The Data Stack Show - 254: Context is King: Building Intelligent AI Analytics Platforms with Paul Blankley of Zenlytic

Episode Date: July 23, 2025

This week on The Data Stack Show, John chats with Paul Blankley, Founder and CTO of Zenlytic, live from Denver! Paul and John discuss the rapid evolution of AI in business intelligence, highlighting how AI is transforming data analysis and decision-making. Paul also explores the potential of AI as an "employee" that can handle complex analytical tasks, from unstructured data processing to proactive monitoring. Key insights include the increasing capabilities of AI in symbolic tasks like coding, the importance of providing business context to AI models, and the future of BI tools that can flexibly interact with both structured and unstructured data. Paul emphasizes that the next generation of AI tools will move beyond traditional dashboards, offering more intelligent, context-aware insights that can help businesses make more informed decisions. It’s an exciting conversation you won’t want to miss.Highlights from this week’s conversation include:Welcoming Paul Back and Industry Changes (1:03)AI Model Progress and Superhuman Domains (2:01)AI as an Employee: Context and Capabilities (4:04)Model Selection and User Experience (7:37)AI as a McKinsey Consultant: Decision-Making (10:18)Structured vs. Unstructured Data Platforms (12:55)MCP Servers and the Future of BI Interfaces (16:00)Value of UI and Multimodal BI Experiences (18:38)Pitfalls of DIY Data Pipelines and Governance (22:14)Text-to-SQL, Semantic Layers, and Trust (28:10)Democratizing Semantic Models and Personalization (33:22)Inefficiency in Analytics and Analyst Workflows (35:07)Reasoning and Intelligence in Monitoring (37:20)Roadmap: Proactive AI by 2026 (39:53)Limitations of BI Incumbents, Future Outlooks and Parting Thoughts (41:15)The Data Stack Show is a weekly podcast powered by RudderStack, customer data infrastructure that enables you to deliver real-time customer event data everywhere it’s needed to power smarter decisions and better customer experiences. Each week, we’ll talk to data engineers, analysts, and data scientists about their experience around building and maintaining data infrastructure, delivering data and data products, and driving better outcomes across their businesses with data.RudderStack helps businesses make the most out of their customer data while ensuring data privacy and security. To learn more about RudderStack visit rudderstack.com.

Hi, I'm Eric Dotz. And I'm John Wessel. Welcome to the Data Stack Show. The Data Stack Show is a podcast where we talk about the technical, business, and human challenges involved in data work. Join our casual conversations with innovators and data professionals to learn about new data technologies and how data teams are run at top companies. Before we dig into today's episode, we want to give a huge thanks

to our presenting sponsor, Rudder Sack. They give us the equipment and time to do this show week in, week out, and provide you the valuable content. Rudder Sack provides customer data infrastructure and is used by the world's most innovative companies to collect, transform, and deliver their event data wherever it's needed, all in real time. You can learn more at ruddersack.com.

Oh, welcome back to the Data Stack Show. Yeah, excited to be here. We're here live from Denver. Yeah, from my house actually. So yeah, I got to catch up with Ben Rogajohn last weekend in person here in Denver and now we get to do this. Yeah, the Denver day, you know. Yeah, exactly.

Awesome, Malia, catch us up from when we last talked. Yeah, there's a lot of exciting things going on on the ZenLinx side. We've been getting a bunch of great new logos like J.Crew, Black and Decker, some of these just fantastic companies to work with. Yeah. And we've just been seeing AI just go gangbusters in terms of overall capabilities of the models. And it changes a lot of how generally this stuff needs to work in the future. And it just changes so fast. So it's like nothing I've ever seen in terms of rate of change of the industry overall.

Yeah. So I'm always curious. These two questions. One, is it going faster than you would have thought? And then like the follow up to that is like, what is something let's pick a six month or a year time from the you're like, wow, this is like, I did not expect this. So I think, I think definitely faster than I expected. And I see a lot of really bad takes where people are like, Oh, well, the model is only getting sort of incrementally better as like, you just use these things to realize that like the rate they're improving. So it's like definitely faster than anticipated. And I think it's also the domains in which they are getting dramatically better is what's maybe most interesting to me. So so it's also the domains in which they are getting dramatically better is what's maybe most interesting to me.

So it's like things that they continue to be sort of approximately human or like subhuman that are a lot of sort of mind vendors are just sort of understanding, communicating these kinds of things that humans do a lot. And then if you look at things that they are already superhuman at and increasingly getting more dramatically superhuman at it's like coding, like you can throw an AI agent and it will win or come in, you know, within in the top five, the best programmers in the entire world who have been training on doing these, you know, same with mathematics, same with with any symbolic sort of task. And the reason for that is that

you can generate a massive amount of training data on these symbolic tasks that you can verify or correct. So it's like you could say, hey, this test case needs to pass if this code is written correctly. And then you can just reinforce and learn that process at a just truly massive scale, way more so than you can sort of solve problems. So while the models have certainly improved in how they handle things in sort of softer domains, how they've improved in terms of code generation or math or physics or other hard or symbolic domains,

that has been maybe one of the most interesting things to me. It's also shaped a lot about how I think about the future. Cause I think within six to 12 months we're going to be superhuman in every respect in symbolic domains. So it's like there will be basically no human coder in 12 months that is a better programmer than a language follower. That doesn't mean that the human won't be writing better code than the language model, just like someone who has more business context, but is not as good at coding is going to do a way better job in terms of actual impact on

certain matters than a language model. But a lot of our job is thinking about how we build an AI employee is how do we give the AI the right context to work with to be able to use the stuff that is superhuman at to really help the human make decisions. That's so interesting. Another question along the AI models, what's your take on the long-running models? Because I've seen some stuff come out where people like these stats like, oh, these long-running models, they don't actually perform that well.

Then other people who are like, it's amazing, the longer it can run, the more it can do. Yeah. I think I'm mad. I think this is something that divides, like, you know, the more it could do. Yeah. No. So, so I think this is something that divides a lot of people, even they used it pretty often. So, so I love them. I think it's amazing. Like you, you just have to do a good job sort of pointing it at the research

problem and the method approach you want to take things like that. So for instance, if I wanted to go and do an analysis of like, Hey, what are the different data sources that people most often interact with sales force it's like I would have an idea of how I want to approach that problem and you know what other stuff fits with sales force in this way and I can go and tell the agent hey this is generally the approach I want you to take and it can go and look at a thousand different sites or a thousand different references and sort of aggregate all that up and then as I look at the report I

Can click and see the citations and maybe print anything I want to take deeper on so I find it to be incredibly powerful I know a lot of people though like just don't engage that in the right way Yeah, and really use it for like quick and dirty banks, right? So I think it's some of its difference in you should patterns as well Yeah, yeah, I've seen some deep research, seen really neat things. Manus, seen some really neat things. So yeah, the thing that I've wanted, and this actually brings us kind of into the BI territory, the thing that I've really wanted to see, to mention in your take, is the long-running analyst task where you're, you know, there's a lot of like, there's a spectrum here. There's like the text is equal essentially,

or the equivalent versus like, I have this really deep problem with data from like a bunch of different places and I want you to like solve the problem. I think with your kind of AI employee concept, you guys are really trying to span both. So I'm interested, I want to dig in a little bit on the harder problem and like, is like long,

I assume like the long running task is part of that, but like, what are some of the other components to solving for those like deep McKinsey or consultants? How? Yeah, I think a lot of it is you first got to give the model interface in which it can, which it can really cook, like what you can really work in and work well. That interface is still got to be governed. Most importantly, it's got to correspond to your actual business context. Because if you aren't understanding the environment you're working in, just like an employee who's very technically capable, but doesn't actually connect to the business,

it's not going to be that valuable at the end of the day, because it's just speaking a different language from what people are going to need to make actual decisions. I refer to that as the data science problem. Yeah, it really is. Because that's what kids companies 10, 15 years ago, they all hired data scientists. And then there is, honestly, I think

a lot of it was a communication gap. These remarkably smart PhD people with all the stats background and computer science background, and a lot of them struggling with the business value. Yeah. And I think with AI, if you don't do a good job or bring the business context, you get kind of the same thing, where you get these answers that are maybe generically good

or good on average, but not actually good for helping you make decision-side of your business specifically. So that's a lot of stuff that I think is really important. The other thing is picking the right kind of model for the job. So like right now we have a model picker, which is not something I'm very excited about. I would love to see us have something that's more of a delineation between kind of fast and slow. Where instead, but right now you have a model picker. So if you pick GBT 4.1, it's going to be really fast.

It's going to ask the question quickly. But if you wanted to go do a sort of comprehensive look of what's going on, it's not heavy. So do you find at this point in time, our customers, like you've ever had probably a subset of customers like really into like, oh, like this is my model. It's like a branding thing, almost like I like this kind of core. I like this kind of car, right? Even though maybe they're like fairly equivalent for the task is divine that the case should be fun? Most people are like, just tell me whatever works best. I don't care. So one of the things that was interesting

as soon as we launched the Model Picker, people loved to just go in and play around with it and find the kind of model that they liked best. And for different companies, because there is this trade-off between latency and sort of comprehensiveness. So GPT-4.1 is going to be really fast. Claude, Sonic 4, or like 03 are going to be incredible sort of comprehensiveness.

and slow or by default do I want it to be faster but maybe less comprehensive. Right. So one of the ways I like using Zoe to go and explore some data is that I'll just go and I'll say like, hey, maybe a comprehensive dashboard of everything that's going on and just like let it crank and it'll just crank and put all this stuff and put it together. And I've got an overall view of everything that's going on and can kind of dive into any areas that I want to focus on. Well, and I have that too, as we progress, there's going to be a mix here where if I'm sitting down, like, all right, I'm going to spawn five background tasks to go explore these things, those run, and then like while those run, I'm going to be running four dot one or something, and just like kind of exploring around. So you can kind of do both.

Oh, totally. I mean, not that you can already do that, but I forget that's a thing. I forget that you can actually kind of multi-thread this stuff. I just forget. Yeah, I know. It's easy to do, because it's as easy as you have opened a bunch of different tabs,

and you just start to find different friends in the different areas. And it's crazy the amount of stuff you can parallelize just to that right now. Yeah, I mean, just like a bunch of tabs, there is the whole focus thing that you have to be careful with because you do still have circumstances where deep thought is the right answer.

And that can definitely be a distraction. Yeah, absolutely. It also kind of plays into where I think we're going in the long run. So I think about us in the long run as an employee, basically, as another person. So it's like the quote is like, don't buy software, like hire talent. You just be thinking about AI agents is almost something that you're hiring. And I think about our place in that as effectively an AI McKinsey consultant.

That's especially good work from a data, knows your business context, and can immediately start launching into helping you answer some of these questions that actually matter. So a lot of the problem that I view that solving is, BI has existed for a long time, is primarily like a collection of facts.

And that's fine if you have someone who's extremely analytically minded and wants to like hop in and. Look at all the facts and kind of build those into a narrative and go with that. But a lot of the problem that analysts have had to solve, and a lot of this isn't really rocket science, you know, here referred to as like data literacy. It's like, how do you go from this collection of facts and turn that into a decision about which products you're promoting that paints, turn that into

decisions about when you're re- paints, turn that into decisions about when you're re-killing inventory for these different SKUs. That is one of the real fundamental problems that AI enables, that we're especially excited about Sol. I got to ask about the AI employee thing, because this has come up more than once. Do you think that in the future, companies will pay for training for their AI employees?

I think they have and they already do. Okay. So in what way? I'm imagining like we have an AI employee and we like send them to a conference or big conference. I know it's not going to be quite that way, but yeah, I think it's actually going to be called data engineering. I mean, it's going to be, it's going to be, because a lot of the job that you've got to think about is how do you broker

business context to a model? And a lot of that is like doing actual data modeling, data engineering, the way we think about it now. A lot of it's dealing similar work to data engineering, but applied to text. Cause like, how do you get the right text in the right spot where this model is being able to act on in a way that matters.

And now we have incredibly capable models that given the right context can perform a lot of rote tasks like really well and pretty reliably now. Right. So it's like the problem and the thing that we have to be very cognizant of is it's like you need basically context engineers who make sure the right stuff shows up at the right time. Okay so context engineers is that a job posting? Are people hiring? Not a job posting, guys.

Not yet, but I think they're still called business engineers. Within a year, it could be. Yeah. Yeah, it could be prompt engineering. Yeah. Man, that's fun. OK, so a couple other. So speaking of context, structured data,

kind of where we've been on structured data, we had some runs with some, you know, databases devoted to the back to your show. Do you think, well, two parts to this question. Do you think the existing structured data platforms are gonna actually be able to kind of retool to support when everybody needs out of an unstructured data storage?

Or do you think that's a whole new, like, greenfield thing where people are gonna have maybe a structured data of an unstructured data storage or anything. That's a whole new like greenfield thing where people are going to have maybe a structured data platform and unstructured and hook them both up maybe to it, to an AI or BI tool. Yeah. Great. Great question. I think it's, it's a hard question to answer because the water is going to be

really bloody, like you have kind of an enterprise search, which is previously down the sort of unstructured thing. You should have all the BI players, like legacy BI players doing like everything just on top of SQL. I think it's inevitable that in the long run, these two things come together. One of the things language models did is they changed what people think about when they think about data. So data used to mean like a table, a CSV, something in a database that, you know,

you have to ask somebody about data now also corresponds like a table, a CSV, something in a database that you have to ask somebody about. Data now also corresponds to contracts and internal knowledge docs and all this other stuff that is immensely valuable but previously wasn't thought about as data. And it's time for business intelligence to actually make good on the business intelligence knowledge of the SQL. Where contracts is such a good one. Because I can't tell you the number of companies

I've worked with, all of the companies I've ever worked for, have all sorts of valuable knowledge in DocuSign, essentially, and PDFs, where they, you know, they customize the contract for that enterprise deal and they keep doing it and they keep doing it. And nobody has any idea what the deals are. Somebody has to dig up the PDF, and the person left, and they have to go get into their,

you know, it's like a whole thing. And then there's an initiative like, hey, guys, get all this in Salesforce. So it'd be so interesting once like, all of that is accessible in the same way, probably even a better way than like a structured data would be. Yeah. Well, I think the other thing that you alluded to is that the part of the benefit and problem with BI the way it's done now is that it relies on very structured data that sits inside of a SQL warehouse that requires a person set up a connector, transform a bunch of data, clean it to the point that it actually corresponds to what the business is referring to. And that's great. That's all a lot of very important metrics and KPI's and how are you actually performing at a business

at this high sort of aggregate level. But it is just so much work that no one would ever bring all the PDFs of all their contact and you would need a 5T connect to the OcuSign. That probably doesn't exist. And if you don't have to, even if you don't have to write that yourself, you still got to like somehow manage all these PDF PS and that sort of engineering work. Some companies will go down that road and do it. Other ones it's they need a more plug and play away to get unstructured data

out of these systems and into effectively searching indexes. So the HPE server. Yeah. That's interesting. So kind of another topic I've been thinking about a lot are MCP servers. Because there's, so now as we're talking about unstructured data, structured data,

and then we've got MCP servers here, we're like, OK, that's an interesting interface. How do you think the MCP servers played into the BI space? I think MCP servers are actually really interesting. And maybe hot take is that an MCP server, if you are building one as the BI product, you are eventually going to get in place. And the hot take there is that BI systems aren't actually

systems of record. They don't have the data. They are just processing layers. And that processing layer can be shown as LickML code, that processing layer can be shown as SQL, depending on which the iPod you're using and how it's structured internally. But it's like if you're basically outsourcing your entire UI to Cloud or to OpenAI or MCP server, eventually you're going to be in trouble. Because you're just a processing layer. You don't have the actual data like Salesforce does. Or like NetSuite right. I think it's only a defensible position if you have proprietary data for

some reason. Like I know of a couple companies that have a unique proprietary source of data and them putting MCP on top of it. It's the same model they have now. They're charging for data via an API. Like, cool, put an MCP. Like if the a model. But if it's not like proprietary or unique in any way, I totally agree. Yeah. And I think it's like for how people work with enterprise data to succeed, it's like you do have to own the interface. And like, if you can't add enough value in your interface

and how people understand what you've done and the intrepid ability piece, how it integrates with dashboard, how it integrates with other parts of the system. It's like you'll kind of get heat. And it's like a lot you, you, the thing that people want, and this kind of goes back to your MCP questions, a lot of people will come and ask us like, Hey, can you be an MCP server? I want to build one agent. And the question to go back to your unstructured comment is like, well, why do you want to

do that? The reason people almost universally want to do that is because they want the one system that can talk to the PDFs over here, and that can talk to their structured data over here. And BI must evolve to do that. Otherwise, it's not gonna actually be bit intelligence. Right, yeah.

It's gonna be SQL intelligence. Right, right. Yeah, well, and the other thing, too, is MCPs are still just a layer over APIs. And we'll have these same problems that APIs have until a different abstraction is formed. Because I think people actually have too high of an expectation on MCPs when it's like, look, there's

a reason we put all this effort in centralizing these things in databases is because it's better for analytic workloads. It's not that we didn't know how to interface with APIs. We knew how to do that. We did this on purpose. Yeah, exactly. And that's what I mean about the value out of the UI.

A lot of people think you just magically throw something into an MCP server, and everything's great. But it's like, there is value in UI. The final UI is probably not one universal chat system that literally everything. It's like, oh, cursor is very successful. Cursor is not the same thing as chat GBT. It has UI that is really beneficial in coding.

Yeah, and one of the, from this Inletiq product, and fairly unique, actually, the ability to switch back and forth, I think, is going to be more commonplace, at least I hope, where you can start with a GUI element of a chart and say, hey, I want to essentially talk to this chart or dashboard and know more about this thing. Or I can start from chat and then, hey, I want to just say what you're looking to learn more about and create

charts and graphs. I think the ability to go back and forth is a big deal from user experience. Totally. And I think that's something that Cursor got really well. Something Andrei Karpadei was talking about in his recent demo day talk, where it's like a lot of AI applications that succeed

are going to find this right balance when giving the model flexibility to do what it wants by where you can kind of control the autonomy of the agent. And I think that's a pretty good way to think about it, where it's like if you bins and limit, right? If you want to be in full control, like agent is not really involved. You can be in the UI clicking on things and same way you did before.

You can also ask Zoe to go and pull a piece of data for you. It's like barely agenting at all. And then it's like, you could also ask it to do a deep research on what's the tariff impact on your gross margin going to be of all these different skewers. It's like take this CSV tariffs on raw materials and blow it through to my final margin impacts. And that's a really big ass and a big ass they need to do. So it's like there's this slider of all harmony, our own product right now. And I'd expect that for products that do well, they master the kind of transitions between that.

So it's really easy and feels really natural to go from, I'm kind of running everything to the agents running some stuff to the agents running a lot of stuff. And you can sort of go back and forth as needed to solve your problem. Yeah, that's awesome. So I want to go back a little bit to the unstructured data conversation. What do you, what's hot there right now? Like, cause originally it was like, OK, everything in vector databases.

And obviously, still some of that. But say I've got hundreds of thousands of PDFs. What are you seeing people do that want to get that type of data into business intelligence, AI business intelligence layer? So I think there's a few different components. There's the deterministic filtering that's going to happen in whatever query language

or whatever database you're using. You probably do need some component of embedding or vectors. There's a lot of complicated stuff that you can do there. Chunking the documents, which embeddings do you use based on the domain where the text is going to show up most often to make sure your embeddings are most fitted to the problem you're trying to solve.

You're also going to need some hybrid component to the search as well. It's probably not going to work that well if you just use the embedding. You don't have also a keyword search as well. So it's like there's, it's going to be a pretty sophisticated system. Yeah, okay.

Yeah, that's what I've sensed. I think a lot of people that desire, and I've seen a lot of people start on that road and kind of of nevermind. It's just because of the complexity. Yeah. Not because they don't want it. Yeah.

Yeah. It's hard. It's hard in a different way from data. I think data is kind of like deceivingly easy. One of our best, one of our best sort of pipeline motions for customers is customers that have tried to build up themselves. Sure. Because it's like their own AIBI agent. Yeah, exactly.

Cause it's like, it's trivial to start. You can, you could literally use an MCP server and do it. Okay. Yeah. Great. And it's like, boom, it can go and it can answer any question and any table in your whole super warehouse, but it will inevitably pick the wrong table. A lot.

It will not have the business context on how you define gross margin. And you won't have the governance for anyone who's not an analyst, like SQL level permissions that you want to use. You won't have interoperability with dashboards or other things that are actually useful assets for data. So it's like, it's really easy to start. But as you kind of go down the road of not only do I not have this integration, do I not have this governance,

most importantly, the business people don't actually know what happened. Because you can't show someone a giant 200-line SQL statement and say, hey, this is what we did. Check our works. I don't know what's going on there. So funny. Well, back to essentially the search problem,

that's been one of the things that I have wondered for years why nobody has just solved the basic search problem for BI. I can't think of a single BI tool that has great search. Oh, yeah. Because it's the reason it matters is I see two things. One, you would obviously not be surprised. Business users all the time ask for things that already exist. Oh, for sure.

But worse is analysts a lot of the time rebuild things that already exist too. So it really goes both ways. So no, it totally does. And I think a lot of the problem is due to hierarchies that get introduced and don't get searched over because it just makes the engineering a little bit easier. So like a good example is like, look, her has explores. If you just need to see revenue and you see a sales, the marketing, you know, pipeline,

revenue explore, like, but what do you click on? Yeah. I don't know. And it's like that, that usually stops a person and they're okay. Well, I don't know. And it's in the folder called leadership. Yeah, exactly. So it's like all these hierarchies that don't really make sense are what caused a lot of

problems with data discoverability. Yeah, super interesting. We're going to take a quick break from the episode to talk about our sponsor, RutterStack. Now I could say a bunch of nice things as if I found a fancy new tool, but John has been implementing RutterStack for over half a decade. John, you work with customer event data every day and you know how hard it can be to make for over half a decade. John, you work with customer event data every day, and you know how hard it can be to make sure that data is clean

and then to stream it everywhere it needs to go. Yeah, Eric, as you mobile, even server-side, and then send it to our downstream tools. Now, rumor has it that you have implemented the longest running production instance of Rutter Stack at six years and going. Yes, I can confirm that. And one of the reasons we picked Rutter Stack

was that it does not store the data and we can live stream data to our downstream tools. One of the things about the implementation that has been so common over all the years and with so many Rutter Stack customers is that it wasn't a wholesale replacement of your stack. It fit right into your existing tool set. Yeah, and even with technical tools, Eric, things like Kafka or PubSub, but you don't

have to have all that complicated customer data infrastructure. Well, if you need to stream clean customer data to your entire stack, including your data infrastructure tools, head over to rudderstack.com to learn more. Let's talk products. Like, so Zenletic, we talked to you about the products

maybe over six months ago now. Yeah, maybe give us a little bit of a roundup of what you guys have been working on, some things you guys have rolled out. Yeah, so a lot of cool stuff on our side. So workflows are maybe the biggest feature that we've launched. Workflows are a way to take intelligent analytics, like intelligence or processes.

So some of our customers really use this for a weekly business review. So they're a little good group, pull a bunch of data, be able to take that to impact. It's like, hey, you know, this skew went down the most. We walked that hierarchy and went into the, you know, class that went down the most, the department that went down the most, the individual skew levels. And you're able to get this very comprehensive analysis that can go step by step and incorporate some intelligence the same way a human would.

Other use cases where people are taking this feature and using it for almost like dispute resolution where you can take, you know, I've got a screenshot or a CSV from SAP and I need to take all the items in there and marry that with something in my warehouse and figure out, you know, based on this, what sort of message should I be sending to each of the suppliers that come through in that CSV I got from SAP? So a lot of different, really interesting use cases when you can kind of take some of the processes that you do for analytics and turn that into something repeatable that you can just schedule to run every week on Monday, you can just incrementally remove this work

from your play. That's one of the big ones. The other stuff is a lot around making our interface more flexible for the language models. So, so, so I think to get a little history of how these interfaces work, yeah, there is, there's kind of text to see one or base like snowflake cortex is maybe the biggest one, but a lot of YC companies are doing this. And it's really flexible. It can do anything, right? It's just whatever you can do in your warehouse. biggest one, but a lot of YC companies what you did in the same way a talented analyst would and in a really high fidelity way to a business user so they can trust the result that they're actually getting. That trust piece is

really important as it's missing in Texas SQL. Then there's Texas semantically or we kind of led the charge in this and you know for a long time we were saying like hey this is how you got to do data and I think we've realized that also does. Interesting. it's too restrictive and it ends up being where you end up just pulling mostly paths that already exist on your dashboards and just instead of finding a dashboard and clicking on something you're just pulling that it's not flexible enough to give you the power you need to actually answer questions so the pro of course is trust yeah it's gone right but it doesn't have the fun which if you had to pick doesn't have the fun. Which if you had to pick, then if you're going to do it. Yeah, you're right.

Which if you had to pick, then you would pick the types of semantic layer. But the unstructured data part, how do you put a semantic layer on unstructured data, for example? So I think this is actually the exact distinction that I would want to get in. This is what we're building on the structuring side,

is it's like the way I think about the problem is that you have to interpret what you did to the user. So to take one step back before we hop into that, the unstructured question, why does it work with unstructured data? Why does deep research work? Because humans just naturally understand the, okay, you said you got this thing from here. There's a link where I got it from.

You can click on the link. It's literally a good look. We've been doing that for a really long time. So you can go and on the link. It's exactly the same thing. Yeah, it's literally a good look. Yeah, okay. We've been doing that for a really long time. So you can go and check the sources, and any human can do that. If you see something that looks questionable, you can say, I don't know, check the source. Maybe that's not quite a fair summary of that. And you can dig in really easily.

That's one of the reasons that deep research is so successful. If researchers came in and said, hey, I came up with this, and you had no way to dig in or question any of the things that came up with. It just wouldn't work as well. Yeah, definitely. And that's exactly how I think about the problem on the data side. You need to get the model the most flexibility.

You need to free the SQL, like let it write SQL, like language models. Like we were just saying, it had become perfect at coding. Um, I think that's inevitable and you need to let them write SQL, let them do what they're good at. But then the hard job of the application is how do you take that SQL that the model generated and has a truly trustable way for the business person to be able to look at that and know what it did

with absolute certainty, the same way they can have certainty if they clicked on a citation link in a text document. So does the semantic layer become a QA layer? Is that part of quality control? I think you have to think about the semantic layer almost in this inverse way, where it's like the whole time up until now, we've been taking the semantic layer and saying, hey, we've got all these building blocks

that humans understand. And we then have to, the semantic layer's job is to compile that down into a bunch of SQL that we brought on the warehouse. I think the real job of it in this new age is the inverse. You have a bunch of SQL that the language model wrote, given the context that's in your semantic layer. Like the semantic layer still owns the context, it's still structured kind of similarly in terms of its input. But the language model has written the SQL, raw SQL based on the input here. The semantic layer's job now is how do you take the SQL language

model as written and effectively invert the problem. How do you take that and take it back into business concepts that a normal human could understand without having to write the read. Yeah. Those kind of translation problem. Yeah, exactly. Which elements are good at. Elements are good at.

And there's a lot of code that humans have the right to do. So it's because again, you can't forget permission. You can't forget governance. You can't forget all the complexity and actually being able to verify that you did stuff the way you think you did. So that's a lot of what's what we're working on and what we're replacing. I mean, speaking of SQL, then we're going to go down the flexibility route.

I think they're both actually really important The reason for a sequel is that sequel the lingua franca of data because it's what runs natively in the data Yeah, right. Like if everyone had the money to just spin up Massive spark clusters to sit on top of the air, you know, right everyone's moving guys for And great, yeah, I thought would be fantastic But it's like as it is now, you've got to bring the data. And that's what's most effective. It's just most effective.

And then you do the aggregation. You take this truly massive amount of data, aggregate down to something reasonable, and then you have Python work on it in the setup of that. And that's exactly how we're architecting. All right, well, I got to ask you that roadmap stuff. What is something like kind of midterm, you don't have to give us details that you're excited about, just maybe directionally. One of the biggest things that I'll give you too,

one's like an improvement on our current experience. And that goes really in line with all the flexibility stuff we've been talking about. Part of our problem and every semantic layer BI tools problem is the setup process. So if you buy Luster, if you buy us, if you buy Ballistics, if you buy whoever, you're going to have to do a lot of work setting up this setup. Part, again, of how we think about the model should be able to generate SQL and we should be able to translate that into intelligible concepts for you. That also means the model needs to be able to

build your semantic layer for you. Sure. Because as it's translating that into intelligible concepts, if you're an admin, you should be able to say, hey, that's now governed concept. That's verified. Whenever you show that to the user, give them the thumbs up. Like that is good. We know that's good. And so now we can actually help you build the semantic model as you go, as opposed to you having to build it upfront. That's a huge difference in terms of that's so good. Like democratization, you know, is overused at this point, but that's like a reality when you can do stuff like that. Yeah, because otherwise you, and this is what we saw, and I'm sure every looker and everybody else saw this too It's like otherwise the semantic model becomes something that the data person is building and trying to guess at what the user actually wants to do

And instead it should be where the user is able to ask questions they want get answers and they know which ones are sort of reverse like approved by the data team and which ones aren't and They can either ask the data team about or take with a grain of salt. I was going to say that, do you imagine there are like gold standard semantic model and maybe some personalization, personalized semantic versions too?

Yeah, so I think the way we think about it is actually with the feature we're calling DynamoFields, where it's like you have the governed semantic measures. Those the model uses all the time whenever you ask about those concepts. But if you say, hey, what does gross margin look like if we take out discounts and we take out refunds and we add in this other like adjusting factor

and it can just do that for you. And it'll say like, hey, here's your adjusted gross margin that you asked about. John's gross margin. Yeah, exactly. It's like, hey, I did it for you. It's not the verified one. And you can see the verified one next to it.

It's not the same thing as that. So it's like, you can't really confuse the two, but it's like, hey, I did it for you. It's not the verified one and you can see the verified one next to it. It's not the same thing as that. So it's like you can't really confuse the two, but it's like you do need that flexibility. And that's one thing that I think Looker and similar sort of BI products are going to miss with this approach is that people don't want to come in and just ask about the same things on their dashboards. They want to ask about new things. They want to be able to do stuff they were not able to do before. Previously, they could only do a conversation with the analyst.

Right. Well, and I think that then at least every BI analyst role that I've been in the past, the like not well kept secret is that most of what you do goes to waste. Yes, totally. It's like people either don't look at it, they look at it once, or they forget that they ask, you know, and yeah, and then you do it a week later and they're like, Oh, I forgot about that. So like, just the inefficiency there is mind-boggling. And I think anybody that's been in the analyst for longer than a couple of weeks now is,

Oh yeah. And then to paint that picture again again a little bit on like where we're going and also on that piece too with the inefficiencies. It's like I view us as a, again, like McKinsey consultant and co-worker. And it's like the way that works now is that we have the business context via the semantic model. This analyst is able to answer questions for you, interpret what those things mean, be really flexible, you know, help you build assets that matter, like these intelligent workflows, these dashboards. And then as we get more advanced, and as the AI systems get more advanced, it's like that increasingly becomes something that you're able to get the deeper questions in your business too, the harder and harder things like Sam Alvin said in the Snowflake keynote, it's

like you expect in next year, people will not just be getting their sort of mundane questions they ask, but give it your harder questions. Give it the really difficult stuff. Let me see that. And then the other thing is you want it to be proactive. You don't want to have to go and ask this and that and everything. You want it to be able to beam on a train.

Here's your inventory. And it's like, let me know when I should be buying stuff. I don't want to ask you when I should be buying stuff. I don't want you to tell me when I should be buying stuff. So that's a lot of work. I'm curious on the proactive stuff, because I've had that come up a number of times. How far out do you think we are in really meaningful, useful,

hey, because that's all another domain, too, that we haven't even touched on, is monitoring and alerting. I've got a little bit of the dead-offs background, like a huge component to that. And some interesting like solves around that, like companies like Datadog. How do you think that comes to be iSpace? Because the iSpace has been bad, honestly,

not very good at the monitoring and you know, anomaly detection and stuff. I think it's also just that it's not, it's because the anomaly that they are explicitly trying to detect, it's not the anomaly they actually care about. So it's not, it's because the anomaly that they are explicitly trying to detect is not the anomaly they actually care about. So it's like, you don't actually care. And that X, Y, Z metric is like two standard deviations. No, very rarely do actually care about that.

I think what you care about is some process that's affected by it. And the ideal interface there is that you can say, like, Hey, I'm worried. Like if any of our product pages go down, it would result in a drop in, you know, conversion rate for that product. It's like apply general sort of human heuristics. And it's like, if it drops to zero because there's no visits, then it's obviously no problem. Yeah.

And it's like, there's a lot of different heuristics that humans have to be able to say, well, is this really a problem or not? And if you just alert the person every time the thing drops, you can't do additions from its last value, you're going to lose all of that. Yeah. And try to, there's so many false positives and people ignore it and it's useless. Totally. And it's like a lot of how I think about as the agents become more ambient, like

the next step in that direction is that you say, make sure nothing bad is going on with conversion rate and it can actually check, hey, well, we've got all the conversion rate. These ones, the number of sessions also dropped to zero. So from like four, that's not really a problem. And it's like it's able to just alert you when things are actually a problem, as opposed to every time some standardization. Yeah. And then you're starting from the end and saying, like, talking about conversion rate or revenue or whatever,

and then talking, well, if it impacts that, that's what I care about. You're not several layers down of, well, monitor each page and check the traffic. And you're going to, like you said, miss something, or something is going to get alerted to, well, no, I didn't mean that. Because you just have to be so precise if you're

going to monitor the old way. Totally. And that's why a lot of the monitoring codes just you're going to monitor like kind of the old way. Totally. And that's why a lot of the monitoring codes just as kind of brittle like sequel or add on like it's written that doesn't find some edge case. And it's like really what you want is reasoning. You want your intelligence applied to the problem. It's not even a lot of intelligence.

It's just like a little no. Previously, that was possible. Right. So it's like I think that's sort of the next step in terms of how things get kind of more ambient and more just for you. And then at some point you onboard the system and the system is just crawling around in the background and you decide how much compute do I want to allocate. Yeah. It's sort of figuring things out for me. And you can just slide. Hey, I want you to just go and find all the stuff and bring it to me.

I'm willing to spend on. Yeah. You're like, just check a little bit. Like, yeah, right. Don to just go and find all the stuff and bring it to me. I'm willing to, I'm willing to spend on, you're like, just check a little bit. Like, don't check that. That's the big stuff. I mean, how far away do you think we are from something like that? I mean, that's like where I work. That's on our roadmap for the end of 2026. Okay. So, yeah, close.

I think it's, it's closer than you think. I think it's because the more flexibility you give the models and the better job you do brokering the business context, they can just do this increasingly magical stuff. Yeah. Through no, through no like, it's not like we or anyone else or cursor builds like the, you know, you don't have to build them all. Right? Right. Your job is to build them all, right?

Your job is to orchestrate all the stuff around it. And the assumption that, you know, us and cursor and a lot of people are making who are taking these bigger bets around AI is that they're just going to keep getting better. Right. Which means like as long as you give them the right flexibility and the right guardrails, then you're set up for them. Just get dramatically better and be able to take on these bigger tasks just naturally, as long as you give them the right tools. And the right way to think about the tools is what tools do the humans have that are really good?

Sure. So it's like, yeah, the humans do have a BI system, which people have a semantic layer that you can look around in and pick the things with the semantic layer. The humans also have a SQL editor. Right, yeah, yeah. And both of those, due to the human's understanding

and communication skills, can become interpretable to the receiver. Yeah, and so it's like, OK, we're going to take those things and give it to the AI as well and not limit it to one or the other. Yeah, exactly. And I think the thing that you're going to see miss from all the kind of BI incumbents

is that they're too fixated on BI. Like, they don't think about themselves as a human as an employee that you buy and then think about what tools you need to give the employee like well we should let the AI agent move things around in the BI interface. That's great that can answer a subset of problems but that's not actually going to replace the kind of work. It's marginally better. It's marginally better. Well, I think we're at time, but this has been a blast. John, thanks for coming over.

The Data Stack Show is brought to you by Rutter Stack. Learn more at rudderstack.com. Music