Enterprise Reporting and Embedded Analytics with AI/BI Dashboards

Summary

AI/BI Dashboards on the Databricks Data and AI platform consolidate enterprise reporting, embedded analytics, and self-service access into one governed layer without requiring separate BI infrastructure or additional licenses.
Three distribution modes—scheduled delivery, embedded dashboards, and direct self-service access—allow organizations to push trusted insights to Slack, Microsoft Teams, or customer-facing applications while Unity Catalog enforces row-level security per viewer.
Rivian migrated a multi-domain dashboard estate to AI/BI Dashboards in six months, illustrating how large-scale reporting consolidation is achievable on the platform.

Enterprise Reporting and Embedded Analytics with AI/BI Dashboards

Modern BI isn't just about ad hoc exploration—it's about delivering trusted data to internal teams and external users, wherever they work. Yet many organizations still rely on fragmented tools, leading to inconsistent metrics, duplicated data, and added cost.
AI/BI Dashboards bring enterprise reporting and embedded analytics directly into the Databricks Platform—with no separate BI infrastructure and no licenses. From scheduled delivery to Slack and Teams to embedded dashboards and Genie experiences, teams can deliver governed insights at scale.
We'll dive into how to design, deploy, and scale reporting and embedded analytics using AI/BI Dashboards and Genie. Topics include:

Production-ready reports with scheduling, parameterization, and consistent metrics
Embedding dashboards and Genie into internal tools and customer-facing apps
Governance, security, and performance at scale
Architectural patterns to scale reporting and embedding without duplicating data, logic, or cost

Talk By: Maggie Li, Associate Product Manager, Databricks ; Raymond Guo, Software Engineer, Databricks ;

Chapters

FAQs

What are AI/BI Dashboards on Databricks?

AI/BI Dashboards are a built-in business intelligence capability in the Databricks Data and AI platform that supports production reporting, embedded analytics, and self-service data access. They eliminate the need for a separate BI infrastructure by combining scheduling, parameterization, and governed distribution in a single platform.

How do AI/BI Dashboards deliver scheduled reports?

Scheduled reporting offloads query execution to low-traffic time windows so dashboards arrive in users' inboxes—or in Slack and Microsoft Teams channels—on a set cadence. Per-viewer filtering can be applied so each recipient sees only the data they are authorized to view, enforced by Unity Catalog identity.

Can AI/BI Dashboards be embedded in external applications?

Yes, AI/BI Dashboards and Genie experiences can be embedded into internal tools and customer-facing applications using the platform's embedding capabilities. Unity Catalog enforces row-level security so each viewer's identity controls exactly what data is returned, regardless of where the dashboard is hosted.

What is Genie 1 in the context of AI/BI?

Genie 1 is an augmented AI assistant within the Databricks Data and AI platform that goes beyond answering questions to producing the artifacts and reports a team needs. It enables natural-language queries over governed data and is designed to work alongside AI/BI Dashboards to support enterprise distribution of insights.

Full transcript

[00:08] Hi everyone. My name is Maggie and I'm one of the product managers working on AI BI and today I'm joined by Raymond, one of our senior engineers. Both of us have worked on the distribution and reporting of dashboards on the AI BI side and our goal today is
[00:20] to share with you what's possible in the space and also the new things the team has shipped over the past year. We've got a pretty big group today, so Raymond and I ask that you leave your questions until the very end. That way
[00:33] we can make sure we get through everything that we have lined up today and we'll have dedicated time at the very end for questions. Zooming out a bit, you know, we've introduced a lot of new products during
[00:45] the keynotes over the past two days, but today we're really going to focus in on two areas. The first is AI BI. This is the dashboards, the reporting side. How do you get trusted data in front of
[00:58] the right people on the right schedule securely? And the second one is Genie 1. This is your augmented coworker that goes beyond just answering questions, but actually producing the artifacts and reports that
[01:10] your team needs. Before we head into the distribution of dashboards, let's talk about dashboards for a second. You know, we're full steam ahead on giving you everything you'd
[01:22] expect from an enterprise level BI tool. These are a few of our more recent releases and a few I would like to call out are more out of the box visualizations like Gantt chart and more
[01:34] interactivity like faceting and parameterized text widgets. Also, a lot of improvements to the look and feel of our dashboards via themes. Everything from controlling the font size, the font themselves and how
[01:46] rounded corners are. We know that the look and feel of dashboards is crucial for downstream reporting.
[01:58] Great. So, you have these dashboards in place. The question is then, how do you distribute these insights to your entire organization and beyond? As we've worked with customers, we've seen enterprise reporting really break
[02:12] down into three areas. Scheduled reporting, embedded dashboards, and self-service and direct access. I'm not going to spend too much time here because we're going to dig into
[02:24] each one of these one by one in detail. So, let's get into it. Starting with scheduled reporting. You know, these are the reports that show up in your inbox. It's your Monday
[02:37] morning sales review. It's your quarterly performance reports. Raymond's going to walk through exactly how to set one of these up. But before he does that, there's a few benefits I would like to call out. The first one is
[02:49] performance. With a schedule, you can offload the query run time to whatever works best. Say your whole team logs in at 9:00 a.m. You can set up a schedule that runs at 8:00 a.m. and take that load off the system. All of that work
[03:02] will be done ahead of time, so when everyone shows up and starts utilizing the dashboards, the dashboards are ready to go. The second benefit I would like to call out is governance. Schedules are secure, they're governed
[03:15] under Unity Catalog, and they respect whatever permissions you have already set up there. There's nothing extra to manage, and because everything is governed, each viewer sees the data they're allowed to
[03:27] see. Let's say you have a single sales dashboard with a few different viewer groups. Each manager will only see the data that's relevant to them and their own specific region, while the executives
[03:39] get a high-level view across the whole business. Exactly Everyone gets exactly the slice of data they're supposed to see, all driven by governance permissions they have set up. So, to recap,
[03:51] schedules are secure, they're customizable per viewer, and they come with some real performance benefits. Now, let's talk about some of the new releases in this area.
[04:07] Previously, dashboard delivery was email only, but we kept hearing feedback like, "I don't actually live in my inbox. I live in Slack." Or things like, "My team is in Microsoft Teams all day, and that's where I actually work." So, we wanted to meet you where you are. Now,
[04:20] you can send dashboards straight into your Slack and Microsoft Teams channels on a schedule that you set, right where your team is already having those types of discussions.
[04:32] Dashboard schedules have always sent a PNG, a PDF, and a link back to the dashboard by default. And those are great when you want a snapshot of the data. But, if you wanted to pull in a table
[04:45] and do downstream analysis, an image doesn't quite get you there. You need the data in tabular format. So, available today, you can attach dashboard widgets as CSV, TSV, and Excel
[04:58] right into the schedule. You can also pick a subset of dashboard pages to send out. If only pages one and two are relevant, that's all that goes out. Now, I will hand it off to Raymond to give you guys a quick demo.
[05:15] >> All right. So, for the demo today, I'll go through some of the common flows that you may go through when setting up uh reporting for your AI BI dashboards. Um so, we have this dashboard here. It's tracking the revenues and metrics of a travel booking company called Wander
[05:28] Bricks. Um and it's a pretty nice-looking dashboard. We got an executive view page uh that gives us the top-line metrics. We also have another page showing us further insights. It's also got a different a lot of different
[05:40] visualizations that we support in AI BI dashboards. And then, the final page gives us a table for the raw data of the bookings details. I'll go ahead to the first page. Let's
[05:52] just say that I'm only interested in the completed bookings, so I'll go ahead and select that in the drop down. And then let's also say we're only interested in bookings of summer getaways. So you can see the dashboard gets
[06:04] automatically updated with the filter data, exactly what we're looking for. So what I did here was I specified global filters. These are filters that apply to all the pages of the dashboard. Uh you can also specify page local
[06:16] filters that uh that are restricted to just that page as well. But I'll just leave it as it is. And let's just say that we want to send this dashboard out as a report now. So in the top right corner, we have the
[06:28] schedule button and this is where you can add schedules. We can go ahead and add a weekly schedule, so I'll have it run every week on Monday uh let's say 8:00 a.m.
[06:43] Okay. And then under the advanced settings, we can also do a lot more configuration. So I'll go ahead and title this schedule weekly report. This is just what gets shown in the UI so that when you create multiple schedules, you know what schedule does what.
[06:55] You can also specify SQL warehouse override. And this is very useful when you want to use a specific warehouse for scheduled workloads in a separate warehouse for when users are interacting with your dashboard uh directly. Um if you don't specify one here, it'll just
[07:08] default to the the warehouse that the dashboard is currently using. So I'll just leave it as it is. Here we have a check box for applying the currently selected filters. I'm going to go ahead and select it. And you
[07:21] can confirm here that whenever the schedule runs, it'll refresh and snapshot the dashboard with the the completed status on uh bookings of summer getaways. All right.
[07:33] You can also specify a custom email or notification subject. By default, it just contains the uh the the dashboard title. I'll leave it as it is. And then the subscribers field is where you can choose the list of recipients that you
[07:45] want to receive the notification or the report. So, I'll go ahead and subscribe myself. And when I say myself, it's really just receiving it in my inbox. Um but we'll also support Slack. So,
[07:57] I'll go ahead and select a Slack channel that I set up with me and my teammates. Um one thing I want to call out is that we do also support email distribution lists. These can be configured as notification destinations by a workspace
[08:10] admin. All right. Let's customize the attachments. So, we have um this include pages. And so, this is where we can choose which dashboard pages we want to include in the snapshot that we send out.
[08:23] Um for the purpose of this demo, I'll go ahead and just select the first page. You can always choose to send out all the pages or any other subset. But, I think the top line metrics are the most important here. And then in case my recipients want to
[08:37] maybe open up the data in Excel and do further analysis, I'll go ahead and attach the uh the bookings details page on that last page. Bookings details widget on the last page uh as a CSV to the email.
[08:53] All right, that's pretty much it. The schedule has been created. So, it will run every Monday at 8:00 a.m. And when it does, it does two main things. The first thing is that it refreshes the query result cache of all the queries that are backing the widgets and visualizations in this dashboard. And as
[09:06] Maggie said, what this essentially means is that whenever a recipient or anybody pulls up this dashboard after the schedule runs, it'll load very quickly because it's pulling the results directly from cache. But, it also sends out the notifications
[09:18] to the recipients. So, I subscribe myself. In my inbox, I'm receiving an email similar to this. We see a snapshot of the dashboard of the executive view. Uh I want to call out that we the
[09:30] dashboard is filtered as we specified. Um we also get a PDF of the same snapshot as well. And then, just like I selected, we also get a CSV of the raw data.
[09:45] So, that's in my email, but I also subscribe Slack. So, in Slack, we're receiving something like this. So, it includes this both snapshots, both in the image as well as the PDF.
[09:57] Um and we always have a link to go back to the dashboard to further interact with it if we need. So, that's how you set up a a basic schedule in the dashboard, but I also want to show how dashboard reporting respects Unity Catalog and data
[10:10] governance. So, I took the previous dashboard and I made a copy, and I made a few changes. Uh the first thing is that I'm publishing this dashboard now to run with individual data permissions. And what this means is that each viewer
[10:22] of the published dashboard uses their own data permissions configured in Unity Catalog to run the queries. And we can see that in the published view. The second uh change I made was adding these two widgets. The first widget
[10:35] shows me who the dashboard queries is running as. So, since I'm viewing the dashboard, it's showing that I'm running the queries. And then the second one I added a company widget showing me which travel company the data in this dashboard is
[10:47] showing. So, since I'm viewing the dashboard, uh I only see the metrics related to Raymond's travel company. And so, this is all because the table configure the table that's backing this dashboard is configured in Unity Catalog
[11:00] with row-level security so that the viewers can only see the data that is relevant to them. All right. So, I already created a schedule on this dashboard, and you can see that I
[11:12] subscribe both uh myself and Maggie. So, very similar to the schedule that I just created, it refreshes the query result cache for both of us. Uh it's important to call out here since we're running the dashboard as with individual data permissions, it's
[11:24] actually running the dashboard as me and as Maggie since we get different sets of results here. But on the notification side, we also receive the snapshots as well, but the data view is different. Um I only have
[11:37] access to the metrics for Raymond's travel company, so this is what I see in my snapshot in my email. But then in Maggie's email, she's saying seeing metrics for Maggie's travel company.
[11:49] And so as I said before, this is all in force in Unity Catalog. But what this really means is that as the dashboard author, when you set up these schedules, you don't have to worry about the wrong data getting to the wrong recipients. You can just re- use the same dashboard here. And you can actually use it to
[12:03] send out reports with data that's personalized to each recipient. Okay, so that's how you can create dashboard schedules directly in dashboards. I'll give it back to Maggie to talk about how you can do it in other ways.
[12:15] >> Awesome. Thank you Raymond for that. Sorry. >> You're good. So that's scheduled delivery. But let's say you don't want a fixed schedule at all. You want the delivery to fire every time an underlying table updates, so
[12:28] that people are always getting the freshest data the moment the data lands. That's possible today with Lake Flow jobs. You can trigger a dashboard delivery off your pipeline, so the moment your tables refresh, your dashboard goes out. And the dashboard
[12:41] task itself got more powerful. Um you can now pass parameters and filters, which wasn't possible before. So you're not just setting one static view, but rather a tailored view for
[12:53] each data delivery. And Raymond's going to walk through exactly how that works. >> All right, so all the features that I showed before, you can do directly from the dashboard, but we also expose these dashboard refresh and reporting capabilities in the Databricks Jobs
[13:06] platform, like Maggie just said. And here you can take advantage of the full suite of features that we have available in Databricks Jobs. So I'm just going to call out a few jobs features that you can use for your dashboard reporting. The first one is
[13:19] jobs triggers. So let's say that you're in a similar scenario that Maggie just described, where you want to refresh a dashboard whenever new data lands in Unity Catalog. So, you can do that by adding a trigger here. Uh Jobs supports a lot of different
[13:32] kinds of triggers. Uh the first one, scheduled, is very similar to what I just set up uh directly in the dashboard, where you can configure it to run daily or weekly. But, the one I want to focus one on is here. This is a table update. You can
[13:45] add a list of Unity Catalog tables, and then Jobs will uh monitor those those tables, and whenever the table updates with new data, it'll trigger the job, and then uh the dashboard will be refreshed.
[13:57] Another thing you can do is run and refresh these dashboards based on a condition. So, I added the dashboard as a task here as the final step, right after a trend analysis. So, in the middle of my data ingestion workflow, I'm doing some sort of trend analysis.
[14:10] Based on the output of that analysis, I can choose whether or not I want to refresh that dashboard and send it out. And then, the final feature I want to call out is Jobs notifications. So, with Jobs notifications, you can
[14:22] configure uh notifications on the job level, so that you can notify somebody if any step in the data workflow fails. You can also notify somebody if the job starts, or if it completes with a success. So, that way you have a full
[14:36] overview of how your data workflow is doing, which includes the dashboard reporting step. And then, one small thing I want to call out within the dashboard task, we do now support uh specifying dashboard filters directly within Jobs. You can do that in
[14:50] two main ways. One is on the task level, so within the dashboard task, you can see that I'm overriding the booking dates parameter uh with this value. And then, the second way you can do it is through job level parameters. So, these
[15:02] are parameters that are specified on the side. These are parameter values that the data workflow pushes to every single task, and so that way all the tasks share the same parameter values. So, in this case, I'm pushing down the
[15:14] completed status to all of these tasks, not just the dashboard task. Okay, so this is how you can add dashboard reporting directly into your Databricks jobs. I'll give it back to
[15:26] Maggie to talk about embedding. >> Awesome. Let's dive into dashboard embedding. Embedding lets you take an AI BI dashboard and place it directly inside
[15:39] your application. So insights show up with the context where work is already happening. Say your operations team uses an internal custom-made portal. You can surface the relevant dashboard right
[15:51] inside that tool. So the data sits alongside everything else they're already working with, fully interactive. So there's two ways to do this, depending on who your viewers are. The
[16:03] first one is internal embedding, which is the example I just gave, dropping a dashboard into your own internal tool, since your viewers are already Databricks users, or they can be provisioned in. And governance carries
[16:15] straight over from Unity Catalog, so everyone only sees data they're meant to see. It's a simple iframe setup, and Genie comes out of the box, so people can ask questions right there, too.
[16:27] The second is external embedding. This one's for reaching your end customers, people who aren't provisioned into your Databricks account, and here you can set up your own application, control your control what the viewer sees by passing
[16:39] in parameters, rather than via UC. So each person still gets their own data. If you want that Sorry, it's uh >> Oh, sorry.
[16:54] >> And if you want the same conversational layer, Genie's available through the API. Now, let's see these embedded dashboards in action. >> Okay, so dashboard embedding demo. So as Maggie said, we support two main types of embedding. The first one is
[17:07] internally for your organization, uh within your workspace, maybe. Um and you all we also support externally embedding for those who don't have a Databricks account. I'll touch briefly on basic embedding here. I took the previous dashboard that I showed you and I
[17:20] embedded it within the Databricks Confluence page. And so this is configured as an iframe. The experience here is I log into Confluence, I'm logged into Databricks, and then I have the full interactivity of the dashboard available to me as if I
[17:33] were opening up the dashboard directly in Databricks. So we can do things such as like, you know, flip through the pages, uh we can filter down the data, uh do some cross filtering.
[17:48] And then zooming into specific visualizations is also possible as well. And then as Maggie mentioned, we also out of the box embed the Genie agent that's tied to the dashboard. So here, this is where you can ask uh
[18:00] questions about the data that's backing the dashboard in natural language to Genie, and then you can directly get those questions answered without having to ping anybody. Okay, so that's basic embedding.
[18:12] How about if you want to share a dashboard with people who do not have a Databricks account? And that's where external embedding comes in. This is a feature that we released in the past year as GA. And with external embedding, we're sharing we're delegating the
[18:25] authentication to an application that you control. So here I have a web application web application that I created and deployed. And so through external embedding, what what's
[18:38] essentially happening is that we created a service principal for your application and then configured it in Databricks to perform an OAuth handshake with us so that the dashboard can be displayed after a user logs in to your application.
[18:50] So in this demo, I created this app. Both Maggie and I are users on this app, but in this scenario, neither of us have access to Databricks. So we don't have we're not in the Databricks workspace, we don't have direct access to these dashboards.
[19:03] I'm just going to go ahead and log in to the app. All right. And then once I log in, it's a basic one. I just embedded the dashboard within it. But you can see
[19:16] that the data that the dashboard is is displaying is for Raymond's Travel Company. I'm going to go ahead and sign out and then I'll sign in as Maggie now.
[19:31] So once Maggie's signed in, she sees the same dashboard. It's the same embedded dashboard, but she sees a different view. She's seeing metrics for Maggie's Travel Company. And this is very similar to the example that I showed before where the dashboard is running with individual data
[19:43] permissions, but the difference here is that night neither Maggie nor I are Databricks users. So but we are still able to get a different view of the data based on the viewer. So the way that this works is that we're actually passing in a parameter value
[19:56] from the application to the Databricks dashboard. So I'll show you behind the scenes. This is the data set query that's backing the embedded dashboard here. It's kind of small. I'll zoom in a little bit. At the very bottom you'll
[20:09] see that we have this filter uh where we're filtering the company ID based on this AIBI external value parameter. And so this is a query parameter that we can use in the embedded dashboard and based on that
[20:21] parameter value, you can choose what data to show to each of your recipients or each of the application users. And you have a lot of flexibility here. In this scenario, Maggie's account maps to value two, my account maps to value one,
[20:35] so we get different views. If you want multiple users in your application to get the same view of the dashboard when they log in, then you can just map them to the same AIBI external value. Okay, so those are the two types of
[20:48] embedding that we support here. Uh I'll give it back to Maggie to talk about Genie. >> Awesome. Thank you, Raymond. So far we've covered two ways of getting dashboards to people, scheduling them
[21:01] out, and embedding them directly into your own apps. The last area I want to cover is self-service with Genie. You know, dashboards get consumed by all kinds of users in your organization.
[21:14] Generally, a large group of those users are business users. Those business users don't necessarily need or even want the full suite of tools available in the workspace in the Databricks workspace. And Genie 1 is
[21:27] built for them exactly. In Genie 1, you can ask questions about your data in natural language. You can consume your dashboards right alongside with the full same interactivity you'd get in the workspace. Drill through,
[21:40] cross-filtering, all of it. And for a consumer or account-level user, Genie 1 is their default and only view, a simpler home base and entry point. Beyond just viewing dashboards, Genie 1
[21:54] can now help you do the reporting in natural language. Genie can run the analysis for you, and it can notify you to your commonly asked questions. So, for a business user, that's things like the daily briefing or
[22:06] the weekly digest, and it can even create these documents for you. It can pull from underlying data, skills that you create, documents, and an ever-growing set of MCP connectors to
[22:18] ground those answers. I think this one in particular is better shown than spoken about, so Raymond, take it away. >> All right, so Genie 1. Let me go ahead and open it up.
[22:35] Okay, so this is the Genie 1 homepage. It's a very simple UI. It's meant for the data consumer, so nothing too complicated here. Um I'm just going to very quickly walk it through walk through it. On the discover tab, you'll see uh that we have dashboards, agents, apps.
[22:49] So, Genie 1 is the one place to see all of your organization's dashboards and data applications all in one place. So, these are all assets that users in my workspace have created for
[23:01] consumption. And you can see that certain assets are certified by the organization and you can open them up directly in Genie 1. So this is all meant for the consumer, but the one thing I want to focus on is
[23:13] the ability to chat with Genie. So this is one of the most important features. This is why we make it the default home page. And you can answer your data questions here with just natural language. And just to showcase that, I'm going to open up a previous conversation
[23:26] I had with Genie. >>
[23:38] revenue, and then the average booking value. And I also tell it to not include the canceled bookings. You can look into the thought process of Genie here. It uses Genie ontology to search through all the knowledge in the
[23:51] Databricks workspace, and it finds the key table that can answer my question right here. It executes some SQL and then eventually gives me back this report. So this report is exactly what I'm
[24:03] looking for. It's giving me these three metrics for each of the years, and then it's also giving me a list of the key takeaways. So I really like this report. I want it to send to me on a recurring basis. So this is where we're stepping away a
[24:15] little bit from dashboards. I tell it to turn this report into a recurring notification. Under the thought process here, it asked me a few follow-up questions. One of them was how often do I want to receive this notification, and I
[24:27] answered monthly. And it spits out this scheduled task. So you can create or you can see all these scheduled tasks under schedules on the left nav bar. And this is where we list all the scheduled tasks that have
[24:39] been created in Genie by you and it. So there's two main ways that you can create a scheduled task. One is creating directly through conversation with Genie, just like I showed you.
[24:51] And so basically, whenever you mention that you want to run something on a schedule or you want something recurring, then Genie knows that it should create a scheduled task for you. You can also edit these scheduled tasks directly through conversation as well.
[25:04] But the other way is, if you don't want to chat, you can just create it manually as well. Here you can specify the title of the scheduled task, a custom set of instructions and prompts that you want Genie to run on the schedule,
[25:16] the connections to the external MCPs, and this is useful if you want to extract data from outside of Databricks, or you want to perform actions outside of Databricks on a schedule. And you can just specify when and how often you want to run this task on.
[25:32] All right, I'm going to open up the scheduled task that I just created with Genie. You can see that it's running on the first day of every month at 9:00 a.m. And then the instructions are already pre-populated based on my conversation
[25:45] with Genie. Now, depending on how the reporting goes and how these, you know, tasks run, I can always go back and edit the instructions and and tune it. Um and I can always run the task to test out how
[25:57] it goes. We also have a run history here, so this is where you can dive into all the past executions of the scheduled task. You can dig into the thought process of Genie, ask any follow-up questions. Um
[26:10] these are all available in the run history. And then for the connectors, I didn't add any here. It was a pretty simple example, but if, for example, I want to update what this report is doing, then I can always come back here and add a
[26:22] connector, and then it will get picked up in the subsequent runs. So, outside of just this scheduled running, uh we also have the capability here to notify. So, based on how you configure the
[26:35] prompt, Genie decides if it it should notify you. Sometimes you don't want it to notify you, sometimes you do. You can evaluate maybe a condition, and then Genie can decide whether it should notify you from there. In this scenario, I explicitly told it that I wanted to
[26:48] receive >>
[27:00] This is all custom wrote by Genie and as you can see it's a report that's very similar to the one that it generated in the interactive chat. Now, every single time the schedule task runs, it runs with the latest data. So,
[27:13] uh as I receive more and more of these reports and notifications, we should expect to see that these numbers update, of course, in 2026. But, the overall report should stay mostly the same because it's using the
[27:26] same template and the same prompts on the schedule. Okay, so that's how you can set up reporting in Genie 1. We're really excited to share it with you so that your consumers can self-service their own data needs and then self-service uh
[27:39] and and sorry, uh and set up fully customized reports directly tailored to them. Okay, I'll give it back to Maggie to talk about uh what's next.
[27:51] >> Awesome. Thank you, Raymond, for that demo. You know, we're always building. So, before I wrap up, I want to give you a quick look as to what's coming. On the reporting side, there's more robust delivery methods, larger file
[28:04] handling support, and much, much, much more landing in Genie 1. And then, on the dashboard side, there's a lot in flight here, too. I won't walk through all of it, but a
[28:16] couple I'd like to call out are dynamic annotation lines and drill down, plus a few new like tree maps and bullet charts. And of course, there's a lot more beyond what's on this slide.
[28:29] To close, I want to leave you with one customer who's put all of this into practice. Rivian came on as a design partner and have migrated a massive multi-domain estate of dashboards onto AIBI in under
[28:43] 6 months. And they have helped us shape 58 new features along the way. Thank you so much for spending time with us today. >>

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.