Getting the Most out of AI in BI: From Dashboard Creation to Intelligent Insights

Summary

This video shows how Genie Code accelerates dashboard creation and maintenance by building complete dashboards from natural language instructions, reducing the months-long BI backlog that teams face when responding to constant new requests.
Built-in AI features in AI/BI dashboards — including forecasting and key driver analysis — reveal trends and root causes without custom modeling, while Genie can be embedded directly inside published dashboards for conversational exploration.
Metric views in Unity Catalog serve as the governed semantic layer underlying AI/BI dashboards, ensuring that AI-generated answers and dashboard visuals draw from consistent, authoritative metric definitions.

Getting the Most out of AI in BI: From Dashboard Creation to Intelligent Insights

AI is rapidly reshaping business intelligence but for many teams, it’s still unclear how to move beyond isolated features to real, end-to-end impact.
In Databricks, AI is embedded across the entire BI workflow. From using Genie Code to build and maintain dashboards, to applying AI-driven capabilities like forecasting and key driver analysis, to enabling conversational exploration directly within dashboards, AI becomes part of how analytics is created, delivered, and used.
This shift allows teams to move faster, reduce manual work, and unlock deeper insights without adding complexity.
In this video, we’ll walk through how to apply AI across the full BI lifecycle:

How to use Genie Code to accelerate dashboard creation and maintenance
How to leverage built-in AI features such as forecasting and intelligent insights
How to embed Genie directly into dashboards for conversational exploration
Best practices for combining AI and BI to drive adoption and real business impact

Talk By: Alex Lichen, Sr. Product Manager, Databricks ; April Dai, Engineering Manager, Databricks ;

Chapters

FAQs

How does Genie Code help build AI/BI dashboards?

Genie Code can build a complete dashboard from scratch using natural language instructions, generating data queries, chart configurations, and layout automatically. This video demonstrates how it reduces manual dashboard authoring time and helps BI teams escape the cycle of constantly responding to new dashboard requests.

What AI features are built into Databricks AI/BI dashboards?

AI/BI dashboards include built-in capabilities such as forecasting and key driver analysis that surface trends and root causes without requiring custom model development. This video also covers how Genie can be embedded directly inside published dashboards so business users can ask follow-up questions in natural language without leaving the dashboard.

What is the Genie Ontology in Databricks?

The Genie Ontology is a context graph that gives Genie the business knowledge it needs to accurately answer data questions. It connects metric definitions, entity relationships, and domain context so that Genie's AI-driven answers align with how the organization defines its key data concepts.

How do metric views support governed BI and AI in Databricks?

Metric views in Unity Catalog act as a shared semantic layer that both AI/BI dashboards and Genie draw from, ensuring consistent metric definitions across all consumers. This video explains how to build metric views that support multi-fact data models and automatically propagate changes to downstream dashboards.

Full transcript

[00:08] All right. Hello everybody. Thank you all for joining today. I'm super excited to be here and you know, welcome to summit. Uh just kind of quick forewarning
[00:20] roadmap and kind of future facing investment themes. So those things may change, timelines may change and that's basically because of feedback from folks like yourself. So as you tell us what you want, we go and change what we build.
[00:34] Uh also please make sure to complete the survey. It does really help us kind of prioritize the content and make sure it gets better and better every year. Uh so you can do that on your uh Databricks event app and then just fill out the survey for the session afterwards.
[00:52] Awesome. So today we're here whoop, my bad. Um we're here to talk about getting the most out of AI and BI. So we're going to talk about our kind of dashboarding products. Uh I talk a bit about our AI and natural language products as part of that. Uh my name is Alex. I'm a product
[01:05] manager here. I've been here for 4 years focused on this kind of BI and AI space and I'm joined by April. She's an engineering manager with our team focused on those similar problems. All right, so let's get started.
[01:19] So what is the problem that we're trying to solve here with kind of BI and AI together? There's really kind of two twin challenges that we've seen a lot of BI teams starting to face. The first of these challenges is getting
[01:31] out of the cycle of constantly responding to new requests, spending a lot of time, you know, building dashboards and things to solve those requests and then getting another one and spending more time and you're kind of stuck in this months-long months-long
[01:45] kind of backlog of, you know, constant dashboards that you have to build. And then at the same time, we're also opening up, you know, capabilities uh like coding agents to a large part of
[01:57] our users. And those users are understandably very excited about these tools and the ability to go and create, you know, HTML dashboards and those sorts of things. Uh but often times those are very ungoverned uh and there's not an easy
[02:10] deployment process. And we don't really have visibility over what they're doing. So, how do we solve both of these challenges in today's era?
[02:22] Well, we really think that BR teams need to focus and kind of spend their time on the core data foundations. Right? So, that is great semantics in Unity Catalog. It is making sure you have the right definitions for the
[02:34] different problem spaces people have. You're still probably going to have some layer of dashboarding there, right? There are still going to be KPIs that you take a look at sort of day in day in out, week in week out, and you're still going to have dashboards for those
[02:47] things. But then that kind of top last section should be self-service. It's the long tail of questions a user might have. And you shouldn't need to spin up a new dashboard each time someone has a random question they want to answer.
[03:05] So, Databricks makes this process really great. Uh you know, with standalone BI, often times you're going to run into challenges trying to make this self-service world happen. And the challenge really comes from the kind of the disconnect between where your agent and your consumption layer
[03:17] lives and where your metadata and kind of your lineage and your history all live. So, with a unified stack, we can really help you simplify that and make sure your agents have great context kind of
[03:29] across the board, right? You ask a question, it understands, "Hey, this is the lineage behind the answer of that question." And in particular, today we'll be talking about two products that kind of
[03:41] help you solve this problem. So, the first is AI BI, that is our dashboarding product product. It's kind of your traditional BI reporting layer, visualizations, filtering, interactions, all of those things.
[03:53] And then the second part is Genie. And there's a lot of different parts of Genie, right? So, a few things in particular. The first is Genie 1. This is you know, you've probably seen at Keynote is an experience for business users to come
[04:06] in and just ask questions. Uh the second are Genie agents. These are agents that are more curated for a specific set of context. And then there's Genie code. And Genie code is designed for a bit more of a
[04:19] technical persona to come in and begin authoring assets like dashboards. So, we'll be showing some of these authoring flows and some of these consumption flows today.
[04:33] And all of this is also powered by the Genie ontology, right? So, this was something that we talked about at Keynote as well. Uh what is it? It's a context graph that Databricks builds for you that tries to understand across all your
[04:45] different kind of activity within Databricks and within external systems, what are common definitions of things. Um you know, who are authoritative sources that can say, "Hey, this is the right answer for what revenue should be
[04:58] defined at, what ARR should be defined at." And it really makes sure that Genie really makes sure that Genie has great accuracy when it's answering a question. It's not just kind of pulling random things from thin air.
[05:15] And in particular, there's a few products here to talk about today. So, the first we'll talk through is Genie 1. Uh Genie 1 is kind of that business user experience. Uh so, we'll show you what that looks like to consume. We'll show you how to use Genie on top
[05:28] of dashboards as well. So, you can kind of anchor your questions in a more like curated, you know, definitive context. And we'll also talk about Genie code. And Genie code is designed for a kind of
[05:41] that authoring experience, right? Uh Genie code can help you author dashboards using AI so all those visualizations interactivity. We can author now more advanced data models as well.
[05:54] And then you can also use MCB connections skills. You can upload screenshots and use this import BI command that lets you import contacts from like a Power BI or a tableau, right? And all of this helps you agentically
[06:07] create dashboards much more quickly and efficiently. Awesome. Well, that was a very quick intro. Most of this will be demoed so I'm going to pass it to April who will be showing you
[06:20] a bit of the authoring experience in Genie code. >> Cool. Thanks Alex. So let me start by showing how Genie code can help you build a dashboard from scratch using natural language.
[06:32] So let's say I'm a dashboard author who is responsible for understanding our New York City coffee shop sales performance. And also identifying the key drivers behind our products and shipments.
[06:44] So I've written down my goals and the requirements into a Google Doc and saved some screenshots of reports that I'd like to incorporate. Now I want to put everything together into one dashboard. So using Genie code, I can simply prompt
[06:58] it. Take a look at this Google Doc and attach screenshots. Help me create a new dashboard. Let me also attach screenshots here. >> Do you want to zoom in real quick? >> Okay. Click enter.
[07:11] >> Do you want to zoom in real quick? Sorry. Just so that >> Oh yeah. >> it shows up there. Yeah. >> Yes. So Genie code will first analyze and understand my intent and semantics of the prompt and look at my Google Doc and
[07:23] screenshots, then begin constructing the dashboard. Okay. Let me share. This is the Google Doc that I shared with Genie code. So I outlined my goals which to analyze
[07:35] the coffee shop performance across our three New York store locations. And also a list of the business requirements that I'd like the dashboard to help me answer. So, for example, I'd like to understand
[07:49] like how much are we selling across our different stores and categories. What's the week-over-week revenue growth rate? And where are our opportunities to improve our business?
[08:01] And at this point, the user can just stop right here and let Genie code take full creative controls. It can automatically determine the dashboard structure, pages, widgets, and also visualizations based on the requirements
[08:14] provided. However, if the user already has something specific in mind like I do, so more instructions could also be included. In this example, I provide a structure
[08:27] for the dashboard. I'd like it to include like three pages. They are sales summaries, transaction breakdown, and shipments. And to be even more specific, I can call out exact widgets and visualizations I'd
[08:41] like to see. Here, I request a Sankey chart to visualize the product sale flow. And a map to display the sales product units by store locations. And a mosaic charts,
[08:53] which going to show the relations between the stores and the subcategory shipments. And Genie code can also look at the screenshot that I've uploaded and use them as a additional context.
[09:08] And this is the screenshot that I shared. I shared with two. So, one is showing the top 10 New York coffee shops by sales. And the other one is showing the top 10 products has been sold.
[09:21] Cool. Now, let's get back to our dashboards. Uh so, let's go to the dashboard. Okay. So, we'll see that Genie code has now found the underlying data sets for my
[09:33] request. Let's click into this one data set, the coffee model. And you'll notice that it is a metric view. So, a metric view is a Unity Catalog asset that defines a business-friendly
[09:46] semantic layer with the measures, dimensions, and also relationships. Because it's stored in the Unity Catalog, it inherits a centralized management, governance, and access controls.
[09:58] It also provides a trusted definition of business metrics, given that all the agents are using the same source of truth in the data analysis. And this is incredibly useful as it
[10:11] helps agents to understand the business meanings behind the data. And you can see that metric views can also be materialized at different granularities. Frequently used aggregations can be
[10:23] precomputed. And this going to improve our query performance and also reduce latencies. And here, you can also see that uh Genie code can create several local dashboard
[10:35] for metric views. And those metric views are defined directly and are scoped only to this dashboard. And they allow the agent to model business metrics quickly without requiring the user to create and manage
[10:48] using metric views ahead of time. And users can also switch between the UI preview and the YAML file to inspect the source table as well as the semantic
[11:00] definition details. And Genie code also establishes the relationships between the data sets. So, this is a private preview feature that allows us to define the multi-fact
[11:13] data models. And in this example, you can see Genie code created one-to-many relationships between store locations, coffee model, and shipments. It also established several measurements
[11:26] for the cross data sets. And one thing you to mention that once all those widgets has been generated, Genie code runs a verification pass to validate that all
[11:38] the data and the visualizations have been created and working correctly. It also supports automatic error recoveries. If a tool call fails during the dashboard creation, the Genie code will
[11:50] identify the issue and try to fix it. Now, let's take a look at the pages the Genie code created. So, out of the box, you can see Genie code can generate a wide range of
[12:03] visualizations. From standardized chart and tables to more advanced visuals. And for example, here are the revenue metrics and also week over week trends that I asked for.
[12:16] And the scrolling more down, you can see that there's a heat map which shows quickly about the Saturday morning 8:00 a.m. is our busiest time, which apparently a lot
[12:28] of people like to drinking coffees on weekend mornings. And let's take another look at the shipments page. So, on this, you can see it also generates very beautiful Sankey diagrams
[12:40] that are exactly what I request. And additionally, Genie code also supports the custom visualizations. So, if the building charts aren't enough, the Genie code can generate
[12:53] completely customized visualizations. And they're going to generate the visualizations through the Vega-Lite specifications, just like this mosaic charts. This is currently also a private preview
[13:06] feature and gives user much more flexibility to build customized experience. And Genie code also created a global futures that I asked for.
[13:19] And let's look at this transaction breakdown page. It also automatically create several page level futures where it fields necessary. And more importantly, everything is connected and interactive.
[13:33] For example, if I click on this is Brussel drink in the bar chart, and that selection will automatically filters the rest of the page and let me drill into the performance specific to
[13:45] this product. And of course, AI doesn't actually replace manual editing. For quick changes, for example, like reading in the dashboard, it is often faster to edit
[13:57] directly. And finally, Genie code also support skills. And skills are mostly like reusable shortcuts for common tasks.
[14:09] Users can define their own skills or use the predefined ones. For example, here we support import BI. This is a skill that brings dashboard from Tableau or Power BI into Databricks seamlessly, so
[14:23] you can furthermore continue enhancing them with our AI-powered capabilities. So, this is looks pretty good. I'm going to just click publish
[14:35] the dashboard. Okay. So, let's go into the published dashboard. You can see Genie code give us a really great example for dashboard authors. Once the dashboard is published,
[14:48] consumers often wanted to go deeper. They may wanted to understand why this metric actually changed, want to explain the pattern, or ask a question the dashboard didn't directly answers.
[15:00] And here's where Genie comes in. Okay.
[15:12] The network is a little bit slow, sorry. Cuz with Genie, viewers can ask questions in natural language and get answers directly from the dashboard data.
[15:25] And Genie also helps user to get started with starter questions.
[15:42] Sorry, there are technical difficulties a little bit. Let's go through here and see if we can pull it up on this view instead.
[15:56] This is the risk of a live demoing AI features. They should try this New York City coffee shops.
[16:20] Here, I I've got I've got a different dashboard that looks very similar. Let me find that one. This is a different New York City
[16:33] Uh this one, New York City coffee shops. As you can see, we love New York City coffee shops. >>
[16:45] Why do you not have access to that one? Okay. Let's try like a Does this one work, do you think? >> Maybe. Yeah, let's see. >> Okay, there we go.
[17:00] >> Okay, cool. So, this one works. So, Genie code can help users get us started with starter questions by clicking ask Genie. And it provides the starter questions on dashboard level, which based on the overall context, or based on the widget
[17:14] level. It based on like the widgets the users exactly clicked in. And here, let me just randomly choose one chart.
[17:27] Here. And I can also right click on this chart and triggering the Genie to help me answer like what happened to this change. Yeah. And the Genie analyzed the
[17:39] underlying data and identify the key drivers behind the movements or anomaly. And in the same semantic data model that Genie code just created.
[17:51] And Genie also understand the dashboard context including the visualizations and also the applied filters if it have any. And for transparency, user can just
[18:03] click into our thinking process and can see the reasoning steps as well as any sequel execution details. And of course, Genie also respects the
[18:16] same dashboard permissions and access control. It only uses the data the viewer is authorized to access. Cool. Let's see if the Genie give me any
[18:28] answers. Ah, it usually takes a little bit more time for the thinking. >> Yeah, you can tell what time >> if we can have any
[18:40] anything in the history. Okay, cool. Yeah, here is the exact questions that we asked before. So, you can see that Genie give me a really clear explanation of what happened on this magic job.
[18:54] And from here, I can also ask the follow up questions to drill in further. And this is great for the consumers to ask questions about a dashboard. But sometimes we wanted to take follow up questions and actions.
[19:08] And that's where Genie 1 comes in. And I'll hand it back to Alex to share more. >> Awesome. Thanks, April.
[19:23] So, um April walked through a bit of the kind of authoring experience and the consumption side for anchoring on a particular dashboard as your context. Uh but oftentimes you're going to have a business user and they don't know exactly where to go to ask my question,
[19:35] right? Like they don't know what dashboards exist, they don't know what Genie spaces exist. Like how do they actually find the right place to go? So, Genie 1 is designed to be a really simplified landing experience for those users to just come in and get started,
[19:47] right? So, you know, continuing off of that same set of New York City uh coffee shop information, I could ask something like, you know, "How were my coffee sales last week in NYC?"
[20:01] And you'll see here, kind of in this thinking, Genie is going to look across a lot of different places. I've got the skill that's already loaded here, but it's also going to go ahead and ask kind of individual Genie spaces in this skill.
[20:14] It can also search across different dashboards that are I have access to. It can search across different tables that I have access to and pull in context from a lot of those different sources. And it does so respecting those permissions, by the way. So, if I'm a
[20:26] user and I only have access to these three tables and that dashboard and these two Genie agents, like that's the only context that I can pull in. I'm not going to find random other information out there. All right. So, it's going to think
[20:38] ahead. It's starting to create our little report here. Let's kind of wait and see the results. We can see it's kind of the thinking it's got. So, store location, total sales, you know, tracking 7 days.
[20:50] It's now going ahead and creating a document. So, a document is a actually like a text editable doc that I can then go ahead and share and edit. So, here you see my document. Let's say
[21:03] I want to change this to, you know, uh Ford Data and AI Summit, right? Uh I can go ahead and make those edits. And then I can go ahead and share it with other users in Genie 1.
[21:15] So, this is a great experience for kind of creating documents and, you know, asking questions. But, let's say I want to see this kind of every week, right? I actually have a Monday morning, uh, weekly report, uh,
[21:27] that I have to go to this meeting for and I need to like talk about how our sales are doing. So, we can do that by using a schedule task. I I'm going to jump ahead a little bit and show what that kind of looks like in a previous chat session from earlier today.
[21:40] But, I can say something like schedule this for delivery every Friday afternoon at 1:00 p.m. It's going to go ahead and create that schedule. And then in my schedules, I can go ahead and see that schedule I've created. And
[21:52] I can see the kind of different runs here. I think I just created this one, so unfortunately it doesn't have a run. Uh, but it will show you kind of the same results here. Uh, and you can also get like a regular, uh, email report. Uh, you can get a kind
[22:04] of a Slack notification or a Teams notification. So, it will reach you wherever you are as a user. The other thing that we can do is I can create a skill. Now, a skill is kind of a reusable piece
[22:16] of context that lets me, uh, make sure I kind of get the same result each time I ask a question. So, I can do create user skill here. And then just say, you know, create skill for above.
[22:28] Something like that. Uh, and it's going to go ahead and make sure that it understands the conversation that I just ran and repeat that each time. I actually went ahead and created a skill earlier, so you can see that skill
[22:40] here in this customizations tab. And this is an example of the skill, right? So, it's quite detailed. It can tell us exactly what the source of data to use are, what the steps to calculate are, what the output should look like,
[22:53] all of those kinds of details. And soon we're working on things like being able to share skills. That way you can have your entire team with the reusable set of prompts.
[23:06] Within Genie 1, we also support a variety of other features. So, something to talk through here is the discover page. This is a place to go ahead and browse different assets that I have access to. So, things like dashboards, uh Genie agents, apps.
[23:18] I can also certify them, by the way, in case you want that kind of stamp of approval. And then we also have agents. So, agents were previously called spaces for those of you who are familiar with that.
[23:30] Uh and agents are really designed to be a very curated set of specific context. So, what that means is that uh let's see here.
[23:43] Oh, not this one. Let me find the coffee shops one. Here we go. Okay.
[23:58] So, what that means is that I've set a particular scope of data for a particular agent. So, here I've got this coffee model that I really like. It's kind of a metric view that we've defined that has great semantics built into it. And then you can add additional
[24:10] instructions. For instance, you know, maybe your fiscal year is offset by a month or something like that. You can define that in the agent. And now every time I go and ask this agent a question, it has that context built in, so it's able to kind of
[24:23] maintain that expert knowledge on a particular narrow topic. And then the kind of final thing to talk through here is that we've added MCP support to Genie 1. So, if you have
[24:36] contacts that lives in something like a Google Drive, uh in a SharePoint, uh some other external system, you're able to bring that context in and really power like more contextual answers as part of Genie.
[24:52] All right. So, that's a quick demo of uh Genie 1 and some of the Genie code authoring capabilities. So, talk a little bit about where I'm headed and what's coming. So, on the dashboard side, there's kind
[25:05] of four key pillars that we're really focused on. The first is these AI elevated experiences. Uh this has a a couple different flavors. So, one is around authoring richer dashboards. All right, we want to
[25:17] really improve the complexity of what you can create. We're also looking at things like automatically surfacing insights on dashboards. So, if there's like something really interesting, if there's a spike in the data, you know, being able to show you that spike and
[25:29] understand why that happened. And then there's a lot of pieces around the just the user experience itself and how it looks and behaves. The second area is around the richness of the visualizations and interactivity.
[25:41] All right, so, you know, very classic uh things like adding tree maps and cohort visualizations and some of these more advanced visualization types. Uh we're looking at things like drill down as well, that's shipping soon. So, a lot more ability coming there.
[25:55] The third category is around enterprise grade reporting and consumption. Some of this stuff is like management features, like version history and being able to kind of go back in time to a previous version of the dashboard. Uh some of it is better usage monitoring
[26:08] and alert integration. And then we're also looking at ways that deliver these reports. So, uh we've got some CSV kind of email attachment capabilities that we've launched recently, but we're looking for kind of larger file sizes
[26:21] um and kind of SFTP OneDrive destinations. So, just more ways to kind of get data to end users where they need to be. And then finally, uh there's a huge uh and um roadmap item around performance. You
[26:34] probably saw some of the Radeon announcements yesterday around kind of that millisecond level BI serving. So, that's been a huge area of investment and it continues to be something we tune. But there's also things around like
[26:46] materialization, for example, where if you have a metric view, being able to say, "Hey, I want to materialize specific grains so that you get faster performance on top of those predefined aggregations."
[26:58] And then the final thing here is this AI-assisted data set deduplication kind of concept. Uh the the idea here is that you have a lot of sprawl semantically in a lot of your organizations. Duplicative definitions, you know, conflicting
[27:10] definitions, and you're trying to reconcile these things. How do you do that? Well, we're going to have AI kind of help you identify, "Hey, there are these duplicative definitions across your community catalog." What are the options that you can then
[27:22] have to resolve those things? So, that's something that we're looking forward to as well. Uh just to call out that kind of first piece in particular around the dashboard customization.
[27:34] We're looking to really go beyond the kind of simplistic layouts, right? Uh so, you know, what we're seeing is people using tools like a cursor or a cloud code or something like that, and you're able to create awesome-looking
[27:46] dashboards, right? Great-looking HTML files. But then like they come to you and they say, "Hey, okay, now how do I deploy this thing? How do I share it with a thousand users? How do I make sure it's live? How do I version control it, right?" And suddenly
[27:58] that becomes a really difficult thing to do. So, with these more flexible dashboards, we're looking to give you that flexibility where you can kind of vibe code your custom components and layouts and side panels and models and
[28:11] visualizations and all those sorts of things. But like they're still governed, right? Nobody's going to put their API keys accidentally into a file. Uh you get to manage and control that and see all that uh information and monitor it.
[28:29] Uh and then the final part of the road map to talk through is around Genie 1, the kind of business user experience there. So, four key pieces of uh focus there. The first is around context. So, this is all the different ways we pull in information to answer your query
[28:42] more effectively. That includes things like indexing, includes a lot of that anthology work that we've previously talked about. Um it includes things like MCV connections and Lake Flow connectors. Second is around just outcomes. This is
[28:55] the things that you can do in the experience, right? It's the things like documents, uh MCPs that can actually write to external systems. So, if you want to like file a Jira ticket uh or take some other action via Genie 1,
[29:07] you'll be able to do that. Uh and then a continuous focus on improving skills, some of those custom agents, scheduled tasks, all of those kinds of pieces. Uh one notable one to call out here as well is actually the branding piece that
[29:20] launched like last week. Uh so, that gives you an ability to like add custom colors, uh custom logo, some of these things to make it feel a little bit more like at home with your organization. Uh there's the governance piece.
[29:33] Uh this is about kind of monitoring and making sure you're understanding what questions people are asking, gauging the accuracy of that, you know, gauging how often people are asking these questions. And we're really trying to give you that control and visibility as an admin in an
[29:47] easy way. So, today there's a lot of audit logging that already supports these kinds of flows. But we're looking to give you kind of a more native UI capability for that. And then finally, the last area focus is around mobile.
[29:59] Uh and there's a mobile application that is available today. You can download it on your phone uh from the App Store, Google Google Play Store, uh Android Play Store. Actually, I don't know. Um and you know, use that interface to ask questions, get answers,
[30:13] view dashboards, all those same things that you see in the web version, you can do in the mobile version. And we're looking at things like push notifications. So, hey, if you get your Monday morning report comes in, it appears as a notification on your phone.
[30:25] Uh things like photo and file upload, so you can kind of be out in the world, take a picture, and then ask a question about it. So, a lot of focus on that kind of mobile area as well.
[30:38] Awesome. And that's all we have for presentation today. Uh we've got about 11 minutes left, so happy to answer any questions you all might have. >>

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.