Talk Description

The rapid rise of AI has dramatically elevated the value and strategic importance of data, transforming how upstream software engineers perceive and interact with data workflows. In this expert-led panel, industry leaders will share their experiences and insights into effectively bridging the gap between data teams and software engineers. They will discuss practical strategies for proactively managing data infrastructure, enhancing collaboration, and ensuring high-quality data to support advanced AI-driven development initiatives.

Additional Shift Left Data Conference Talks

Shifting Left with Data DevOps (recording link)

  • Chad Sanderson - Co-Founder & CEO - Gable.ai

Shifting From Reactive to Proactive at Glassdoor (recording link)

  • Zakariah Siyaji - Engineering Manager - Glassdoor

Data Contracts in the Real World, the Adevinta Spain Implementation (recording link)

  • Sergio Couto Catoira - Senior Data Engineer - Adevinta Spain

Panel: State of the Data And AI Market (recording link)

  • Apoorva Pandhi - Managing Director - Zetta Venture Partners
  • Matt Turck - Managing Director - FirstMark
  • Chris Riccomini - General Partner - Materialized View Capital
  • Chad Sanderson (Moderator)

Wayfair’s Multi-year Data Mesh Journey (recording link)

  • Nachiket Mehta - Former Head of Data and Analytics Eng - Wayfair
  • Piyush Tiwari - Senior Manager of Engineering - Wayfair

Automating Data Quality via Shift Left for Real-Time Web Data Feeds at Industrial Scale (recording link)

  • Sarah McKenna - CEO - Sequentum

Panel: Shift Left Across the Data Lifecycle—Data Contracts, Transformations, Observability, and Catalogs (recording link)

  • Barr Moses - Co-Founder & CEO - Monte Carlo
  • Tristan Handy - CEO & Founder - dbt Labs
  • Prukalpa Sankar - Co-Founder & CEO - Atlan
  • Chad Sanderson (Moderator)

Shift Left with Apache Iceberg Data Products to Power AI (recording link)

  • Andrew Madson - Founder - Insights x Design

The Rise of the Data-Conscious Software Engineer: Bridging the Data-Software Gap (recording link)

  • Mark Freeman - Tech Lead - Gable.ai

Building a Scalable Data Foundation in Health Tech (recording link)

  • Anna Swigart - Director, Data Engineering - Helix

Shifting Left in Banking: Enhancing Machine Learning Models through Proactive Data Quality (recording link)

  • Abhi Ghosh - Head of Data Observability - Capital One

Panel: How AI Is Shifting Data Infrastructure Left (recording link)

  • Joe Reis - Founder - Nerd Herd Education (Co-author of Fundamental of Data Engineering)
  • Vin Vashishta - CEO - V Squared AI (Author of From Data to Profit)
  • Carly Taylor - Field CTO, Gaming - Databricks
  • Chad Sanderson (Moderator)

Transcript

*Note: Video transcribed via AI voice to text; There may be inconsistencies.

" This is, this is the behind the scenes view at the, uh, at the, the Gable operating Center. So you got, you got the view of the, here, I'll show you.

This is the second you, you, you got this view earlier. Yes. Now, now you're getting the door. So congratulations sitting behind the scenes. There's, there's panelists around here. I see Joe, I see Vin. I see Carly.

Oh, nice. So they're, they're around. They just didn't want to come to the stage and help me out of that. Joe, you were just waiting in the background. Is that what you were doing here? I'm just letting you dig yourself deeper in that one. It's fine. Flounder panel. It's fine. Appreciate that. You sounded great.

You sounded great. Demetri just hoarding. What are whatcha talking about? I was, we're putting that through some chat tea. Alright, here, I'll let, that's Im saying you guys take it. Classic. All right. Well, you know, I think that's a, that's a fitting, that's a fitting way to end us off. I feel. Uh, this is, I think this is gonna be the, the casual session.

I hope so. Um, I'm, I'm super excited for this one. Well, Joe's pretty hardcore and uptight. Shut up. No. Yeah. True. Um, well, well may maybe we'll just start with introductions and then I've got a few questions, but we can probably just riff a bit and do whatever you guys are interested in, honestly. So maybe, I don't know, Carly, do you wanna start with the intro?

Sure. Yeah. Are we live right now? Uh, we are live. Oh, nice. That's awesome. Hey everyone, my name is Carly Taylor. I am a field CTO at Databricks as of Monday. That's crazy. I'm already killing it. It's been five days. Good. You get your performance. You I worked in gaming. Yeah. Okay. I worked in gaming for the last five years on Call of Duty, so I guess I'm here to hold down the gaming data and AI corner of the internet.

Cool. Uh, VIN. I'm Vin Vish. It's middle of the afternoons and I'm old, so I, I have no idea where I am right now. We're starting off a casual session, the right place. It's actually alert for him right now. Um, founder of V Squared we're one of the older sides, speaking of old, older data and AI focused consulting companies.

Started out in technology almost 30 years ago again. Wow. I'm really kinda hammering this old theme on myself, aren't I? Started out, I can see your hair graying in real time as we speak. Started out thinking I was gonna graduate from college to do AI and, um, well, let's just say nothing was ready for that.

So did software engineering, software development, very different type of gaming. Casino gaming. Started seeing BI analytics come into focus in about 2012. So started my business, wrote a book, helped clients make a ton of money with data analytics, machine learning, pivoted to strategy. It's been a fun ride.

Joe, you wanna round us out? Hey, what's up? Uh, I'm Joe. Good to see everybody. Thanks Joe. Hi, Joe. Okay. Um, okay, well, uh, subject of the day obviously is, is shifting left. Um, I think we've spent the last, uh, five or six hours hearing from a lot of folks who are sort of shifting left in practice. We talked about data contracts, talked about data engineering, shifting left, talked about DevOps.

Um, maybe we'll just start with a simple one and see where it takes us, but, you know, what does sort of shifting left mean to the folks on this call? And, uh, what do you, how do you see it sort of changing or, or improving sort of the slice of the data pie that you work with? You want to answer first? I'll go first.

Whatever. Um, shift shifting left is an interesting thing. I think to me, it, it means, um, the obvious answer would be that it it's about involving gear upstream stakeholders in, uh, the downstream, uh, impact that they may have, right? But I think it's, I think it's more than that. There, there's obviously a technological angle to it, which, um, Gable's working on.

And, and I think that there's also, um, there's an empathy component, which I don't think it's enough press either. And so that's something where I've, you know, I know having worked as a software engineer and a, a data person and other roles where, um, uh, to me shifting left, it's, it's, it's a, it's a vibe man.

Uh, but it really is, um, it's, it's a way of working with people, right? And, um, I think, um, understanding each other's needs and to me that, that, um, if you can have that plus the technological aspect, you, you've, you do have a true shift left. So curious to deal this year, think. I like that approach. I think that there, I hadn't thought of it as empathy.

Um, I was thinking of it more like, I don't know if we're allowed to curse, like shit rolls downhill, right? And so like if you shift left, you're kind of solving some of the problems that stop that from happening to begin with. Uh, but maybe Joe's answer's a bit more eloquent than that. I like the idea of it mainly because I have been on the receiving end of, you know, bad data and no data ownership for so long.

Um, specifically in gaming, you know, we think about everything on in games runs on data, right? And the data producers, this is often not an easy question to answer because you have so many hands in the game development. Um, when I was working on Call of Duty, I think in any one time we had 3000 engineers committing code.

And so you never really know who made anything that you're using and you have no idea if it's believable or reliable. And I don't think that there's a way to really work within those confines, right? And unless you understand your data producers and you're working with them, um, everything else is just, you know, gonna start falling apart.

I don't know. Vin, what do you think about it? You waited till I took that sip, didn't you? That was nice. Yeah, I did. I like that. Was that, was that the, your geral that you, you drank? Yeah. I gotta, you know, wash it down with a little bit of coffee for the midday. I was hoping I said something so exciting. You were gonna spit your coffee out all over your computer.

Oh my God, that's outrageous. Oh, we missed it. We should have done that. I have a, I have really a kind of a total dig, different take on shifting left. I think shifting left is shifting towards value because we talk about a whole lot of technology. And so when you start talking about data, yeah, there's technology.

I mean, gay, well obviously there's technology involved in data, but we have to monetize it. We have to deliver something to the customer. We have to deliver something to the companies we work for, to stakeholders internally to C level. I mean, all of these people are kind of getting to the point where they're sick of just throwing cash at data, throwing cash at ai, and still having their technology organization be a cost center.

And increasingly we're looking at it, that's a cost center code becoming a cost center. Now data teams can't do the same thing. I mean, it's time to break the cycle. And I think that's what shifting left really is, is looking at this from a different perspective and shifting towards not just moving technology towards stakeholders, moving technology towards the people that are creating or owning the data, creating data generating processes, but really talking about shifting towards the value of it.

What's the point of gathering all this data if it doesn't do anything for us? You know, gaming's kind of a good example because you, you immediately see the value of the data. It, it's kind of a, you know what the value of data is? Why? Because you're using it immediately. It's like a race car, running a race, you know, did the thing you did last week improve the car?

Well, did you win or not? So there's a obvious feedback loop and we need to create more of those. So it's probably shifting down to the bottom line. Oh, you could say we are shifting up and down to the top and bottom line. Yeah, it's, um, it's a circle shift. It's, it, it, it's actually really interesting. So there was, there was a friend I was sort of talking to, um, last time I was, I was in Washington and we, we were discussing shift left.

And one of the thing, like I, I saw sort of light bulb go off in her head is, is we were speaking and she was saying, you know, you know, what's interesting is that for the last however many years, the kind of confines of my ownership in the data space has been, uh, you know, I've got a set of reports or I've, I've got this like BI tool that I own, and, um, I've got dashboards that go out on some cadence to my leadership teams and, and I'm responsible for that.

But if I were really to, to sort of frame how, where data is used that delivers the most value in the business, um, it's oftentimes not even within the scope of the, of the warehouse or the analytical database. And this person, this particular individual worked with like trains and things like that, and they were saying, you know, we've, we've got our, our, our big board that shows the status of all the trains and whether or not they're, they're on time and that it's, it's just data, right?

It's data kind of moving around and when there's issues with the data, it has a really, uh, a, a really massive direct impact on the business because you're, you're showing that a train is like not on time and it is, or, or, or the inverse and customers will get really mad at you. Um, but businesses haven't really thought about those types of things as data problems.

And then what you're saying sort of makes me feel that's like part of the shifting left is starting to recognize those as not just software problems, not not just codes or application problems, but, but it is, it is part of the data value chain as well. Definitely.

Cool. Awesome. Um, well, yeah, I mean, I I, I would love to hear from you all as well. Like, so, so we, we sort of talked a little bit about like what, what kind of shift left means, um, that, that's great. Where do we see AI starting to fit into that and how does AI potentially start to help across the three axes that you guys mentioned of like, or, or does it help, right, sort of driving empathy with various people in the organization.

Um, sort of the, the technological shifts, the, the ownership and accountability and, and driving value. Like does it, what, what sort of change does it make? I think the biggest change that I've seen so far is just, um, uh, in people's respective disciplines. I. Other disciplines are now a topic of conversation.

So, historically, if you're a software engineer, if you were a, a data person, if you are a machine learning engineer and so forth, a data scientist, uh, whatever. But, um, you're kind of stuck in your own domain working on your problems. And so as part of the, the siloing that we're, we're trying to, uh, uh, you know, knock down with the shift left movement and so forth, but, uh, you know, AI sort of blew the doors open where all of a sudden interest in ai, uh, was almost universal no matter what, uh, discipline you're in.

I think that that more than anything raised an awareness. Okay, so we need, we do need to think about, um, data. We do need to think about code when you think about a lot of things. Um, and so, you know, I, I, I, I, you know, I look, uh, three years ago, four years ago, and they're, uh, I know a lot of software engineers, not too many of 'em are interested in machine learning or AI right now.

Every one of 'em is magically an AI expert. It's crazy. Um, data people the same way and, and, uh, and vice versa. AI people are cognizant of, of software, uh, practices more so it's, it's kind of what I described as a convergence. So I've been writing about this for a number of years now, but it's what I, what I see happening is it's, it's sort of, you can call it shift left, or maybe it's just sort of, um, you know, this, this melding, uh, I think of a lot of the, um, uh, disciplines and, and things are starting to converge.

Um, and discussions are happening that if you were to ask me five years ago if they were happening, they'd say, there's no way, there's no way software engineers that care about data and ai, why would they? Uh, but I think AI has been the forcing function for a lot of, um, rethinking and, um, uh, new ways of, of working.

But it's still very early days. So we'll see. I mean, I've got an article I'm dropping this weekend where comparing basically where we are now to the information super highway, um, of the, uh, of the internet age in the nineties, and Carly still has her a OL account, so she probably still uses that. But, you know, gen Z is referring to the nineties as the late 19 hundreds hate.

That's pretty funny. So much Vin and I are just like, Jesus. Yeah, we're pretty old. Um, hold on. I, I got a shift lift. Depends. I didn't mean to ruin anyone's day. Yeah. Okay, so mold, great. Yeah, yeah, we are. What, what are you, what are y'all seeing in that? Uh, Carly and Ben, um, sorry, my dog just came in here. Get outta here.

Um, it's interesting for me in gaming, because AI has been a thing in gaming since the beginning of gaming, right? Mm-hmm. And now when we talk about it, we're talking about. AI as an academic, you know, as an academic term, as an umbrella term. But there's also in-game AI that's part of this academic discipline of ai.

And it's funny seeing the adoption of academic AI into gaming to drive in-game ai, which I guess is just an application of ai. It's very circular and confusing, and everyone just calls it ai. And I guess what's been interesting to me in having all these conversations is that we're kind of defining the scope of our work, and we're defining the world as it's unfolding.

Like it's literally building the airplane while it's in the air, right? Like the, you know, a year ago maybe, I don't know that anyone even had an acceptable definition of what AI was or that anyone had even all, you know, we'd all reached this conclusion as an umbrella term and people are using it differently.

And it's just been so weird to live through that now we all kind of have an understanding of what AI is, but now we're like, how do we use it? Well, how do we integrate together well in gaming companies? It's been interesting because, you know, like VIN and was saying like, data gaming companies generate a lot of data and you know, immediately if you're using it well or not, but what you're doing with that data is different.

It used to be, you know, you built an AI and it was pretty static based on some rules of things that happened. And it could get smarter if you built, you know, more if than functions. And, but it was, it was all, all hand coded. And now you're seeing, you're getting pushes from players on what they want out of their game.

You're having to respond to that in real time. You're seeing in real time what people are actually buying, where they're spending their money. Um, as Rick Klein talks a lot about how attention is the new, like economy, like attention is worth money. AI drives attention. So now AI is worth money. 'cause it drives attention, right?

And gaming is very, at the center of that attention economy. You're fighting, you know, you think of a video game and you're not just competing with other games, you're competing with Netflix, you're competing with TikTok, you're competing with all these other things that are competing for people's attention.

Um, and you can't do that effectively unless you're leveraging data, right? Because the tiktoks of the world understand this. They have their algorithms dialed in to maximize attention. And if you're not doing the same thing, you will lose. So how do you do that effectively? And I think everyone's realizing that, right?

Even LinkedIn's realizing that they're forcing videos on us, which is strange 'cause I already spend time there. They don't have to, they don't have to gimmick it. I'm already there. But I guess they do for some people. So, I don't know. I, I guess it's seeing this integration into the wider society that's been interesting to me and how we commoditize it.

I mean, looking at ai, what has AI changed? I mean, I can charge more AI's made things more expensive. Uh, I, I think the hype's good because. We needed more attention on a lot of these issues. If there's money flowing into something, attention flows into it, you have to do it right. But I think that's part of what we've forgotten is, okay, so we've done the hype part, now we have to deliver product.

And if you look at some industries, you look at gaming, yeah, they've delivered product. You know, the AI either works or you're getting shredded on social media because your game's terrible. You know, the game experience is trash. Again, immediate feedback loop in a lot of situations like retail. If your AI's terrible, no one uses it.

And you never know. You just never find out. You know, it's, it's kind of, okay, we released this AI feature, what happened? Crickets. Okay, is that good? Is that bad? So there's no feedback mechanism. So one of the things that AI hasn't changed, I guess not answering the, answering the question by not answering it, is that we haven't brought forward the accountability piece of this.

But now shareholders are throwing CEOs out because, oh, you know, we don't trust your AI roadmap. We don't trust your data products and your data plan, sorry, you lose your chair. Thanks for coming by. Uh, enjoy your parachute. And, and that's, we've had more CEO departures last year and in the beginning of this year than we have, I think since 2009 or 2002 or something wild like that.

This is the reality. CEOs are losing their jobs when they lose their jobs. All of a sudden they're looking at their data team, their technical team, and they're saying, you are gonna lose your jobs before I do. How about I put it that way? So now we have these expectations. So we need to figure it out. We need to figure out how to, uh, retail companies, working with retail clients right now to literally change the website into more of a game experience where it is building things more dynamically and more in real time, responding to an individual's needs, and try to figure out how to use AI to do things that we've never been able to do online before.

So a lot of what shifting left means is if you don't have the data about your customer, well, good luck. You can't make an AI experience magically appear with no data. And that's slowly permeating into the CEO mindset. No data, no ai. I'm sorry, I, I like, I can't code ai. Nice try. It doesn't work that way.

You can't train data with vaporware either. The infrastructure has to work. You can't deliver insights that are just a couple of data points on a dashboard that doesn't work either. Insights are information, they're contextual. There's all of these pieces that. We've never had to confront before because the expectations haven't been there.

Now suddenly we've all been slammed into expectation territory, and this is the find out phase where a lot of companies just can't, you know, if you can't get it together, your CEO is toast. But unfortunately, the person that's going first is probably who's in charge of your data team or your technical team.

So it's a, it really is a partnership. I mean, we either figure it out together or we're all getting fired. I preferred the fuck around phase, man. I'm not ready. Oh yeah, that was fun. That was great. Yeah. Well, I think it depends too, what type of company and you're at and what game you're playing, right? I wrote about this nine months ago where I described data teams as either existing in enterprise land or product land.

And AI is gonna be a lot easier to get value out if you're, um, in product land, right? Because the end game, no pun intended to Carly, but it, you know, that that is the product. A game is a product, right? Or Uber, for example, or any of these other apps. I mean, the data is using AI or machine learning to do its thing, right?

So, but as been points out without data you're lost. But Enterprise Land's different. Enterprise land is where I think the biggest challenges remain because data has always been seen as a cost center. It hasn't been adding the quote value, um, directly to the bottom line. This is why there's always confusion about ROI attribution and so forth and enterprise land.

And so I, I think that it definitely is worth making the distinction between the two, um, paradigms. 'cause they couldn't be more different. But what the, the trouble is you're seeing enterprise land trying to catch up now and, you know, integrate AI into, uh, the business, um, which we'll see. It may work it, it may not, I think it, um, but it's gonna require a, um, you know, a different mindset again, sort of shift left attitude, um, like we've been talking about.

So yeah, it, it, it, it's, it, it feels, it feels a lot to me, like when I talk to data leaders and data teams, that they've sort of been corralled either by force or, um, sort of of their own volition into a very kind of small, you know, corner of the, of, of sort of what systems they own and, and what they're accountable for.

Um, I experienced this personally when I, when I was sort of running the, the platform team, um, at Convoy, and it was sort of made pretty clear to me that, you know, the, the primary value that I was gonna add was, was reducing the spend of our snowflake bill. Um, and like when that is your primary focus, right?

That, that that's not, that's not actually adding value, right? If you're, if you've got multiple people sitting around trying to optimize your compute, it's like, okay, cool, you know, you can, you cut off a, a few hundred k in the bill, but meanwhile there's some engineering team or some sales team or some marketing team or some data science team or whatever that like at the end of the quarter says, Hey, I just made you $5 million.

Um, you, you know, you, you can't compare to that. And like you said, VIN, I think that a lot of these executives, once they're starting to get pressure from their bosses on, Hey, you need to make these investments into AI and you need to make the number go up into the right, and they start looking at where their biggest inefficiencies are, they're gonna start with the folks who, who don't have that clear connection to a dollar sign in a bank account.

And, uh, and I, and I feel that, um, that the, that it doesn't need to be that way. You know, like, like I was saying a second ago, I, I think almost every company has unbelievably valuable use cases for data, but data teams oftentimes don't insert themselves into those conversations, and so therefore they get no credit.

You know, like if you've built an API and that API is unbelievably useful and customers are paying for it or whatever, um, that seems to me to be a data system, right? Like, you are consuming data, there are things that could go wrong with it. You have data quality rules and expectations, um, and there's a whole host of issues, uh, a whole host of things that I think data teams can say like, Hey, here's where I can sort of help make this better.

But because they're kind of in their own little corner, um, oftentimes relegated to bi, they just, they don't, they don't get the chance to, to do that. And I think to your point is like, that may have to change soon.

Okay. Cool. Um, alright. So, uh, I think we've, how much more time do we have, I dunno, seven, seven minutes? Uh, may, maybe in the last few minutes here. Um, I would love to get like, what, what's, what's, like the big data hot take. So thinking out next few years, um, what, what is the thing that, if you could say, the industry right now is thinking about some problem the wrong way.

We're, we're going down the, the wrong path, um, and here's what we should be doing instead, or sort of common wisdom or whatever it is, is sort of saying one thing. What, what, what is, what is the hot take that folks maybe should walk away from this conference with.

I think my hot take right now is prepare for a lot more gaslighting, um, with ai. Um, if, if you think the, the hype's been hot, now wait until we get into, uh, I think, uh, give agents about another give year or so, uh, to, um, reach a crescendo. But, but so to me, the dichotomy right now is, um, there's a couple things.

Uh, we're going to be, supposedly software is gonna be flipped on its head where it, everything is basically a prompt, right? Um, at the same time we're, we're talking about things like data contracts and, and standards. And, and I'm wondering, and this is actually somebody, somebody brought this up in my discord group, but like this, this is something I've been thinking about the dichotomy of the world that we're in and they throw vibe coding on top of this, but yeah, we're so speed, um, you know, creating standards and um, you know, building robust systems.

And so I think we're, we're actually, we're at a very interesting, um, I would say philosophical, uh, inflection point in the industry where on one hand it's vibes, on the other hand it's about discipline and rigor. Um, and so, uh, I, um, yeah, I'm very curious to see where this goes. The two warring forces, well, could they be, or could, is there a situation where these work together as well?

Right. And so that's, that's one thing I've been noodling on. Um, yeah. So anyway, that's not necessarily a hot take, but a uh, um, maybe hot garbage. Yeah. So anyway, um, my hot take is kind of short and I think anyone familiar with me is not gonna be surprised. I write about this a lot. I think that we are doing security with AI the wrong way.

Um, and I have yet to see anyone really set forth a comprehensive set of best practices for what it looks like right now and what it needs to look like in the future. Because I foresee the same issue that happened with gaming. We have a scaling problem with ai, right? Like there will come a day where, and these questions are already being asked, why can't we just do inference on the user's hardware?

'cause it's expensive to do it in a data center. It's expensive to do it server side and then push predictions down to the user. So just let their GPU do the hard work, right? But GPUs are not secure. We haven't really had a good discussion about what AI chip security looks like. And I mentioned gaming learned this the hard way because we learned this with anti cheat.

Anyone who knows anything about multiplayer gaming knows that there is a problem with integrity. Users systems and chain of custody and being able to trust anything that comes back from a user's personal device. And I think we haven't even begun to have that conversation yet, even though I know it's costing gaming companies millions of dollars.

And it's scary because it's a very, very real problem. And the solution is gonna seem real shiny and nice once people realize, Hey, everyone's phone, everyone's computer has a capable chip. We can just use their stuff instead of our own. And they're not gonna think through the consequences. Yeah, that's, I mean, kind of tangential to that is my hot take, which is ai.

You can't just keep throwing hardware at this. We, we've already sort of reached a point where hardware's at a ridiculous level, if we keep throwing hardware at the problem, we're never gonna solve our real problem, which is optimization. We have to optimize ai. We have to see more small models, more sort of the age agentic paradigm, which is multiple small models working together with guess what?

Apps and APIs and data sources. You're not going to just completely replace everything with AI immediately. Will that happen later? Yeah, but right now, unit economics, I mean, you can't fight unit economics. It's too expensive. So we're not gonna be replacing a whole lot of apps and code with AI immediately.

You're going to be doing the things that you couldn't do before with ai. Your other bigger, ch bigger challenge when it comes to optimization is data is a cost center and it will never be anything aside from a cost center. Information on the other hand is a revenue generator. Information's different.

It's contextual. We're in the BI paradigm way of thinking, which is these data points on a spreadsheet or you know, on a report are somehow valuable. But that's not how AI works. It doesn't train on data points. It learns patterns, it needs information, it needs that context. And if it doesn't have context, that data's useless when it comes to ai.

So data itself is a massive cost center unless you can do something with it. And those two go hand in hand. We have to optimize and we have to change the way that we manage our data in order to do a lot of that optimization. You can only get so far with the math and with the GPUs and everything else, you're going to have to make it so that the data requirements go down.

And that means a whole lot of transforming data into information. But that, how do you vibe with that though? I think it's a strong, like data's dead vibe, but a really, really happy information is the future vibe.

Well I think that's a, that's a great thing to end on. Just good vibes that all around, you know, we're vibing, we're vibing to the death of data engineering. Yeah. We're vibing to the death of data. That's, and that's vibing to the, uh, to the life of ai, uh, replacing all our jobs. So, yep. Um, awesome. Well guys, thank you.

Thank you so much for this and, and coming in the middle of your day and potentially the end of some of your days to, uh, to be a part of the first shift left data. Of course. We really appreciate it. Yeah, thanks. Great. Great conference. Great to see everybody. This panel panel right here was a vibe, I'll tell you that much.

I was, you're a vibe, man. That's sweater. Look at that. That's cool. Oh yeah, that's a, that is a huge vibe right there. That's amazing.

So we've got to finish this. You all gotta get off this stage. We're done. Yeah. We'll shift left by, I mean, you can hang out, but, hi, I won't mind. , however, I'm about to end the live stream.

We gotta say goodnight."