00:41:01 · Guillaume Moubeche, Ron Gabrisko
[00:41:01] S1: Can you can you explain like... Yeah. Sorry. Can can you explain like why you think like you're gonna need the new type of databases with with AI? Like, what does that change from a conceptual perspective? [00:41:16] S0: Yeah. The biggest part is we separate storage and compute. So similar to what we did in the analytics space with the Lakehouse. Right? Like, if you have to expand storage and compute at the same time, it makes your Databricks super rigid. Right? And so a lot of these AI and agent based applications, they wanna be able to spin up a database in, you know, milliseconds and then shut it down. So you don't wanna have to pay for it when it's not running. Right? Because, you know, think about it, like in the future, you have millions of agents hitting these, these applications. Right? It's gonna be, you know, today applications, you know, are for humans. There's so many humans that are using it. Now it's gonna be agents. It's gonna be 10 x, a 100 x. And so the difference of separating compute and storage is super, super important. And so, you know, if you think about our Lakehouse concept, where it's like take all your, you know, it was all your data for your data lake and your data warehouse, put in an open format so you can do all your analytics on one set of data. We're gonna do the same thing with Lakebase for databases. Right? And so, I think it's a huge transformational shift in the database market, and that's what's needed for AI and agents.
00:42:34 · Guillaume Moubeche, Ron Gabrisko
[00:42:34] S1: So Yeah. That... That's really cool. And do you also feel... Because I know, like, a lot of... I mean, some software company basically leverage Databricks in the back ends, and they use it, you know, for their customers also. Yep. So in the future, you can think per what you say that agents are gonna also sometime be using, you know, this kind of interface. And do you feel like there is a way for Databricks at some point to kind of, like, also charge the agents and not the user in some way? [00:43:12] S0: Yeah. I mean, again, like agents, because, you know, we have Databricks apps too, right, which is similar to kinda, you know, vibe coding, but, like, there's literally tens of thousands of Databricks apps out there already. Agents for us, again, it's consumption based. So agents, again, are running queries, asking questions, building pipelines, doing ETL, building models. You know, it's for us, it's the same as a as a human in terms of how they use consumption, and so we love agents. You know? [00:43:47] S1: They're good for revenue. [00:43:49] S0: They're good for revenue, and they automate a lot of things. Right? Like, they're they increase productivity for our customers, and like I said, I think every person in an enterprise or in a startup or in a D and B company should be using Databricks, and so that's a big reason why we created Genie and AgentBricks, So
00:43:49 · Ron Gabrisko, Guillaume Moubeche
[00:43:49] S0: They're good for revenue, and they automate a lot of things. Right? Like, they're they increase productivity for our customers, and like I said, I think every person in an enterprise or in a startup or in a D and B company should be using Databricks, and so that's a big reason why we created Genie and AgentBricks, So [00:44:09] S1: Nice. And and you said you've you've basically replaced a lot of the apps you were using. What what are, like, the the apps you've replaced and where now you feel like you couldn't go back to the old apps per what you just built?
00:44:27 · Ron Gabrisko
[00:44:27] S0: Yeah. I mean, just to give you example of, like, we call it our GTM hub. You know, we still store our data in Salesforce, but, you know, how our reps and managers and even me interact with it, like, I can get pretty much infinite access to whatever data I'm looking for. And so, you know, I have a dashboard. It measures all my forecasts. I can measure those forecasts, you know, by region, by customer, does data science models on that, it tells me which customers would churn. It also tells me like, hey, it recommends next best action, like for AE that has a retail customer, might said, hey, you you sold them like, you know, some ecommerce, you know, functionality, but you should also be talking to them about, you know, next best action for their their retail store and things like that, promotion analysis, things like that. So, we do next best action, churn analysis, revenue predictions. I can see which products we should be trying to sell to which customers, you know, it's it's been all customized for what I need to see as my business. And so I think every company would want that. Right? Like, if I'm selling beverages, and we've sold this to a bunch of customers, they want their reps to be able to see, okay, which stores are consuming which beverages. They even have it, like, I'll take a picture of the shelf, and it'll make recommendations on how I should... Where I should put certain products. It'll make recommendations on promotions I should be running inside of that store. Like, it's your own little cockpit, you know, for your business. Another example, like rental cars, you know, they're like little franchises, and they're trying to run their business while also doing all their financials and all that stuff. And so you might be like, oh, my customer sat's going down. How come? I'd be like, oh, know, your your couple cars weren't cleaned, so go, you know, make sure they're cleaning the cars or, you know, you should be trying to upsell gasoline more, whatever. And again, all these dashboards real time predict... In the predictive and they have Genie attached to them where I can just ask questions in English or whatever your language is and get answers back and get, you know, predictions and graphs and whatever you need. Like, it's crazy.
00:47:02 · Guillaume Moubeche, Ron Gabrisko
[00:47:02] S1: Crazy. Yeah. That that's insane. And did did you had to to, I don't know, like, let go some data analyst or these kind of things? Because for me, I feel that this is a position, you know, that whenever I need typically, like, to ask question that's data related for many years, you know, I had to go through either analyst or scientists that would take time to analyze, like, all our data. They would do, like, very complex queries to make sure that they get everything right. To be honest, now, I think it's, as you said, you know, it's like, you can ask just in plain language and get replies. So how how did it affected, you know, like the the team? [00:47:48] S0: Yeah. No. I mean, listen, we're still growing super fast. Like, I'll, you know, I'm I'm hiring a significant number of people again this year. I think the biggest thing is like, it's more of increasing access. Right? So there's still gonna be a lot of complex things that you want humans... I mean, you want a human in the loop when necessary. Right? Okay. But increasing access to where, you know, the simple stuff is just right there at your fingertips and everyone can just ask those questions is like pretty game changing for companies. Right? And so I think productivity will go up significantly for every employee, not just the analysts. So...
00:47:48 · Ron Gabrisko, Guillaume Moubeche
[00:47:48] S0: Yeah. No. I mean, listen, we're still growing super fast. Like, I'll, you know, I'm I'm hiring a significant number of people again this year. I think the biggest thing is like, it's more of increasing access. Right? So there's still gonna be a lot of complex things that you want humans... I mean, you want a human in the loop when necessary. Right? Okay. But increasing access to where, you know, the simple stuff is just right there at your fingertips and everyone can just ask those questions is like pretty game changing for companies. Right? And so I think productivity will go up significantly for every employee, not just the analysts. So... [00:48:30] S1: Yeah. I agree. So far, so... Yeah. I agree. I think, like, everyone has been talking about, you know, like, AI replacing jobs, but from my point of view, everyone using AI is just, like, working more because they are more productive and can do more stuff. So... Yeah. [00:48:44] S0: Yeah. I saw, like, I read an article, was like, they thought like a lot of the automation on, you know, radiology was going to, you know, replace radiologists, but radiologists actually went up, it just meant more people were actually getting tested, Right? And so I think more people will use data now to make decisions because they have access to it, and that will increase better decision making, increase productivity, and so think that's where the market will go.