 Hello, and welcome back to theCUBE's live coverage here. I'm John Furrier, your host in New York City from MongoDB Local. This is their flagship inauguration event of a 20 plus city tour. We're going to go outside and take it to the streets where the developers are. Bottoms up, organic, that's the way the market is. We love that. Got two great guests here, going to unpack what's going on at Mongo, how to stand up applications, how to run them, how to build them, AI apps, and in the modern developer data platform. It's killer, Andy Lieberman, CIO at MongoDB. So we're going to ask you all the hard questions. It's Aaron Hernandez, VP of Developer Productivity at MongoDB. Thanks for coming on theCUBE. Thanks for having us. Thank you. So first of all, love your title, CIO, which means you're in charge of a lot of stuff. Developer productivity, the hottest topic on the planet. It's, you know, yet another in a long line of buzzwords around, what do you do to ship software? I love, it's all about software. This is what I love about this event. It's not about this, that analytics, some dashboard, it's about good old fashioned pure software and data coming together. And I think Mongo has really hit a home run here. I was just talking to Dave, the CEO, and I heard the keynote, I wrote a story about it. I think the developer-led data platform is a genius strategy, it's needed, and it's not new at Mongo. This is like been going on for many years. There's a trajectory, there's economies of scale involved. You have a customer base, that is developers. So it's not new, this is a big thing. Now everyone's like, whoa, platform's a hot pot. Congratulations. Well, what's your reaction to that? You agree, obviously. Well, I mean, I think it's safe to say that, you know, Mindy and I kind of represent the practical, like, aspects of that whole thing, right? You know, Mindy owns technical operations and all the corp side stuff, which very tightly couples with, you know, R&D developer internal tooling and development and when, you know, when the challenge came on with what about this AI thing? You know, Mindy and I sprang into action and we're having all these conversations, you know, on both sides of that internal corp oriented versus internal developer usage. Both of those things on the way towards leading to what's the external, you know, engagement that we're allowed to enable through that work. You know, I was joking around, sometimes it's better to be lucky than good or luck is where preparation meets opportunity, right? I think that that's the ladder, is where you guys have this platform, has been actually tailor made for what happened so it's almost, you know, skate through where the puck is going to be. You guys hit it, home run, you're right in line with the trajectory. 100%, I mean, I think it's prescient because when we think about developing software and you're right, it's all about data, it's about applications on top of data. I mean, I look at it for IT as a portfolio. There's some that is built, there is some that is bought. When you can't go to the market and find things that are fit for purpose, you have to build. And to have a developer platform available and I get good pricing, that, it can't be beat. Well, I want to just make a disclaimer and everyone knows I'm biased. I love data in Covenant since 2010 when theCUBE started with big data at Duke. We saw that the maturation of what was really happening was less about the people selling software mechanisms that no data experienced, the people who had data. But DevOps was the growth. Then DevSecOps, so the question is, will there be a DevSec data ops? So now you have a whole nother level of, because security is a data problem, okay? Dev's a data problem now with a Dev platform and ops is a data opportunity problem. So, or is data just in everything? I mean, I think it's- I'm kidding about the data at SecOps, but my point is data, okay, data is ubiquitous and you have to always think about careful ways to manage it because there is, I mean, data is ubiquitous across your entire enterprise. So in fact, one of the most important things that I do as a CIO, I mean, they call it a chief information officer because it's all about the information part, right? You have to figure out ways to gather it, to disseminate it, to cleanse it. So it's all about the data. Right, and for my developers, when we want to do testing or analysis or other things to help with the customer engagement, we have to be mindful of how are we managing the customer data and how do we build the guardrails to make sure the developers aren't looking at stuff they're not supposed to. And vice versa, when we're talking, to make sure that our data is also saved in a safe way without getting in the way and that's what DevSecOps is, that idea of shift left is interesting not because it's making the developers more acutely aware of security, that I mean, that's a little part of it, but it's that we ideally have created a platform where they don't have to worry about it, but they get the benefit of the protections that were built in to how they do their development. I'm glad you brought that up because this is exactly what I've been saying and seeing and this is what I hear here at the conference is that the developer want, they want easy. They want to have the guardrails. We saw that in security. That was, that's the security guardrails. SecurityOps is doing great right now. There's love between Dev and Sec right now. Data is an unknown. There's no real hatred for data because everyone loves data, but it's new. It's a new way to program in the app. So the ops side of it gets interesting. I mean, I just see something where I'm a developer. I would be like, compliance, no. I got to do all, no, I don't want to do it. I just want to write my code. I want to build my app. That's the value of a platform. When you can have one solution that everybody needs, those are killer, killer features. And Sec is certainly one of those. And you know, and it comes back honestly, you think about DevOps, you know, if you ever talked to Jess Humble, he says DevOps is a community of practice, right? It's not a role. It's not a job description, right? It's an approach. It's a culture. And so, you know, one of the things around Mongo, like Mongo is a developer tools company at the end of the day, right? So our customers are developers. We use it internally, dog food, making sure that we take advantage of that opportunity. Cultural aspect also goes into how we think about things like security and compliance. You know, I'm on the R&D side, Mindy's over on TechCops. We talk very closely. We talk a lot. The security application folks when the product development teams were all part of these discussions because we have to be able to have a common language, common understanding so that we can move not necessarily in lockstep, but collaboratively so that we can understand where are we hitting these edge cases that we need to, like either Mindy needs to boost them up or maybe my team needs to boost them up so that those guardrails are not just created but curated in a way that has the developer mindset thinking that security isn't just people telling us no. Right, that way lies death. And it's, you know, in this day and age when things go so quickly, you can't go slow. Well that's, well the culture is a great point because everything revolves around culture. The team formation, the discipline, the practices, the empathy and the compassion of your teammates but more importantly speed. Remember agility and speed has been the hallmark of DevOps. And so now with AI that's only going to get faster but that could get worse too. You could go off the rails. 100%, it's a Goldilocks problem. You move too slow and you haven't capitalized on the opportunity. You move too fast and you become a headline so you have to find the just right. And that's where there's a really good partnership across MongoDB to tear and I work very closely together. Well I'm really glad you guys came out because I've been riffing hard on this all week and then I've been ranting on my podcast around the role of ops and operationalizing the infrastructure. Amazon's got bedrock, Microsoft's got open AI and all this other stuff. So a lot of people saying look at run your AI on here then you've got the developers saying I just want to code. And they're still in discovery mode. They're trying to figure out what I'm going to work on who I want to hang out with or which culture I want to align with and open source has proven that the developers are driving the culture. The bus, they're driving that bus. So the question is, what is the good enough operations to enable and not foreclose developers from being curious, playful and experimenting on the code? But you want to take this one? I'm going to take this one because how part of our strategy aside from providing developer tooling for us to experiment and find out what works versus what's hyped part of that is collaboration and a very light touch at coordination. So basically we are saying yes to safe cases where we are trying to shorten the process to make sure that people can get their hands on things early and often and just we test and learn and then expand. And one of the beautiful things about MongoDB that I love, it's one of the reasons I came that MongoDB is based on open source technology at its heart. You can go grab the community build, you can go to GitHub and you can grab the source. So one of the first things we said is, took a quick pass and you want to play with ChatGPT, point it at the public repos because the guardrail is, it's already public. The internal stuff we can work on as fast as we can but there's no reason that we can't have a reasonable set of policies to get started. That's right. And then we get to yes, use case by use case. Exactly. I mean, I always say never bet against open source especially in this day and age. No way man, never. Open source is now the software industry. It used to be a community thing. Now it's the communities, plural. And I think open source is now, I mean anything done that's proprietary is either under the covers or deep in some product differentiation that's not impacting much other than having a feature. So open wins. You look at open source, the fastest innovation is going on right now with AI and open source. You're seeing the proprietary foundation models getting replicated and decimated literally in open source. In what, three months? Tara, three months? Or if that. If that, yeah, faster models. Okay, so all the carbon footprint issues. Thin code, nice clean, fast, all done in the open. That's going to trump everything. Open source, it ultimately always wins. On the journey though, we are hyper conscious of protecting customer data. Yeah. And that's the difference between the pure open source environment and the enterprise encapsulation of it, right? It's our responsibility to understand those differences. But to your point, if anybody who doesn't understand the level of influence and power that the Linux foundation, the Apache foundation, those very key open source foundations have over the industry as a whole, I mean they're going to get hammered. I mean I hate to say this, but I'm old. But I remember when I was growing up in the computer industry, they were standards bodies. IEEE, IETF, W3C, and they were stewards of that old proprietary, and I won't say watchdogs as much as they were just transitioning and building the open source culture. The thing about today, there is no standards bodies anymore because a lot of the physical layer stuff's done. At the logical layer, it's the open source, it's the developers. They, no vendor can say we're the winner. If the developers don't like it, they're not going to use it, number one. So software's going to get easier to make, I shouldn't say that, it's going to be easier to rate software and the winners will emerge, but having data is not going to be as easy. So the data's the value, the software's the mechanism. Okay, so now you have this dynamic where, a lot of people I talk to are like, hey I love all this stuff, but anyone can, like I can spin up something that, I go to open source and I can get Reddit. Reddit's going down, I might as well replicate Reddit, but I don't have the data. I mean the data is your business, right? Whatever it is that you do, ultimately the data is your business. And so the manipulation and management of that data and in the industries where it matters, how carefully you cherish it and protect it with the software offerings that are engaged in it. Yeah, it's not too reductionist to say that the function of an IT department is to get the right data in front of the right eyes at the right time. And the UI is the delivery vehicle, but the data is the real jewel. You know I never said, I've said in theCUBE Dave Vellante and I have said there'll never be another Red Hat, I said that. Okay, because it's going to be hard to replicate what Red Hat did. We look at what's going on in open source right now. What Red Hat was to the Linux kernel, you can start to see why the data platform makes so much sense. You can go play in the open, that's where community is, that's where you can see the code, but when you want to run things and actually ride the wave and build, a lot of things are taken care of, because think about it, I'm a developer, when I go to the airport, I want to be in TSA Pre. Okay, I don't want to be in the other lane, okay? I want to be in the, I don't want to take my shoes off, I want to go right through. That to me is like the data platform. It is the model, right? And you think about it, developer gets in there and like I said, they grab the community bill, they got it running on their laptop and they're doing their thing and now they're telling their CEO and the CFO, it's like this is what we want to do and the CFO says well I don't want to hire 30 SREs, so we're going to throw that thing into Atlas and take advantage of it, not that it's SRE free, but it is definitely SRE reduced, right? Well come down to switching costs, right? The optionality fact, this is where I love your role coming together, because I see this very clearly, you got to have the operational understanding of what impact is. Costs and also switching costs and then you factor in culture, it's an easy path to say yes. That's why I have both the best and the worst job in the company. Yes, same. All right, what's the coolest thing you're working on right now? Oh, I'd say. Sandbox? There are a couple of really cool things. We are working on the, enabling the consumption workload centrism of our business. I mean, what is a workload and how do we take it from a concept into something that we can use to run and sell our business and forecast our business, do all the things that you would normally do with your ops teams. The other really cool thing is the one that Ter and I work on all the time, which is Gen AI and how do you use those innovations to safely, sanely juice up your operations because the promise is amazing. The potential is massive. And what's your vision on the impact on AI? Obviously, a lot of heavy lifting can be differentiated way, but also could accelerate value too, don't you think? It's going to accelerate value. I mean, the most precious commodity is the brain of your smart individuals. If we can make easy things easier, if we can save time, that will return dividends to the organization in terms of productivity. And I think it's interesting because you hear different sort of doomsayer sort of conversations like, oh, AI is going to cost all these jobs. And I would say, well, possibly, but I think a company that looks at AI as a way to say, well, I'm going to cut my support organization by 70%, it's probably not really thinking about it, because AI is really not AI. We really want to talk about pattern learning, machine learning, it's about up-leveling, right? More than replacement. The developer uses AI to write code. The AI is not going to debug the code and the customer picks the phone and says, hey, I've given you a whole lot of money for my support contracts and you were giving me a chat bot? No, that's not going to happen. I mean, when it changes functionality like that, that's a shift, that's a structural shift. And I heard the same thing when the web came. That's a kid's toy, that was the PC. The web was, oh, it's so slow. It's the terrible user interface. Hey, that 14.4 moto, man. Oh, God, it's so bad. But you're absolutely right. It's going to shift the way people staff about, it's going to shift how people write code, how people debug code, how people deploy code. But you're still going to need the same number of people. They may be doing different things. Different, exactly. That are higher up in the stack. That require more cognitive ability. So I love the developer platform. Let's circle back now to developer productivity, because right now that's kind of the topic that's on the table. There's a lot of supply chain issues with software, S-bombs, we've had that conversation before. Our chatbots got that indexed heavily. I'll ask that question. But you take care of all those under the covers, like paper cuts that a developer would look at and say, I don't want to get involved with that. Like I said, they went through the security problem. Same way. Given the guardrails, thank you very much. How does AI and how does the Atlas and the Mongo data platform make developers more productive? How would you answer that question? Well, there's a couple of things that Dave and Mark and Zaheer were talking about earlier today. There's the aspect of taking a set of developers who are new to document DB style, no SQL style databases. Speed of onboarding, right? To be able to get the assistance to help transition your mental model from a SQL base to a no SQL base. There's testability. It's like, hey, I'm writing a bunch of functions and I want to be able to get those functions testable as fast as possible, right? Upleveling that. Or I've changed things. Where are all the places the tests need to be updated or run, right? Kind of taking that idea of dependency management and up-leveling it. But even beyond that, security, what's the number one thing people complain about as developers is security right now? It's like, we got to produce these S-poms and we have to look at all these CVEs. 95% of the CVEs don't have a patch or they're not critical, they're just noise and now you're having to address them. AI around that to tell you what you actually have to care about in a more comprehensive way that's going to reduce the cognitive load, right? I think reducing cognitive load in order to enable velocity is going to be kind of the thematic benefit that we're going to see. Coding, testing, security compliance, all those things are going to benefit. And also one of the things that's come out of the AI push is that the notification bombing. How many notifications are we getting? That's right. That are firing off. Yeah. Yeah, separating wheat from chaff. Yes. Because there's a lot of chaff out there. Well you're in the middle of all the operation stuff so I was making a riff really, I want to get your reaction on. When Amazon came out it was EC2, S3, SQS, basic primitives. And I remember when EC2 had no custom domains that URLs were this long, but it was great. I loved it. Okay, it was good enough. I can do my local host, I can put it in the cloud and swipe the credit card. That was good enough. Now it got better. But all that was going on now is the Gen 1 SaaS was happening. I see the same movie going on here in kind of a way where the focus is what's the good enough picture ops wise and how do I get coding to take advantage of the benefits of the new environment? Which is agility, speed, developer assistance, whether it's humans plus AI. That's what everyone wants right now. And that is the challenge of the job. I mean you also have to remember that expectations have shifted too. What would have flown or would have been good enough five years ago is not good enough any longer. And so you have to know first of all what problem you're solving always and then know what is the best way to solve that problem. What is a good enough solution? Yeah, in a way, I was thinking about this recently too. You think about going back to Martin Fowler going on top of the mountain and coming down with the tablets of agile methodology, right? And you think about like, okay, and you talk to the state of DevOps report, what's the turnaround time? What does it make to be elite? But the complexity of what it means to run distributed systems in cloud, it's like actually it's increasing that cycle time. So in a way, I'm looking at these AI up-levelings actually gets us back to where we were before we made things this much more complicated. So it's actually taking something really complicated and then simplifying it down again maybe in one way. In one way. And it's also about iteration because again, what is good enough for a start may not be good enough over time. And so IT departments also have to go hardcore agile because they are developers as well and they had to use DevOps and agile methodologies. And going back to what you were saying around open source. I mean, I don't think MongoDB, I'm speaking for Dave here for a second, but I don't think we want to get in the business of making MongoDB specific models, right? And so that idea of where can we share models as an industry around things like DevOps or DevSec, Ops or DevSec, BetaOps, right? Well, if you're a platform, as you know, enabling value is what you do. That ecosystem picks it up, open source will pick it up and you're in an inflection point now, the three business model tried and truth method was reduce the time it takes to do something, make it go faster, make it easy to use. I mean, like that's what you got to do. Like those are like basic principles. 100%, which is why we're building a lot of custom tooling, particularly in our move to go for workload centrism, there aren't tools in the market. The market is catching up, so we're building our own and we've got a really good database platform to use. Well, we're getting the hook here. We went way over, this is a great conversation. Glad we got it on video. Final word, both of you, please share the vision of this event, the folks watching who aren't here, beginning of a 20-something city tour. What's this Mongo local about? What's the core bumper sticker? How would you describe what's happening right now? I start, you're the C. Oh, oh, you're looking for wisdom. We've got the best data development platform out there. We love our developers. We are going to make, just watch this space. And I think, you know, this is all about enabling, right, it's about how do we help you get what you need to get done in either enterprise, startup, you know, basic entrepreneurship, like we want to be here for those developers, all of them. Many great conversations there, great conversation. Coming together, operational and devs, productivity. Thanks for coming on. Live from theCUBE coverage here in New York City from Mongo local, we'll be right back. I'm John Furrier, host, stay with us. Got a great set of guests coming up next.