Scaling App Development Beyond the Data Team

Summary

Databricks Apps has seen 50% more customers, doubled apps, and tripled users, driven by a wave of coding-agent-built applications increasingly created by teams outside the core data organization.
Three challenges—enabling citizen developers, governing app sprawl, and controlling compute cost—are addressed through Genie App Builder, App Spaces, and serverless micro apps with scale-to-zero micro VM technology.
Thales, a French aerospace company, built a production Databricks app serving hundreds to thousands of users for predictive maintenance analytics on inflight entertainment systems.

Scaling App Development Beyond the Data Team

Coding agents have triggered a Cambrian explosion of new data apps, increasingly built by users outside traditional data teams. While this is an enormous opportunity, it also creates headaches for admins who need to manage cost and govern app sprawl. In this talk, we'll do a product deep dive into how we're helping scale app development while keeping admins in control. We’ll cover our new serverless architecture, app spaces, and the Genie App Builder.
Talk By: Jon Levine (JPL), Product Manager, Databricks ;

Chapters

FAQs

What is the Genie App Builder on Databricks?

The Genie App Builder is a native, AI-powered vibe coding experience that lets non-technical users build data applications directly on the Databricks Data and AI platform. It reduces the barrier to entry by enabling plain-language app creation without deep coding expertise.

How does Databricks handle app governance at scale?

Databricks introduces App Spaces, which separate resource configuration from app creation, allowing admins to centrally manage shared compute, identity, and access control. Cost management and audit capabilities are built into App Spaces to prevent uncontrolled app sprawl.

What are serverless micro apps in Databricks?

Serverless micro apps use micro VM technology with scale-to-zero capability, so compute resources spin down automatically when apps are idle and spin up on demand. This architecture helps control costs while making app deployment simpler for teams outside the core data organization.

What is a real-world example of Databricks Apps in production?

Thales, a French aerospace provider, built a Databricks app to deliver detailed analytics on inflight entertainment usage and predict when seatback units need maintenance. The app serves hundreds to thousands of users and is a core part of the Thales product offering to airline partners.

Full transcript

[00:08] Hello everybody. Thanks so much for coming. I'm going to be talking to you today about how we're going to help all of you scale app development beyond just the data team. My name is John Levine. I'm on the product team here at Databricks working
[00:21] on Databricks apps. This talk is going to be a little bit of a sneak preview of what you're going to hear in the keynote tomorrow. So, thank you for keeping it quiet until you know, like 9:00 a.m. tomorrow.
[00:35] So, here at Data and AI Summit, you're going to hear a lot about agents. I'm sure you've heard a lot about agents this morning. You're going to hear about our mission, which is to democratize data and AI. And
[00:47] I think one thing that's amazing about working with agents is that they're helping all of us democratize data and AI. They help us build data pipelines, analytics, dashboards.
[01:00] I think at the end of the day, if humans have a role to play in all of this, then we need interfaces to see, touch, and feel the data. And of course, Databricks is helping
[01:12] provide those interfaces through AI BI, through dashboards, through Genie 1. But sometimes you need to go a little bit beyond that, and you want to build a custom interface to interact with the data. And that's where apps come in.
[01:25] Apps are a bet on humans. So, I stole this line from Andre. I don't think we have Andre in the room. Andre is the tech lead for apps. He's the creator for apps. He likes to say that apps are a bet on humans. If you think that humans need to work with data, you want to
[01:37] build custom interfaces for humans to do that. And people are building amazing things with apps. So, um Thales, this is one of my favorite
[01:50] examples. Thales is a French aerospace provider of aerospace technologies. One of the things they do is they build uh the like seatback entertainment systems. I'm sure everyone's used these on on the aircraft that they've been on.
[02:04] And they want to provide analytics to their airline partners and super detailed breakdowns of like what are people watching on the flights and like even like per route like what movies people are watching and how much time they spend watching them. And they build
[02:17] models to like predict when these units fail so that the airlines can do predictive maintenance on that. They do that all with an app. And it's pretty amazing. And of course an app like this it has hundreds or thousands of users and there's like a dedicated team of people.
[02:30] This is like a core part of their product that they use and it's built on Databricks apps. Um and while these use cases are amazing, what we're seeing is now that there's so many people inside of your organizations who have
[02:43] ideas for apps. Right? Probably not just all of you in this room but people outside of data teams who are not like data people who want to make apps. And if they can describe the app they want, they can bring it to life at Weave with a front end. And so what
[02:56] we're seeing is that apps are really democratizing app agents are helping democratize app development. Uh but there's a small problem which is great. People are able to There's so
[03:08] many tools that can help Weave code an app. You can describe the app. You can You can like get the interface you want but how do you make that thing real? How do you actually connect an app to your production data? So great. You Weave code the UI. Um
[03:22] how do you deal with authentication and SSO single sign-on? How do you define governance rules about the people who sign into your app, what data can they actually access? Uh are these Weave coders are they like now SREs? Are they managing the
[03:35] infrastructure? Are they deciding, you know, the containers that things are deployed in? What about a database? Okay, great. We have Lake Base. We can persist state in the app. Uh what happens when your data model changes and need to migrate a
[03:47] database? Have any of these people migrated a database before? Uh Uh, what about observability? Have you thought about observability for your app? Have you thought about audit logs? How about supply chain risks? You know, you're pulling in dependencies, are those secure?
[04:00] So, those are things that folks who are vibe coding apps probably are not thinking about if they're not professional app developers. Now, Databricks apps were really built to solve a big chunk of these problems.
[04:12] Uh, >>
[04:24] plane where your Spark jobs or your data pipelines live. It's secure and it's governed, so you can control access to apps through Unity Catalog. And it's open, but it's opinionated, so we support a wide variety of Python and JavaScript
[04:36] frameworks, but we have our own SDK. It's a very opinionated way to build apps. And I would say that, you know, the app started about two or three years ago, and the focus really was more on the professional developers, people who
[04:48] understood a lot of those problems and understood how they wanted to solve them. But, everything has changed and really just in the last 6 months. Uh, I'm sure many of you are seeing this. Um,
[05:00] I have a couple graphs here. These graphs go back uh, about a year, basically to last year's data and AI summit. And you see this inflection point, actually even less, like something like 3-4 months ago. Um, so the top one is the number of
[05:12] active apps that are running every week. And the bottom one is actually the number of users who are signing into those apps. So, what we've seen is just in the last 6 months, so since December 1st,
[05:25] we have more than the number of customers who are using apps at all has grown by more than 50%. The number of active running apps has more than doubled. And the number of people who are interacting with apps has more than tripled.
[05:37] So, we think this is really amazing. We're super excited about this. Uh, hopefully many of you are excited about this, but from talking to a bunch of folks, maybe some folks who are here in this room right now, I'm hearing some concerns as well.
[05:49] Um so often I'll talk to folks who are like, you know, platform admins and in in the same conversation I'll get feedback like uh this is first one's from one of our retail oh one of our retail customers. You know, apps have changed the way that we provide value as
[06:02] a data team. So incredibly excited that it's transformational for them. But on the flip side, uh you know, we're working hard to keep up. We're working day and night to put in fixes. So good to the bad. Um another customer
[06:16] who's a uh digital native software firm, um they ran a workshop where they had people around their company vibe coding Databricks apps. They had executives building apps. And they said this was the most excitement, the most adoption they'd seen of Databricks in a long
[06:29] time. But they're also very scared because random apps are able to proxy permissions and do whatever people want. So how do they prevent this? So if you've ever seen the like two guys on a bus meme, it's basically this meme but
[06:41] like one guy looking at the different windows. Um So what's going on? What are people so concerned about? Um and I think there's like there's three things. There's three problems that we know we need to go solve.
[06:56] So problem number one is what we're calling this citizen developers. This is this new breed of developers. These are folks who are not professional developers. Like they don't know what Git is or CI/CD, right? They don't know what ETL stands for. Um they want to be able to vibe code
[07:09] apps. That's great. The problem is and we have some great partnerships by the way with um uh like folks like Revelate and Lovable. Um they can build the front end but how do you how do those apps actually understand the data and understand Databricks? Understand
[07:21] Databricks apps. That's challenge number one. The second problem is that again app administration and app governance is again really built for this world where every app was special and like there was a development team that thought
[07:33] carefully about it. And so the sort of governance and settings and the access control is like coupled to the app development itself. It doesn't work in a world where there's like 10 times more apps getting created.
[07:45] And speaking of 10 times more apps, we've heard you that apps are expensive. If I vibe code an app, and let's say I just want to access it for a moment just to use it, um I'm often going to be paying for the app to run all the time, even if I'm not
[07:57] really using it. And you know, at Databricks, it's never our intent that, you know, you're paying for things that don't provide value. We want you to pay for apps when they're actually useful. Okay, so this is now the big reveal what you're going to hear about in the keynote tomorrow. So,
[08:11] surprise, we're going to fix all these things. Uh we are going to be releasing the Genie app builder. So, this is going to be a native vibe coding experience that's powered by Genie. So, if you if you've ever used something like Lovable, it's going to be
[08:24] kind of similar to that where you're going to be able to describe the app you want, and Genie is going to have an interactive flow to actually build that app. Now, how does the app builder know what to do? Um and a challenge we're seeing from
[08:36] folks is like you have your app developers, maybe they can describe the app, or they may not even have access to the underlying tables or the warehouses they need to build that app. But, how do you give that access to the app? And this is a problem that's solved by
[08:49] app spaces. So, you can think of I'm going to We're going to spend a lot of time in this talk going into lots of detail about app spaces, but the basic idea here is we're going to separate out the configuration and the resource management for apps from the process of building an app. And
[09:02] we're going to let you apply that those policies, that configuration to groups of apps, and assign the ability for people to create apps within an app space. And finally, we're introducing a new compute
[09:15] primitive, a new type of serverless compute which we're calling serverless micro apps. It's based around micro VM technology. So, what we're trying to do here is kind of push the Pareto frontier, if you will, of like fast startup and lightweight and isolation.
[09:27] So, what we what we want is for these apps to be very secure, very lightweight. Um they should scale to zero when people are not using them, but more importantly, they need to scale up to one quickly when someone does access
[09:39] one of those apps. The team's been super hard at work to make this technology possible so that you're paying for apps when they're being used, and this supports this new the fast fashion era of apps.
[09:52] Okay, so I'm going to hold something for tomorrow. I don't want to give away all the surprises. You're going to get it up. Clicker's a little slow. Go back. There we go. This is my non-demo of the app builder. You don't
[10:05] have to wait till tomorrow to get the app builder demo. It's going to be great, I promise. What I want to do with you all today is go deeper into app spaces. Okay, so what does it actually look like to configure one of these app spaces? What's what's
[10:17] actually in that? So, I'm going to switch over to a demo now. Please bear with me here. And can you all see this?
[10:30] All right, let me just make this go away. Okay. So, has everyone seen the new By the way, how many people here have created a data works app before? Okay, I great. I figured most of you. I didn't we didn't get too much into the
[10:43] basic like apps 101, right? We're going we're going right into the the 201 level. So, you've probably noticed that we've recently moved. We are now in the app switcher. There's an what we like to call the apps app.
[10:55] And maybe some of you are like, why is there an app for apps? That's because we need we need a home for some of the new things we're building. So, um I got I I'm not going to demo the app the app builder, but
[11:07] you'll there'll be a new like place in the apps app to to start to build these apps. You see that when I start using the app builder, it says JPL test space. What is what is this thing? So, if I click on this, this is a list of spaces I have access to. Of
[11:20] course, this is our demo environment, so there's like a thousand of these, but you know, it probably most people might only have access to like one or two or three of these spaces. Okay, so what is an app space? Let's just get into it. I think the easiest way is just to show you the
[11:32] configuration of one of these. So, I'm going to go to one that I've already run. Can everyone Can everyone see this? Let me make it a little bigger. Is that better? Yeah. So, okay. So, right off the bat, um
[11:45] what is this? This is the overview page. So, oh, yeah. So, between when I made this and the the demo, uh the apps got deleted. But, you can imagine >>
[11:57] have been created in the app space. And this configuration applies to all of them. Made more sense when it said like two out of seven, but um but the whole point is like apps, you know, some of them are running, some of them are not running. People can create apps in this app space. There's many apps that you're
[12:09] governing. It's not just like configuration for one app, it's a configuration for a set of apps. And we'll also include some analytics about like how many people are accessing the apps. And we'll we'll build more of this out over time. Okay. So, what's what do you actually do
[12:22] in the app space? So, let's walk through what each of these boxes are.
[12:34] do your app devs have the ability to grant access to their app to SQL warehouse? You want to just define that up front. Like, what is the warehouse that the apps in this space are going to use? Um there's also secrets, there's also AI
[12:46] gateways. Um anything that's a workspace resource you can manage. So, if I click into that, if you've ever built an app, you're familiar with that interface. It's very similar where I can actually choose the shared resource I want to add and add that into
[12:58] the app. So, if I go back, you can also grant access to shared data resources. So, in addition to like a warehouse or an AI gateway, you can give direct access to a catalog, schema, table,
[13:11] a lake base. These would be like existing tables that are already out there in the world that you want apps to be able to use. So, in this case, I have granted access to a couple of different tables. Now, one thing I hear uh really commonly
[13:23] from many of you is uh I've heard, you know, I don't want apps to embed credentials in the app. There's a concern that, you know, you don't want a very powerful developer to be able to expose data to users that
[13:35] users should not have access to. And you want apps to only work on behalf of the user or OBO. And the way you would configure this is by simply not giving apps access direct access to
[13:48] data, right? If you don't configure access to the data, the apps won't have access to that data. And instead, what you can do is you can configure API scopes. So again, this is one of those concepts that I don't I expect like the citizen developers, like are they in the details
[14:00] of like OAuth and API scopes? Like probably not. But as an admin, you will be able to set what are the scopes that an app can request. And so here I've requested SQL and API gateway. We're actually request
[14:13] We're actually creating this new scope, um which is this restricted query scope, which is going to like only limit the set of SQL operations that you can do. I can save that. Um oops.
[14:25] Good right here. Um finally, there's also there's also cost management. You'll be able to add serverless budget policies to this. Um and of course, there's there's user management, right? You can set who can actually create apps within this app space and who can manage
[14:37] and actually change the settings of the app space. So I'm going to switch back to the slides. There's a couple other features in here which we This is This is the current working internal version.
[14:50] We're hoping to have a preview out for you to try this all very soon. We'll share more about the timeline at the end. Um I want to switch into Let me see if I can go back to the slides. Awesome. So I'm going to get into some
[15:03] more of the details about app spaces. So first of all, like how are people going to use these? Um Well, I mean, I think one answer is you're all going to tell us, but there's a few patterns that we've seen emerge. One is this idea of like a personal app space. So it's very common that you want
[15:17] users in your workspace to be able to create apps for themselves. You You not be excited about those getting shared super widely. uh but you at least want the people the able to the ability for people to like to vibe code their own apps. And so, to do this, you could create a space that
[15:30] um doesn't have any built-in data access, but gives access to, you know, like a starter warehouse or an AI gateway, um and lets people request API scopes at least for themselves. There's also the idea of like a team space. Like, for example, at Databricks,
[15:42] we have like the product management team, or we have our field engineers who work on like, you know, manufacturing in the Americas. And so, if they cover that vertical, they might have a shared group of datasets that they want to collaborate on together, and you might create an app
[15:55] space for that team. And finally, we're seeing a lot of our customers, let's say you're, you know, uh consultancy, and you have fixed-length engagements, you know, of 3 weeks or 6 weeks you're working with a customer with a client, and there's a
[16:08] set of datasets that you're working on, and you might want to create an app that's very focused around that engagement. That's another pattern we're seeing where this could be really helpful. Okay, I want to shift gears a little bit um and get a little bit into the
[16:21] details. Um I will stop before the end and make sure there's time for everyone to ask questions, but I'll try to do my best in the next few slides to like cover like the most common uh questions that people have about this. Um
[16:33] So, let's talk about the cost piece first. So, I mentioned that um with app spaces, the apps that get created are going to run on this new lightweight serverless infrastructure. It's going to support scale to zero. So, the thinking here is like, if the apps
[16:46] are running when people are using them, we're hoping that that will um really alleviate the cost of running these apps, right? And serverless usage policies will then be able to control to let you uh tag the usage and be able to
[16:58] attribute usage not just of the app, but also of the resources in the app back to like a specific cost center if you want to if you need to support chargebacks.
[17:11] Um identity and access management, this is a really big topic. So, I think one thing when I talk to folks who are trying to deal with this explosion of apps, one of the most common things I hear is, "Can you please stop auto generating service principles?" Yeah, some people at least
[17:24] have have dealt with this. Yeah. Yeah, we've heard you. Uh sorry about that. Um Uh so, I think we are still working through all the details of exactly how this is going to work, but the basic idea is we want to make sure we're not
[17:37] in a situation where every app that gets created is this new service principle where you actually have to manage the grants on that service principle. And so, we're exploring a few different approaches to this. Um Uh I think you know, as we break out,
[17:49] we'll have like smaller group conversations. Uh as we know more about this, we'd love to get feedback on on some of the things we're exploring here. Um but the basic idea is that an app space should have declarative access to data and resources. So, you should be
[18:01] able to configure the app space like through the UI or through API or more importantly using Terraform or or like infrastructure as code to say, "Okay, apps created in this app space, this is the set of resources they have access to." And you shouldn't be like, "Oh,
[18:14] okay, like I have to click here to create the service principle and like change the grants on that." Um something else that probably won't be in the first version, but we know we're going to need is some way to give like unique resources per app. So, what I
[18:26] mean by that is and this wasn't in the demo that I shared. We know that apps very commonly need like their own database. So, an app needs like its own like base that it manages the life cycle of. And so, that's something that we know we're going to need to provide as part of an of an app space. So, you can
[18:39] imagine like a checkbox, "Okay, every app gets its own like base or every app gets its own volume that it can manage." Um another very important thing on this piece is audit.
[18:51] So, it's very important that even if we're not like creating these uh service principles, there's still a way to uniquely see the app IDs for the purposes of auto logs. So, every action that happens
[19:04] by an app, you should see it logged as, "Okay, this app took this action on behalf of this user." Right? Um and finally, yeah, we talked about the OBO scopes. I think there's a bit of nuance here and something we've heard is
[19:17] so scopes, if for folks who aren't familiar, scopes, I think of an think of an a scope is like what is the API surface that this app can access? So like SQL and like within SQL you might limit that a bit more to like read-only operations. Uh you know, there's like a
[19:29] set of APIs around jobs or like around managing UC endpoints. We know that we also need to control the um the data sets that apps can access even with OBO. So for example, if
[19:42] someone signs in using OAuth through an app, the app now can't shouldn't just be able to see like anything that user can see. You may have to put constraints around like, "Okay, well, only data in these catalogs are like fair game for access with OBO." So we're going to need
[19:55] to put in those kind of constraints as well. >> All right. Let's uh did we go too far? Okay, great. So the next one is about security. So this is a really big topic here. It's a
[20:08] lot to cover. So I think the first thing, for me the like the baseline for security is many of the companies, I think you all work for, have policies around like what code can get deployed to production, right? People can't just like push things out that like go into
[20:20] prod, haven't been reviewed by a human, haven't passed their CI/CD. A lot of you are using GitHub Actions to already enforce this today. Apps can't just circumvent that whole governance model. And so we know it's really important
[20:33] that even vibe-coded apps, before they get deployed, before they get published, before other people can use them, have some way to integrate with CI/CD. So I'm not sure if folks are aware, but today we have
[20:46] Git support in apps. Actually, how many people are using Git in apps? Okay, only a few. For me, some of you are, okay. Yeah, so I encourage folks to try it out. Um this is another one where I'll say we're working through the details of how this
[20:59] will work with app spaces. So today, um to use Git with apps, the app developers need to link their Git credentials to Databricks, and then you can enforce that, you know, when you know, the app one gets deployed, like when it gets pushed to like the main branch.
[21:12] And we're again working through the details for app spaces, but one thought is you could set this up at the space level, and there's a few ways to do this, but you can imagine like perhaps the app space has a Git credential where for it
[21:25] to deploy, it has to merge things into a main branch. And you can all set up, you know, GitHub actions or like CI policies on that branch or like require human review for things to get deployed. So, I think that's really important. Um but we don't want to just push this all on you.
[21:38] Like we know that um there's a lot of responsibility that Databricks has as well to make sure things stay secure. Um so, jumping ahead a little bit, uh supply chain protection. This is a This is a topic we think a lot about. Um Databricks already today at the
[21:50] workspace level lets you configure default uh package registries for Python. We're going to be adding support for Node and for probably for other languages, other ecosystems over time. And the idea is that apps should be able to inherit from that, and you can you
[22:03] should be able to just set that control up at the workspace level. And then going back a bit on the on the networking side, we want to make sure that you can control We already have, you know, serverless egress policies at the workspace level. Those need to be
[22:15] able to flow through to an app space. So, if you do want to say, "Hey, I want to ensure that people are not able to download packages from these locations that I, you know, that I don't control, that everything is going through the the defined locations for for packages, for
[22:28] example." Uh another one is observability. So, apps today have We've recently launched uh support for OTel export. So, today at a per app level, you can configure like a catalog and a schema where your logs,
[22:41] metrics, and traces live. AppKit, which is our SDK, uh lets you configure and customize those a bit as well. And again, we're going to have a way to do this at the app space level so that you're not like relying on developers to
[22:54] remember to turn this on, that, you know, for that app space, okay, we make sure all apps out of the box come with observability. Um on the roadmaps, some things I think we know we're going to need to do would be uh credential scanning. So, again, I
[23:08] mentioned at the top like how do you make sure your developers, your vibe cutting an app, it doesn't just like shove your pet token in there and now you're now you've just circumvented all the governance. We'll probably need a way to like make sure that apps don't embed developer credentials that we that we stop that
[23:21] from happening. Vulnerability scanning, again, if you haven't done like you know, set up your CI systems for this like we want to have some basic level of like checks on the packages that get that get included in apps so
[23:35] that we can check if there's any vulnerable ones that get included. Um and finally, we're thinking about like a built-in security proxy. So, many of you are probably using like a laugh or some kind of proxy like this. We'd like to have at least a few built-in protections in place so that like out of
[23:48] the box we've checked for like you know, very obvious like SQL injection risks risks and like XSRF and things like that. >>
[24:00] Next one. Yeah, so interoperability. So, um one thing I do want to uh clarify is we we talked about these three changes that we're introducing um scale to zero, app spaces, and Genie app
[24:13] builder. We're really thinking about this as a way to enable a new set of folks, what we're calling these citizen developers, to build apps. And so you'll when you set up an app space,
[24:25] the only way to create apps at launch will be through the app builder. We are going to add support for Git so that you can bring your own tools to develop in the app space. And the app spaces apps will all have this scale to zero functionality powered
[24:37] by this new like micro VM architecture. Um but at least at launch time that will be separate from um some of the new features we've offered for pro developers like horizontal scaling or these very large and dedicated instance
[24:49] sizes. We do want to reconcile these over time. So, like in the fullness of time I do expect that um uh you know, you'll be able to have all of the sort of compute functionality within an app space and that your standalone apps can scale to zero, but just want to
[25:02] set the expectations on that. Um, and but yes, we are planning to support Git. So, if you do need to migrate between these app types, that will be a way to to do that.
[25:20] There we go. Okay. So, um, timeline. We were hoping to have some version of this ready for you all to try by Summit. We didn't quite make it. We know we need to get this right. Um, the whole point of this is to make sure that you can govern apps, to make sure that these are secure. Um, it's not our goal to have people deploy lots of
[25:33] insecure apps. That would that would defeat the point. Um, so we know we need to get this right. We are hoping to have a version for you all to use um, end of this month, most likely early next month, um, in private preview and to go to beta
[25:46] and public preview shortly after that. Um, I do want to say the reason we're sharing this early is because we know a lot of folks that we've talked to are like starting to build out other solutions. And of course that's great. Like we encourage you to do what you
[25:58] need to help support apps. Like that's great. We want to help you with that. But some of these things I think you do need support from the platform from Databricks to do. Like, you know, having scale to zero apps, um, to have the
[26:10] ability of like an app space that can give direct resource access to apps. That's very hard, we know, for our customers to build ourselves. So, we want to give you we want to tell you that this is coming and and and plan for this.
[26:22] Um, and I do want to note like when we do go to beta, um, in the first version we probably will be limited on the regions we offer this in because we are offering this new compute type. It'll take us a bit of time to get that and it'll take us a bit of time to get support for all of the like security and
[26:35] networking rules that we have. So, thank you for your patience and for staying tuned on that one. So, yeah. Just want to say, um, talked to many folks, uh, maybe in this room about some of the challenges. We think
[26:48] it's very exciting what people can build with apps, but we hear you that uh, the last 6 months have been uh, maybe uncomfortably exciting with everything that's getting created and deployed and we want to help you come on this journey
[27:00] and help really take advantage of that. So, thank you for thank you for betting on humans with us. Appreciate it. >>

Learn more about the Databricks Data and AI platform.

The information provided herein is for general informational purposes only and may not reflect the most current product capabilities or configurations.