies to where we are heading and essentially this movement, this low- code, no- code movement. It is sometimes controversial in the developer's sphere. Developers like to have full control. They like to start things from scratch and first principles, et cetera, but not everybody likes to do that and it can be slow. We would certainly say that leveraging a framework or leveraging a product is always a quicker way to go about doing things. We have a really interesting company called Unqork and we have Nick Gamble, the head of evangelism at Unqork, talking to us about what they do and how their company helps users to rapidly develop and deploy their apps and platforms with minimal coding. Let's hear from Nick when he chatted with Mike Lynn at MongoDB World and learn how Unqork allows enterprises to seamlessly integrate all of their systems and data and increase productivity and innovation through automation.

Nick Gamble: Okay. Nick Gamble. I head evangelism at Unqork, which is a role I recently started after spending over three years at the company, being a solutions architect, developing with the platform, leading teams that actually even developed features of the Unqork platform with the Unqork platform, which was pretty exciting. And now just sharing the love that I've grown for Unqork with the rest of the world.

Mike Lynn: Yeah. Fantastic. Well, it's great to have you on the podcast.

Nick Gamble: Thanks.

Mike Lynn: I'm really excited to talk to you about Unqork. It's a great platform. Explain to folks what it is. Apart from a low- code, no- code development environment, tell folks what the benefits of using Unqork are.

Nick Gamble: Sure. The benefits of Unqork, we're really framing in this umbrella of codeless architecture now, which is how we're rethinking no- code, which is effectively an approach to no- code, but an open approach. So Unqork uses Mongo under the covers to store many things, both the data that our customers are collecting, as well as the representation of your business logic. So codeless architecture is effectively a JSON based declarative document that describes your application's data, your application's business logic, and your application's presentation, effectively separating the logic from the implementation. There's a lot of other no- code, low- code platforms will tie those two things together. Whatever you do in their interface is going to either generate code or generate some proprietary format. Unqork is taking an open format to it, leveraging the power of Mongo underneath that in, order to rapidly develop and confidently deploy your software.

Mike Lynn: Great description.

Nick Gamble: Cool.

Mike Lynn: It's clear that you've done this in a presales context, as well as an evangelism context.

Nick Gamble: Yeah, it's funny you say that, because I actually did start at Unqork on the presale side. I had spent my career in software development before Unqork. I was in product development teams, leading engineering teams, and it wasn't something I really wanted to do anymore. And I wanted to take my passion for technology and open it up and start to deal with a more people focused approach to technology.

Mike Lynn: Okay. I'm a developer, from way back. And I had exposure to a low- code environment early on, looking at an application, looking at how you were going to manipulate and manage data, and then leveraging some lens of low- code or no- code to accomplish that. I vividly remember some frustrations around limitations. Do you recognize that a no- code environment can be difficult for someone who has written code and is that something that's been thought about in development of Unqork?

Nick Gamble: Sure. Certainly we're trying to keep things as pure no- code as possible. And again, going back to that codeless definition of that declarative representation of your logic. Admittedly, we've gotten very far in doing so. We're dealing with very large enterprises, large banks, insurance providers, even governments like New York City and developing very complex applications using purely no- code. As well as I even, for fun, built a video game completely with no- code. So I think we've shown that you can do a lot, but yes, there are certain times where there might be various UI interactions that may not be supported directly or there might be some other logic that it might be better done in code. And that's that's okay, I think, in the end. And it's about, right now as companies are digitally transforming, they have to do so in a very composable manner. You can't just expect that you can move everything over in one big shot nor that one tool is going to be right for any job. Even when you're thinking it from a developer perspective, different programming languages have different benefits and you're going to use them in different cases. Different databases have different benefits and use it in different cases. Like Schemaless is great, MongoDB's great, but there are probably times where you might want to use a different data store. So you can look at that at the same way with our platform where we have a very strong integration framework, so if you do need to go code something somewhere, Unqork can tie very seamlessly into that in a very visual manner to help support some of those cases where maybe we don't have all the perfect capabilities that are going to achieve your need.

Mike Lynn: Yeah. Now, does Unqork give you the ability to have a break in the no- code experience and customize with code?

Nick Gamble: Technically there's a secret back door, but it's very much frowned upon. And we do steer clear of it. But we are thinking about other ways in terms of how we can allow people to introduce code if that's something we want. So as we open up this codeless definition that I was talking about a little bit earlier, we're going to give people the opportunity to bring in some of their IP and represent it in a codeless manner. So as long as we keep some separation between that creation experience and then how their data is stored against it, we can potentially allow some code to be brought in, but then to the end user, or the developer's perspective inside of the Unqork platform or any other platform that may subscribe to the codeless standard, it feels very visual on this entirely no- code to them.

Mike Lynn: Yeah. Now, is there a set of applications or use cases that are just really in the sweet spot of Unqork?

Nick Gamble: Sure. Again, we're working with a lot of large banks and enterprises and financial service... financial services, insurance are our two primary markets. We're also getting very big into healthcare. So it's a lot of onboarding flows. Anything that's a multi- user workflow oftentimes with integration with other systems, that tends to be our sweet spot. But again, we've proved out some pretty interesting stuff. Those highly regulated workflows, it tends to be where we hit it hard.

Mike Lynn: Okay. So forms and processes and multipage?

Nick Gamble: Right. So multipage forms, which might have multiple different users interacting with it potentially at the same time or in sequence, having calls down to the server, which might talk out to other APIs that are going to enrich the data or do checks against the data that you're capturing, those are going to be a very big sweet spot. And then also on top of that, once you have that data, being able to visualize the data in dashboards and running reporting against it in metrics and things like that, and understanding how your workflow is going and where you can potentially optimize it.

Mike Lynn: Yeah. I'm going to dig in myself. I have a couple of things that I'd like to experiment with. How do you get more information about Unqork and where do I begin?

Nick Gamble: Sure. Well, we have Unqork.com. Unqork. com/ creators would be a great place to go. That's our creator hub, where you're going to get a lot of awesome information about, want to sign up to learn even more information through our newsletter, see some spotlights from some of our power users and the things that they've created. And then we'll also be able... We're soon going to be opening up our community, which is community. Unqork. com, which you'll be able to get tons of information, introductions to our documentation, to a stack overflow question and answer forum. We also have a Twitch channel, twitch. tv/ Unqork. We've been going on a weekly basis on Thursdays, but we're going to be ramping that up a lot more. You can see a lot more from us, and everything that we do on Twitch also gets posted to our YouTube channel, so you can find some archives of some of the content we've had up there.

Mike Lynn: I'm curious about, when I create an application with Unqork, I'm creating a JSON document, which is declarative, as you mentioned, is it possible to share those? And is there an open source concept in Unqork?

Nick Gamble: Yeah. Effectively everything is visible, everything is open. Everything is API enabled, so when you design a module, let's say that's... A module is effectively a page within a workflow or an API endpoint. And that's the primary building block of these applications. Everything that you do they there, the JSON that comes out of dragging and dropping the various elements onto a canvas, is fully accessible and visible in your API. We've had examples where we even push that outside of Unqork. So there's a few clients who want to bake Unqork within their own CICD systems. So let's say they're using GitLab for example, and they're controlling all of their CICD all their deployments from there. We're actually able to push our JSON documents into their GitLab, where they can do merges and polls and whatever else they want, and then call back to Unqork for promotion. We understand that you're going to want to get at your data. You're going to want to see it in other systems. And we're getting into these organizations that have processes in place and other systems in place, and to be a part of that, we need to be extremely open with what we do so that we can integrate as best as possible.

Mike Lynn: Yeah. As you're talking about that, I'm thinking about native versus compiled, or emulated versions of the code. What's happening in Unqork's mobile space?

Nick Gamble: Right now we're mostly focused on mobile web. There's two primary, what we call, run times to Unqork. Everything that you build in Unqork is going to be rendered in a browser, or it's going to be executed on the server, but we're exploring a number of other run times. And that's one of the ideas of opening up this codeless definition that we have, which represent us, is now when people see the open standard, they understand what everything in that declarative document means, we can start to open up the world of people who can build new run times, mobile potentially being one of them. Right now we have seen simply dropping an Unqork application into a mobile wrapper where you can still access some of the native features, the biometrics and other things that come with mobile, but certainly once we open this up, we expect to see a lot more traction building dedicated mobile run times.

Mike Lynn: I want to dig into that. What does it mean to open that definition up? Is this a standards approach that you're looking at?

Nick Gamble: Yeah, exactly. An open standard, so effectively again, it's JSON, so it's built on top of an already existing open standard that you have benefited from tremendously and we are as well. So then, basically for each and every component, each and every element of Unqork, there are key value pairs that describe its behavior and describe its business rules, the validations, the actions to take if someone interacts with a given component, so those key values will be defined and you'll understand that if it says required true, then you need to validate this field at any point in time. So you'll be able to see the documents, you'll be able to see what they mean, what each declaration means and how you should respect that if you go build a run time on top of it.

Mike Lynn: Sort of as schema, for applications, effectively.

Nick Gamble: Yeah. Yep.

Mike Lynn: Well, is there anything else you'd like to share with the audience when it comes to Unqork or maybe even yourself?

Nick Gamble: Yeah. I mean, look, I think with low- code and no- code, a lot of people tend to focus on citizen development. That's like, okay, well, we're going to do no- code, so then everyone in an organization can start building software. But we don't come at it that way. Not to say there isn't a wonderful aspect of low- code, no- code that democratizes software development for a lot of people, but software development is software development in the end. And there's a lot of very difficult things. I'm sure a lot of your listeners being, very hardcore developers, they don't necessarily want a citizen development platform, or like to be able to build complex things. So we're definitely looking at this being a very developer focused no- code platform, in that you still have a lot of the power that you get when you code you, but you are abstracting away a lot of those mundane things that you're always doing as a developer, whether it's you're a DevOps engineer and building out your servers all the way to a front end or backend developer, writing code. We're understanding what you've been through and taking the concepts that you're used to, and giving you a new visual way to go about it in a declarative way. So don't think of Unqork as just a citizen development platform. It really is a general purpose, powerful development platform, that myself, again, as a career developer have just been drawn to and have seen what you can accomplish when you put someone with that kind of experience in front of a keyboard using a platform like that.

Mike Lynn: I hate to go backwards, but as you were answering that question, I'm again, thinking about the output of my work in Unqork. I'm dragging, I'm dropping, I'm creating a form, it's multipage perhaps, I'm applying business rules, we're going to be capturing data. When I'm finished, when I'm ready to publish the app, what is the output, aside from the JSON document? Is it compiled? Can it run natively on a server or does it require an emulator environment?

Nick Gamble: It is just the JSON document. That is the only output. There are multiple levels of environment, so you're going to have a development environment, a QA environment, and a production environment. So ultimately when you're promoting your applications, you're just pushing that JSON document from one environment to another. And then our browser runtime, effectively, just interprets that on the fly. Just like an interpreted language, so it's just run right on the fly.

Mike Lynn: Does it have to run in the Unqork environment to be available to the general public? Or can I run it on my own servers?

Nick Gamble: Right now, the express application, which is our browser runtime, that is kind of blended a little bit or attached to an Unqork environment. However, we are exploring concepts of web components, so actually taking Unqork and not needing an entire page in order to render it, but actually be able to render just a piece of a page in a web component which you can plug into maybe your React framework that you have, that you're building on your own. And potentially SDK and things like that.

Mike Lynn: Terrific. I'm curious about the number of users of Unqork, and the scale that you're you're working with at Unqork.

Nick Gamble: Sure. I'm trying to remember the exact number. I believe we have in the tens of thousands of trained certified Unqork engineers, and that's across folks who work for the company, folks who've been trained in our customers and then a large amount of SI partners we have that do a lot of the development for us. So yeah, the numbers are pretty large and they're from all different backgrounds. We have a lot of folks who are, like myself, engineers. We have folks who didn't necessarily have an engineering background who have been able to be effective. And then you have the folks that are kind of in the middle where you have maybe your QA engineers, folks who don't do that on a day to day basis, but have the really technical mindset that now we've given them a platform that they can really be effective in.

Mike Lynn: Is Unqork in Atlas?

Nick Gamble: Yes.

Mike Lynn: Okay. So you're leveraging the Atlas platform. What other platform features are you leveraging? Are you leveraging auto scale for example, or search capabilities or charts?

Nick Gamble: Yeah. We are using a lot of the query optimizers and the index suggestions. Because we have customers building such a wide array of applications on top of our platform, that every one is almost unique. And we need those insights to be able to pull things out. We leverage the BI connector to get data from Mongo into other client's analytics tools, so they can do additional analysis on top of what's possible inside of Unqork. And then auto scaling, for sure. And again, the two things I just mentioned earlier, we're very excited to try to bring in as well.

Mike Lynn: Great. Well, Nick, we're here recording at MongoDB World 2022. And I just have to ask you, have you heard any exciting news or announcements from the show so far?

Nick Gamble: Sure. There are two big ones that I think is really getting Unqork excited. One is cluster to cluster syncing. That was a very big one. We're very excited about that, as you're building all of these highly available systems and also working with data that exists in production and lower environments and having to get those real life examples down in your test environment, so that was a very big one for us as well. And then also the relational to Schemaless. I think there was an automatic conversion that y'all talked about. We're working with a lot of enterprise that have a lot of legacy products and the more capabilities that we can leverage from Mongo to help to transform those is going to be better, so that sounds like a very exciting one that we hope to be able to leverage.

Mike Lynn: Terrific. Well, Nick, thanks so much for stopping by and sharing details about Unqork. Anything else you want to mention before we wrap?

Nick Gamble: Michael, thanks a lot for having me. This was a lot of fun and it's been really exciting to be here with you all at MongoDB World.

Shane McAllister: Well, that was a fascinating conversation there with Mike and Nick. I really enjoyed it. As a developer, I'm yet to say you have to fully embrace or explore low- code no- code tools, but it certainly looks promising. And, from me, and really what Unqork are promising is that, they're developer focused and they're following an open standard and, as you know, we're big fans of open source here at MongoDB, obviously, but for me the biggest thing is extracting the mundane and repetitive tasks for a developer. And I for one am certainly all for that, so I will certainly watch the progress of Unqork with keen eyes. That's been another superb episode. That was episode three and we will be coming back with more. This is a nine part series in the interviews that we did at MongoDB World in June, 2022. I think that's about it. Right Cedric?

Cedric Clyburn: I believe so. So remember, as always links are in the show notes, and please remember to review and subscribe and stay tuned for our next episode for MongoDB World. For myself, Cedric Clyburn-

Shane McAllister: And me Shane McAllister, it's been a pleasure and we hope to have you join us on a future episode very soon. Thank you all.