Great. Thank you to everyone for joining us as part of the Needham Growth Conference. My name is Mike Cikos. I'm the lead analyst here covering MongoDB. Pleased to say that we have with us the management team for, what should be a 40 minute fireside. I have some prepared questions on my side that I'll try to get through, but obviously you guys have a ton more questions which are way smarter than mine. So please feel free to lob those in, and we'll get them to Michael and Serge while we have them. The boilerplate stuff out of the way, Michael and Serge, thank you very much for participating in the conference today. We really do appreciate it. Thank you. Thanks for having us. It's great to be here as always. The easiest question you'll get out of the day, super high level, what does MongoDB do? Sure. You wanna go, Serge? You want me to go? It's your last rodeo. You go. Okay. So for those who don't know, we're the leading modern general purpose database platform. So if you think about the database market, it's a very well-established large market, you know, $90-plus billion annually being spent. So sort of real actual TAM, not like someday, one day, maybe it'll be there. And that's a market that's existed for years. And it was really based off of a technology today called relational technology that was important, and necessary and critical at the time in the 1970s, when it was pioneered. But things have evolved, and so there's a need for a modern alternative. And that's really what we at MongoDB provide. And if you think about the database market, not only is it large, but it's also growing pretty significantly, kind of, you know, low double-digit growth. And normally you think, isn't this market's been around for decades? Shouldn't a market like that grow more like a GDP? You know, why is the market growing so, so quickly, you know, given that it's so big? And the reality is, the reason for that is because databases are at the heart of all applications, right? Each application has a database, at its core. And you hear phrases like software's eating the world or every company trying to become a software company. And those all point to the fact that the way that companies derive competitive advantage today is principally as a result of their proprietary applications, the technology that they build, right? If it's off-the-shelf technology, like a standard SaaS application, it's available to everyone. That doesn't really provide a lot of benefit. And so the way they derive their benefit and derive competitive advantage is by virtue of building software. And that database at their core of that application determines how scalable, how agile, how nimble, how quickly can they innovate, all those things. And so that's part of the reason why the market is so strategic, and we see such significant growth in the market. Lastly, maybe I'll just talk about sort of, you know, why was there a need for a new alternative, right? You know, we've had relational alternatives around for decades. Why do companies, why do developers need a modern alternative? And the reason is because the way that relational databases were so successful initially was they were solving for the constraint of the day, which was storage, right? In the 1970s, storage was exceptionally expensive. And so if you could store things efficiently, that was very helpful. Unfortunately, that's the way that you store things, is you break them out, right? And I won't even use any database speak, but I'll just sort of, you know, try and give a generalist kind of a view. If you owned a parking garage and you wanted to store as many cars as possible, when the car drove in, you would disassemble the car and you'd put all the steering wheels with the steering wheels and all the fenders with the fenders. And, you know, you'd break apart the car and all that, and you get way more cars into your parking garage. The challenge is the parking lot attendant, when you gave him your ticket, had to reassemble your car and remember, where did I put, you know, the steering wheel for this car? And then they'd look at the sheet and they'd say, oh, it was basically in cell, you know, BC 478. And I have to go look and retrieve that. And that's basically the challenge that a developer has when working with the relational databases. All the data is stored very efficiently, but it's not natural and intuitive to a developer in terms of how to use it because developers use object-oriented programming languages, and so they have to map all this complex data. Whereas today, MongoDB storage is cheap, and so we, MongoDB, to continue with the parking analogy, just park your car as one whole car. We store it as a document. And then the developer can think of that document as an object. And so it works naturally and seamlessly. And so one of the reasons why we've had success, breaking out, 'cause we certainly weren't the only one who saw this opportunity. There were many companies that drew a lot of funding and everything else to go and kind of tackle this market. But the reason why we, you know, have excelled and kind of broken out is we had a better product, that captured kind of the hearts and minds of developers. So there's, you know, been developer preference for MongoDB. And then we've married that with solid execution. And that's sort of what's, you know, brought us here today to, you know, just under $2 billion in revenue. It was still a big, large opportunity, and single-digit market share, you know, where we are. So why don't I stop there? I can go on and on, but hopefully that's a helpful framing of it, Mike. No, no. Thanks. Thanks for that. Thanks for that, Michael. And then again, just in the interest of kind of dusting off numbers for folks, given how, how long ago the, the quarterly print felt, like at this point. But what were some of the key highlights from the October quarter earnings? Just again, quick recap. Yeah, sure. Yeah, yeah, yeah. No, no, we're obviously at the end of the cycle, but still that's the most recent quarter as we head into the quiet period and everything else here. So revenue for us was up 22% year over year, above the high end of our guidance. Atlas, which is our database services and offering, grew at 26% on a year-over-year basis. It's a little over two-thirds of the revenue, at 68% of revenue. Non-Atlas, which is mostly comprised of Enterprise Advanced and some other things, was up 14% on a year-over-year basis. That benefited from more than $15 million of multi-year license, versus a year ago. It was a strong operating income quarter. So non-GAAP operating income was just over 19%, and sort of had healthy, you know, free cash flow. So we were, you know, felt like we had a strong quarter, you know, ahead of our expectations. The new business environment continued to be healthy for us and executed well, both for Atlas and EA. The non-Atlas business, as I mentioned, significantly exceeded our expectations. Atlas consumption was sort of slightly ahead of our expectations, in, you know, what we would call sort of a stable environment. We saw modest seasonal rebound, versus, Q2, and consumption growth is slower, on a year-over-year basis, the same as it was for the first half of the year. So I, I think those are probably like the key highlights. Right. And I know we're in the new year as far as calendar. I know everyone's doing their work. There seems to be a growing sense of optimism. It seems like IT budgets might be up slightly versus where they were last year, but interested to the extent you can provide some color. What is the tone of customer conversations today? Where are they centered? Does there seem to be an uptick in any capacity, or is it again macro has not significantly changed, but just curious what you're hearing out there. Yeah. So generally, I would describe the conversations as constructive. We continue to, you know, see good sort of customer our value prop resonating well with customers. I would say I think that we're not necessarily perfectly indicative of some sort of like macro IT budget, right? Like back to my earlier comments, we're still a pretty small share. And so to the extent that there are like tweaks up or down, you know, in IT budgets, we're still a net share gainer and have so much opportunity that, you know, we can have success. And that's not necessarily an indication that budgets are, you know, growing, because we're just, you know, we're such a small player, even at almost $2 billion, you know, relative to the opportunity set. But certainly people are judicious with their dollars. You know, people are thinking about, you know, how to get the most bang for the buck, if you will. That's obviously generically always true, but probably more true over the last, you know, two years, as people think about it, but we feel that the value prop and where we kind of sit in the technology stack continues to resonate and be an important part. We've talked about, you know, the success we've had in new business, despite the kind of, you know, macro downturn, you know, coming up on two years ago or three years ago. And, you know, we've continued to persevere there given the mission criticality of the platform, what we offer, our position, you know, as a modern player in the technology stack, the relevance for MongoDB while it's still early for AI, and how we can help and, you know, be an important part of people's technology, you know, going forward, and sort of the future-proofing aspect of it. So, in general, feel good about those conversations, but, you know, I wouldn't take anything like, I don't, I'm not sure that I can give you like a crystal ball on like macro IT budgets just because, again, we're a pretty small player amidst the grand scheme of things, and we tend to be at the higher priority stack of opportunities. Got it. Got it, and I'll, I'll throw myself on mute after the question just 'cause I hear the fire trucks in the background. New York City, you gotta love it. Yes, indeed. But question for you on the new capabilities and functionality as well. Like you guys have obviously announced Vector Search, or streaming for that matter. Is there any significant change for like new logos acquired as far as, are you noticing that they're coming to you for these newer features or no, is that still maybe additive, but not why they would initially land with Mongo in the first place? And the genesis of the question is I don't wanna overinflate the innovation taking place, but if you could color that, that'd be great. Yeah, maybe I'll take a crack at that, so I think it depends a little bit by product. First thing I would say is that they're all still relatively early on, and when it comes to vector search and stream processing, they're both playing in markets that are nascent as well. Unlike our core market where, you know, it's well-established and large, these particular features or products compete in small, but, you know, fast-growing markets, and in vector search, we've had very positive feedback in terms of NPS scores and some of the sort of independent surveys out there. It's obviously a part of our sort of AI value proposition, but not the entirety of it. There is some evidence that on the self-serve side, customers are coming to us, maybe not because of the vector search product, but that they are taking it up. The new customers are taking it up sooner, which maybe indicates some amount of proclivity towards it. And then on streaming, it's really just exceptionally early stream processing. So it's hard to tell whether that's gonna drive new business or just new workloads and existing customers or even just add-on spend to existing workloads. Probably some combination of all, but too early to tell. Got it. Got it. And not to over-index on the prior quarter, but again, just wanted to get a better sense if we think through Atlas for a second. MongoDB discussed stronger business, solid new workload acquisition. Can you help us think about the cadence of business through the quarter or workload trends, if any, to call out? Like does October tend to be pretty, or the October quarter tend to be pretty evenly distributed, or is there anything to consider? Yeah. I mean, I'll come at it two different ways. The first way is a little bit of sort of walking and tackling inside the quarter. What we've said before and what's true now is that the second half of the quarter tends to be better than the first half because the seasonal recovery that we usually tend to see in Q3 happens, you know, in the back half. It's really usage growth post the summer holidays that sort of we see a bit of a pickup, and then second is, however, that seasonal recovery was more muted than expected. I think Michael mentioned that, and so there was a bit of an improvement, but not as much as in years prior on an intra-quarter basis. And that partially impacts our guidance and sort of how we think about sequential growth of Atlas 'cause consumption in Q3 really gets reflected in revenue in Q4 'cause consumption is sort of like the most real-time measure of the business and revenue comes, you know, a little bit after that. And then the other way that I would put it is, and obviously this is top of mind for the investor community and we get it, is what Michael has said is that consumption has been slower, consumption growth has been slower than the same period prior year in Q3 as well as Q2 as well as Q1. And generally we'd describe, obviously we had a disappointing Q1 and sort of revised our Atlas consumption expectations then. And I won't belabor the puts and takes, I'm happy to, but you know, maybe as a follow-up. But since then we sort of generally found the consumption environment to be stable. Nothing, you know, on the second derivative to call out one way or the other. But it does, consumption growth does remain below prior year. Maybe one element to hedge out there, but I know we're, we've spoken about macro, but why do you think that consumption element would be slower? Like, what are we attributing to that slower growth from a consumption standpoint? Yeah, so there were three pieces to it that we called out in Q1 and, you know, all remained true. First is we've seen a broad base, modest, but a broad base usage deceleration sort of across the base, which is reminiscent of other times when we've sort of seen a bit of a macro pressure in the business. It's smaller compared to the last two times we've seen this, one at the beginning of COVID and the other one, whatever that was, fiscal year 2023 midway, but it was noticeable and just because of its breadth, it was important for us to call it out. That's, so that's this in the, if you will, macro category. The second piece is we have seen workloads we acquired in fiscal year 2024 grow more slowly than expected. And those workloads are relevant for consumption this year because they are still early enough in their journey that they are growing meaningfully while they're not growing from zero like they tend to in the beginning. So they're in this kind of sweet spot. And so the fact that those are growing more slowly has an impact on consumption. And then finally, and only in Q1, we had a disappointing quarter in new business. And if you think about that cohort, the fact that it was smaller, that's just, you know, the tax on your consumption growth going forward. So that remains the case. And those elements add up to consumption growth being slower on a year-over-year basis. And to that, I think it was the second point, when we were talking about the new workloads acquired last year, just growing slower versus what previous cohorts have demonstrated, right? Has there been any improvement on that front or no, that still holds from where we are? No, it makes some minor puts and takes, but overall they're performing sort of along those sort of, revised growth curves after Q1. Mm-hmm. Then again, let's come back to the fact that October quarter did cite strong new business trends. What is management attributing to that strong new business trend then? Is it the fact that has there been some sort of change from a go-to-market standpoint that we're executing against? Was it just execution? Were there additional marketing campaigns? Just anything there? No, I would say it differently. Overall, with the exception of Q1, we've been happy with our ability to acquire new business, and I think Michael sort of mentioned that in his opening remarks, so it was, it wasn't so much that Q3 was the exception or something unusual happened. It was more that the one time we didn't kind of live up to our standards was the exception. That's it, but it's sort of a combination of all the things that we do day in, day out. I think maybe one thing, Mike, that's related to that that's sort of helpful to the investor crew is one of the things that we got and kind of took away or processed in the immediate heels of our December earnings call was this question of like, are you changing things in go-to-market? Because we, you know, we provided an update on things we'd said at the beginning of the year, but somehow people didn't hear the update part, and I think maybe connected some dots that don't get connected together in that, you know, with my departure, pending departure, we elevated Cedric into President of Field Operations and elevated one of his deputies, a guy named Paul Koumpouras, to CRO. I think people mistook that as some sort of much bigger reorg as opposed to sort of the natural evolution. That combined with the fact that we were providing an update on some of our changes that we made at the beginning of the year in terms of go-to-market about moving up market and things like that. People have kind of heard that was new and said like, why would you make any new changes in the fourth quarter? Is something going wrong or are you rejiggering things or everything else? We spent, you know, time in those first few days at the relevant kind of conferences and one-on-ones trying to help people understand that maybe they were not hearing it correctly or, you know, connecting the dots in the wrong way. And so I think we got to most of that, but we do still hear that, as kind of part of the cleanup and just wanna make sure that people understand that we really haven't changed anything. There are no changes in strategies. We were simply providing an update on what we talked about at the beginning of the year, and things like that. So hopefully that helps a little bit, but we haven't gotten that question or at least adjacent questions along the way. So we just wanted to properly address those. Got it. Got it. On the consumption front too, just wanted to call out again, most recent quarter consumption trends were slightly better than the revised assumptions that you guys had out there. And just to make sure, the consumption that you guys did see broad-based across all workloads, was it maybe specific to a geography or a specific fleet of applications? Can you provide some more color on that front as well? Yeah, it was slightly better and really nothing to call out in terms of a particular source. And I know this is a little bit of a cognitive dissonance, but I'm gonna try it anyway. It is slightly better than we expected consumption growth. It was slightly better than expected in Q3, but it was nonetheless weaker than a year ago. So both, both things remain true. Okay. Okay. And then just the route, like I know, we've been talking about Relational Migrator. Can you provide an update? Like, are you seeing an increasing volume of workloads coming through that Relational Migrator route or is it still in its infancy from where we sit today? Yeah, it seems like Michael is having some Wi-Fi issue. So I'll kind of take a first stab and hopefully we get him back. So, I would. No, it's a good question. Yeah, sorry about that. I would divide the relational migration story sort of into two chapters. The first chapter has sort of been around since before we've been public, which is we always had some amount of relational migration business. You know, we talked about in our filings, roughly 20%-25% of new business on EA usually is from relational migrations. It's less on Atlas only because the friction on onboarding new applications is so much lower on Atlas that that's really what our reps focus on. But relational migration has always been a portion of the business because for some number of applications in the normal course of business in a market that's $90-plus billion, the pain on running on relational because it becomes so acute that the company or the IT decision makers make the decision to transition. It's either because the app is so slow that it's impacting revenue, or so fragile that it's impacting whatever regulatory environments you might have. Cost is occasionally the issue, but it's usually something about just the thing is crumbling. It's not working anymore. So I'm gonna hold my nose and replatform it. And I say hold my nose because replatforming is not an easy thing to do. So that's the historical sort of story and it's been with us for a long time. More recently, really in the past year, we've become increasingly bullish about our ability to combine some of our tooling, some of our professional services and the AI tools that exist out there to meaningfully lower the cost and the time and the risk of relational migrations. And this is what we talked about a couple of pilots that we ran early in the year, we basically found customers who were willing to partner with us, dedicate their own resources alongside our resources to pick a few of their existing relational applications, very old ones, and work hard to replatform them using the combination of services and tools that we sort of brought to the table. And the early results are very encouraging in that, not only are we reducing cost and time, but we're reducing risk using AI tools because you can really use AI to test application to understand how it works. And then you can, when you rewrite it, you can test the new application to make sure that it works exactly like the old one. So that ends up being a very powerful story for the IT decision maker and one that's resonating frankly not even at the level of you know database administrators or developers but at the C-suite level because the brittleness of an architecture the cost and the complexity of it is a meaningful problem. So we're seeing demand. We're doing more of those engagements A for us obviously to help customers but more importantly for us to keep getting better at the process. We're encouraged by the interest that we're seeing from the customers. And right now the focus is really just on doing more of them developing a playbook ensuring that we get all of them right 'cause that's very very important. But the size of the market is that there's sort of, in the fullness of time, there is significant demand for this, but we still have a lot to figure out. Understood. Understood. And I know on the, again, MongoDB has been citing, I guess these strategic initiatives and one of them, you can call it go-to-market, but, I guess this investment in the enterprise, not surprising. Larger organizations have stickier budgets, might be tougher to crack that nut initially, but once you're in, you're in. And we've seen this across our broader coverage. People are just moving up market. How has this expanded strategic account program tracking for Mongo versus expectations? 'Cause you guys haven't piloted or tooling around with this before making it a more concerted effort, right? Yeah, yeah, yeah. Yeah. So first of all, thank you for the setup 'cause you got it all right, but let me just expand a little bit so that everybody's on the same page. So we started our strategic account program in fiscal year 2021 as a pilot. And the general idea was pick some of our most promising accounts, and give them more resources, either more rep time or more technical resources or customer success resources or marketing resources or, or, even some support resources and to see if we can get a disproportionate return on that investment because to your point, we already have significant traction in the account. And by and large, that effort has been successful. Not every account worked, but on average, the incremental investment, the incremental return more than justified the incremental investment. So we've been slowly and deliberately growing that program over time, and really what we've learned that is key is that you gotta keep the bar high of what makes a strategic account. You need to have a certain amount of spend. You need to have visibility into some near to medium term pipeline. You need to have very strong technical champions. It really helps if they already signed your, their cloud services agreement, which means you've gone through their procurement and security org. And then, you know, you might be in a position to make the incremental investment and get the incremental returns. Like the way that I try to paint a picture of it is, strategic account investment is like pouring gasoline. If you pour gasoline on a dry piece of wood, nothing happens. But if you pour gasoline on a piece of wood that's already on fire, you get a bigger fire. That existence of fire is sort of key ingredient. And so what we're seeing, what we're doing in fiscal 2025, but really going into fiscal 2026 is increasing the size of the strategic account program. That's our version of dedicating more resources up-market. So it's not generic. It's not broad-based. It's really at the very high end in our largest customers. And it isn't because we're seeing poorer returns elsewhere or that we are seeing something happening elsewhere in the market that requires us to pivot. What we are seeing is a greater opportunity to invest in strategic accounts, not because we're lowering the bar, but because we're seeing more accounts clear the bar. And more accounts are clearing the bar partially because we're keeping, you know, evolving and maturing as a business in the eyes of IT decision makers. So more people are willing to bet on us. But also what we just talked about, relational migration is getting, you know, kind of more C-suite attention, which is allowing us to just have a more holistic conversation of what we can do for them. And then finally, AI. Customers are, you know, working through what their AI strategy is and looking for companies that they can help us, they can help them. And certainly some of our largest, most promising customers look at us as a partner to not just provide them the OLTP layer, but just to be sort of a strategic resource as they think through all the pieces. So the tenor of the C-level conversation is changing, and that's the opportunity to invest more in the strategic account. And what I like about what we're doing is that that isn't just some sort of top-down idea that, you know, the senior management came up with and is rolling out, but there's actually demand from the field and our bottom-up planning that was complete support from the field organization to do this. But that's what it means to us. And then the second chapter of that is, money isn't infinite. So if you're gonna invest more of it, you know, in the strategic account program, you gotta pull back somewhere. And it's classic capital allocation. We're adding to our best payback channels and reducing payback in channels that are sort of near the bottom of the pecking order, which for us happens to be, and by the way, always has been the mid-market. So that's what I know a lot of people are quote unquote moving up market, but to us it's, it has a very specific flavor. And it's driven by specific, you know, positive signals that we're seeing in the market that didn't exist a year or two ago. I think that's the thing, and hopefully it's obvious, and folks can probably connect the dots. But just to make it super transparent, as Serge was saying, those strategic accounts are the ones we always fund to the maximum that makes sense initially, like in any planning process, because if they are your highest returns and your fastest payback, like why wouldn't you prioritize those? The key thing is based on a whole bunch of factors, including investments we've made over the last few years, relational migrations, all the things that Serge was talking about. There's just more opportunity to do that now, and we're trying to take advantage of that. Great. And has the company, and apologies for not having this, but has the company in any way quantified either the size of the initiative from a manpower perspective or how many people are being reallocated for this or no? No, but keep in mind this is really a sort of an elite program for a small minority over 7,000 direct customers. Mm-hmm. And the thought process then too, I know you said it's nothing against mid-market, remains fine there, but is the thought process then for mid-market better served by channel or maybe in a self-serve motion? Is that how would you think about that? As well as scaling some of our human teams to cover more of it, so it's seeking efficiency as opposed to, you know, pulling the plug, if you will. Okay. And the second piece as well, again, I'm just trying to go through the initiatives that I have on my side, but one of them was maybe around building up the pro-serve capabilities, either directly or through partners. And I just wanted to get a better sense. Is that, should we interpret that as helping enable in any way the strategic account initiative or no, does that potentially? No, that's related to relational migration. So that's related to the app modernization 'cause services is a part of the equation. Particularly early on, you really need people there. Tooling is not enough and AI, you know, models aren't enough either. You need your people on the ground to connect with the customers. And the reason is because relational environments are very heterogeneous among customers in terms of programming languages used in terms of the actual relational database use, ORM layer on and on and on. It's very multiplicative. And we're early on, you know, we're still have done this for a relatively small number of customers. So each is still in bespoke. Although obviously as we sort of acquire more knowledge in this process, there'll be more templates and more things that we can automate and kind of do effectively and automatically, but for the time being, there is a human element and that's what the professional services investment is about. Got it. I think one of the other things that I know the management team in here has called out is the idea that yes, there's a lot of experimentation for some of these AI-based applications, but the majority, it's still unproven whether or not how many of these attain product market fit. And so I was wondering, do you guys, is there a parallel that we can draw for, I don't wanna say non-AI, but the original or initial cohorts that maybe didn't have such an AI-based flavor to them, like do the majority of those not have product market fit? And then eventually some of them hit, like how do I, is there even an analogy there or no, is it just really working its role? Yeah, I think, let me, let me start with just telling you how we see the AI market at this moment in time. Mm-hmm. First thing I would say is, not very many people in the world knew what ChatGPT was 24 months ago, and so this whole AI revolution is really, you know, two years in the making and two years it seemed like a long period of time if you're reading the newspaper every day, but in the world of enterprise IT, it actually is, you know, things are just getting started. Mm-hmm. However, we are living in this very interesting cognitive dissonance moment from our perspective, and this is the following. There's a relatively small number of very, very popular AI applications that have caught popular imagination, right? Number one, ChatGPT and number two, Anthropic, right? You know, those have billions of dollars of revenue associated with them and everybody's using them, right? We are using them, our children are using them and on and on and on. And they paint the picture of the world of AI. But once you move past those relatively few highly popular and visible applications, there's relatively few that have truly made product market fit as revenue measured in the tens forget hundreds of millions of dollars. And so if you think about the company to be a $7 million seven-figure customer for us, probably does need to have, you know, $10 million in revenue. It won't be enough for you to have database spend on $1 million plus, right? So like you need companies to generate their own business and their own revenue to be a meaningful contributor to us. And simply the number of them right now isn't particularly large. We talked about one seven-figure customer in our call, but it's really an indication of the relatively few customers, relatively few AI applications or companies that have reached that meaningful mass. Will that change in the fullness of time? Absolutely. Will that number be higher a year from now than it is today? Absolutely. We think for reasons that Michael covered that we have every right to win a disproportionate share of those, meaning disproportionate to our current market share, right? That's the opportunity we're excited about. I don't know that I can give you a parallel around prior sort of product adoption cycles or technologies and sort of what it means for product market fit, other than to say that nothing that we're seeing with AI is unusual in that new technology is being deployed. First, it needs to make it into production, then some and then more will make product market fit. It's really just about the moment in time that it was. The only thing that's perhaps different is that there are these very relatively small number of very large, very successful applications that are maybe painting popular perception around how far along we really are in this AI journey. Mm-hmm. Understood. And I do wanna be true to my word here. We did get a question in from a client that I wanna ask while we have you. Question goes, I know it's early, but can you speak to customers that have Gen AI in production and are therefore doing inference? Is MongoDB being used on that inference component? I would think large organizations that are starting to tinker might be using Mongo for that purpose. Anything you can share? It's really premature. There's really very little in terms of patterns that we can infer either in terms of, you know, the intensity of the use case or anything that sort of generalizes those. Obviously, we're eager to find out ourselves, but the moment in time that we are in is still early. Got it. Well, we've talked about this, you know, and it's not unique to MongoDB, but it's broader in that there's a lot of enthusiasm, there's a lot of experimentation, obviously in investor settings and conferences and everything else. People talk about it a lot, but that's sort of like reasonably ahead of where the day-to-day reality is. And I think Dave gave an example of, you know, one of the largest, you know, financial services companies in the world, tens of thousands of developers. And I think last time I heard they had something like, you know, 20 AI applications in production, you know, despite having tens of thousands of developers and zero of them were customer facing because they just didn't trust them, right? So, there, there's a lot of enthusiasm, but a lot of concern, you know, especially about how it impacts, you know, reliability, hallucination, you know, all those kinds of things. And so, there's, you know, it'll take time for it to play out, and that's not unique to MongoDB thing. I think that's sort of a, a broader thing, but we're, you know, you know, eager to, you know, capture our, our kind of more than fair share, if you will. Excellent. Maybe time for one more here, but this is kind of just a little question I had. I'm happy I can stress test it against you guys while we have you. With Gen AI, does that place a greater emphasis or greater demand on time series-based capabilities given the need for real-time insights or not necessarily? No, time series is a specific use case. You're right on real-time, but real-time and time series partially overlap but not fully. So, we're happy with the traction that we're seeing on our sort of time series offering, but it's a particular type of use cases that we build and we would expect AI to be broader than that. Understood. Okay. And I will keep it. I said it was the last one; that'll be the last one. But thank you very much for the time, guys. I really do appreciate it. Michael, been great working with you. Thank you again. Yeah, no, likewise. Thanks for everything. Apologize for a little bit of Wi-Fi issue in the middle there at the hotel here, but thanks for everything and appreciate you always taking the time. Excellent. Have a good day, guys. Take care. Thanks.