Snowflake Summit 2026 | Baris Gultekin, Snowflake & Nicolas Doyen, Viral Nation

Clips
More from Snowflake Summit 2026

Baris Gultekin

VP of AI

Snowflake

Nicolas Doyen

VP of Product and Data Strategy

Viral Nation

play_circle_outline Viral Nation overview: social media agency using AI and data for brand activations

play_circle_outline Snowflake Rebrands Intelligence as CoWork: Personal Autonomous Agent for Scheduling, Prioritizing and Running Tasks Across Email, Calendar, Slack, Natoma

play_circle_outline AI-Augmented Agencies: Human Creativity and Automated Agents for Scalable Multimodal Social Data

play_circle_outline Governance and security: data access, context quality, and guardrails for trustworthy agents

Info
Transcript

Baris Gultekin, Snowflake & Nicolas Doyen, Viral Nation

Baris Gultekin

VP of AI Snowflake

Nicolas Doyen

VP of Product and Data Strategy Viral Nation

In this interview from Snowflake Summit 2026, Baris Gultekin, vice president of product and AI at Snowflake, joins Nicolas Doyen, vice president of product and data strategy at Viral Nation, to talk with theCUBE's Dave Vellante and Rebecca Knight about how governed agentic AI is moving enterprises from passive analytics to intelligent systems that coordinate and take action. Gultekin walks through the three pillars of the Snowflake CoWork announcement: a personal work agent with governed connectors to email, calendar and Slack, enabled by the Natoma acquisiti... Read more

explore Keep Exploring

How does Viral Nation use AI and Snowflake in its social media work—particularly for brand safety, data collection, and informing strategic decision-making? add

What were the major announcements, and why do they matter to customers like Viral Nation? add

How will the future of marketing/advertising agencies look in an AI-driven world—will AI agents replace people, how will agencies use AI to plan, test, and measure campaigns, and what role can Snowflake play in enabling that? add

Who within an organization is able to work with AI? add

bolt Powered by CUBE AI

Baris Gultekin, Snowflake & Nicolas Doyen, Viral Nation

search

Rebecca Knight

>> Hello everyone. Welcome back to theCUBE's live coverage of the Snowflake Summit here at the Moscone Center in San Francisco. I'm your host, Rebecca Knight alongside Dave Vellante, co-founder and co-CEO of theCUBE. I would like to welcome our next guest to the show. We have Baris Gultekin, VP of AI at Snowflake. Welcome back, I should say.

Baris Gultekin

>> Thank you.

Rebecca Knight

>> And Nicholas Doyen, VP of product and data strategy at Viral Nation. Thanks so much for coming on the show.

Nicolas Doyen

>> Thanks for having me.

Rebecca Knight

>> So I want to start with you, Nicholas, because Viral Nation sits at the intersection of AI, data, and creative. Why don't you start by telling our viewers a little bit about your organization?

Nicolas Doyen

>> Sure. Yeah, absolutely. So Viral Nation is a social media agency. So we work with brands to help them activate on social, work with creators, launch products, talk about their brand, et cetera. We're using AI pretty much everywhere in the business at this point, as many businesses are these days. But we started out by first focusing on brand safety as a core focus. And that allowed us to start to collect all the media from social, start to analyze it. And it turned into a dataset that now is powering all of our strategic decision making. It's driving us to create new creative ideas and identify how brands should go and activate on social and reach the audiences that they're looking at. Snowflake's been a big part of that and yeah, really excited about the future of data and AI.

Dave Vellante

>> Brand safety is what? Don't step in it on social media? Explain that.

Nicolas Doyen

>> It's a great question. First and foremost, these are partnerships with creators and creators are not necessarily professionally oriented. And so they're always posting about all sorts of different topics, having opinions. And so when brands work with creators, they want to ensure that creators are brands safe and remain brand safe both before, during, and after their activations. And so we really work with brands to help them feel safe going onto social and especially with the rise of agentic, engine optimization, which is using social for a lot of that decision making and brands are losing control of how their brand shows up online. It's really critical that they create partnerships with creators that talk about them in the right ways and present the brand in the way that they want it to be presented.

Rebecca Knight

>> Okay. So Baris, Snowflake announced some major new CoWork innovations at Summit this week. Can you walk us through what those big announcements are and why they matter so much to customers like Viral Nation?

Baris Gultekin

>> Absolutely. So the first announcement was the name change itself. So went from Snowflake Intelligence to Snowflake CoWork, mostly representing the nature of that change, essentially expanding the charter of Snowflake Intelligence to broader getting work done with data and security at the core of it. So the announcements are, one, Snowflake Intelligence being CoWork now is your personal work agent. That means a lot of connectors to work context, email, calendar, Slack, and other connectors. With the Natoma acquisition, now Snowflake has access to those plus connectors in a secure and governed way. So the agents will be able to use that along with the data on Snowflake and on other platforms to give high quality answers. Also supporting more advanced capabilities like skills, understanding memory and learning about the personalization of the user is part of that umbrella. The second bucket is making Snowflake CoWork an autonomous agent so that it can go run things on your behalf. You can schedule tasks, it'll prioritize your emails, look at your Slack for things to do, prepare your meeting briefs and so forth. So doing more in a scheduled and automated way. And the third pillar is enabling more BI oriented experiences with AI as a first class citizen by making artifacts, dashboards that you can talk to. So admins can create these dashboards and they share it with their accounts on Snowflake CoWork, and business users can look at these dashboards, can talk to them, drill in and understand their business.

Dave Vellante

>> And Cortex Sense was also new or was that-

Baris Gultekin

>> Cortex Sense is new as well.

Dave Vellante

>> Okay.

Baris Gultekin

>> So Cortex Sense is the pillar for builders essentially to build high quality agents with ease. So up until now, to build a high quality agent, you have to build a semantic model that can take a lot of effort to build high quality semantic models. Cortex Sense is a runtime for context that essentially connects to a series of underlying data sets and dashboards and other information to create a context runtime for agents. So anyone who's building agents can now out of the box get very high quality really quickly to build production agents.

Dave Vellante

>> So I was really excited about Cortex Sense because as an analyst, you're always looking to the future, right?

Baris Gultekin

>> Yes.

Dave Vellante

>> And to me, that's the future, right?

Baris Gultekin

>> Yes.

Dave Vellante

>> Being able to harmonize all that data and potentially the process knowledge.

Baris Gultekin

>> Exactly.

Dave Vellante

>> But when you talk to the customers, it's like, okay, we'll get there, but we got to get these agents to work and we have to make them trustworthy and we've got to rethink our processes. So when you see these announcements, how do you respond? Where can you apply them in your business today? Where are you at on that journey?

Nicolas Doyen

>> Yeah. First we meet it with excitement. At the end of the day, as a social media agency, our job is to be at the forefront of public opinion and understanding what people are talking about. And so the CoWork announcement in particular was really exciting for us. We did an activation earlier this year with Snowflake Intelligence at the time where we were looking at social conversation around ads during the largest publicized sporting event in the US. And what we were doing was trying to understand how are people reacting to the ads? How do you identify if the brand has achieved its goal by running these super expensive ads and these ad runs? And in order to do that, we were taking in all that information, and we used Snowflake Intelligence to help us parse through it live on that show to talk about how everyone's talking about it. So that was the foundation of a lot of our BI transition that we've been doing. CoWork's that next step where it's now not just BI inside of our business, but it's actually guiding the way in which we make decisions across the whole business and we actually work and operate. So it's really exciting. Cortex Sense, we haven't had a chance yet to play with it, but we were excited by the opportunity. We deal a lot with really large context windows that we run out of context. So being able to figure out different ways that we're able to use all the information that's on social in this decision making process more effectively is something we're super, super excited about.

Dave Vellante

>> That large sporting event, we can't say the S word.

Nicolas Doyen

>> You can't say the S word.

Dave Vellante

>> Okay. Wow, you got the W coming up.

Rebecca Knight

>> The big W, right?

Nicolas Doyen

>> Yeah.

Rebecca Knight

>> Yeah.

Dave Vellante

>> So I'm excited about that. I got tickets at Gillette. I got tickets at MetLife.

Rebecca Knight

>> Wow, very cool.

Dave Vellante

>> We got France Senegal at MetLife and we got France and Norway, which is going to be amazing.

Rebecca Knight

>> Yeah, at Gillette.

Dave Vellante

>> It's going to be an incredible social media madness worldwide.

Nicolas Doyen

>> We're excited and ready.

Dave Vellante

>> That's really good.

Rebecca Knight

>> So Baris, when you hear Nick talk about how they are putting AI into their marketing and creative workflows, what does that signal to you about how the enterprise is thinking about AI in general?

Baris Gultekin

>> Yeah. I mean, great to partner with companies like Viral Nation where AI is at the forefront from the beginning of the data journey to the end. So they are using AI from kind of processing the data, understanding, capturing, and extracting information to analyzing and building agents that you can talk to. So I get very, very excited about real life use cases. We are expanding our capabilities to multimodal now, and Viral Nation, of course, processes a lot of multi, multimodal data. So great to bring those products to our customers and then have them use it and get feedback to us.

Dave Vellante

>> So ... Go ahead, please.

Rebecca Knight

>> I was just going to ask, because there is such a debate right now about AI and its use in more creative and more artistic fields. I mean, how do you think about that? I saw today that Martin Scorsese is talking about using AI and becoming more of an advocate for it in Hollywood, which has really been verboten. And so how do you think about just those critical questions about AI and taking out the humans in that more creative work?

Nicolas Doyen

>> It's a tough question and something we struggle with a lot to figure out our answer to. What I would say is we've started by first focusing on identifying the white space inside of social media, trying to understand where are brands showing up, how are they showing up differently from you to maybe expect. One of our favorite examples is we were working with a major CPG and we were doing an analysis on soda, and we were looking for Pepsi and Coke and the likes. We found out that there's an entire space on social of Korean cooking where they're using Pepsi as a sweetener in noodle dishes. And that type of-

Rebecca Knight

>> In noodle dishes?

Nicolas Doyen

>> In noodle dishes, exactly. Yeah.

Rebecca Knight

>> Okay. All right.

Nicolas Doyen

>> And so that type of example is great for the brands to kind of identify new ways they can enter into the market, show that their products are being successful, et cetera. And so we're using AI a lot for trying to identify those types of cases. We also use it to help augment our strategy teams that work with the brands to figure out what's the best way to activate on social, how should you be thinking about activating it? And of course we're testing in the studio and in the creative, are there ways in which to use AI? I'll say so far we feel that leaving it to the creators who are the ones that create all of this art and bring it into the market should continue to do so in the way they see fit, but trying to streamline that creation process, at least at the beginning and ideation and alignment to the brand, is something that we're really focused on.

Dave Vellante

>> Nick, we're in the AI bubble here in the tech world and we're like, "Yay, AI, let's go." But of course, we hear at the graduations, the commencements, we hear the boos, there's a lot of negative sentiment around AI. How do you think about that? What are you advising brands in terms of their positioning and how they talk about AI?

Nicolas Doyen

>> It's another great question. I'll say at the end of the day, brands need to make their position first, and our job is to help them figure out how to bring that position forward to the market in a way that people are welcoming and open to. But if their position is one that's antagonistic, our job is to help them still send that message and then deal with the challenges that come with that. So we don't try and take a position as to how they should be thinking about AI necessarily in their business, but really thinking about how do you actually go and deliver that message and put your brand out there in the way that you want to at that time.

Dave Vellante

>> Do you guys brainstorm much about the agency of the future? I mean, you're an agency. We talk about agents, agentic systems, AI taking agency, taking action. What's the future of an agency look like and how do you guys think about that? How will Snowflake help you?

Nicolas Doyen

>> Yeah. I mean, at least for us, we're really thinking about it as a AI powered organization. And so we don't feel that in the marketing space, agents are going to do everything, but they are going to speed up the process of testing different ideas, validating different places that you want to show your brand, and then helping brands measure that after the fact to actually see the ROI of their marketing spend. Especially on social, that's been something really complicated in the last many years. There's no longer a way to do proper identity resolution on social, and so you have to prove success in other ways and that's where AI is really starting to come in. Dealing with unstructured information that Snowflake's really effective at is one of the critical parts. Taking the transcripts, taking the different ways in which people are talking about the different subjects or different brands on social requires us to have that computing power and AI models in place to analyze it. So the agency of the future, it's still going to be people driven, but supported by AI. We feel that way definitely. But it's allowing us to do a lot more and test a lot more theories and a lot more go to market actions before we necessarily go and take that action.

Dave Vellante

>> Well, the creative piece obviously stays with humans. I would think the coordination is another piece where Snowflake can help at least automate or speed up. I mean, how do you see that?

Nicolas Doyen

>> Absolutely. I mean, CoWork, at the end of the day, is going to be streamlining a lot of our communication processes, a lot of the ways in which we manage the coordination of either events or of posts going live. Validating that the right asset has gone live on social is actually something quite complicated at scale. Historically, we had to have people that would be sitting there watching the video when it goes live and we activate thousands of influencers every month and that can be really, really expensive. And so CoWork as a way to actually start to have those conversations in an agentic way is definitely going to be important. We do a lot of live events. Unfortunately, agents can't be there and human, at least not yet. We'll see how the robots come through, but there is still definitely a need for that. And I think there's still the conversation too about human conversation and the feeling of how somebody activates and works with people. We have creators that can be very complicated to work with for brands because they don't have a lot of the historical behavioral patterns that brands might expect. And I think agents are still going to have some trouble kind of navigating those types of conversations. So that mediation will still need to be there. But a lot of the day-to-day work, definitely we're looking at ways to automate that, streamline it with AI and with Snowflake.

Rebecca Knight

>> So Baris, when you hear Nick describing this agency of the future and the ways in which AI agents will work together with humans and take on some of their work, but also leave other work to humans, one of the big things we're hearing is that Snowflake is making it easier for people to work with AI, eliminating the friction. So how do you think about that in terms of how you're expanding? Who gets to actually work with AI in an organization?

Baris Gultekin

>> What Nick is explaining is very much what we're seeing across the board. AI transformation is happening both in kind of the way people work, changing functions like marketing, sales, as well as just changing lines of businesses altogether, just bringing a lot of automation to lines of businesses. In terms of the platform itself, again, the role of a human is going from kind of manually doing things, writing queries, just reporting on what's happened, to guiding agents, guardrailing, and ensuring that the right things happen. So from a platform perspective, that means governance is still at the core of everything. Security is really, really important. And agents are as good as the data they're provided access to. So a lot of the work is how do you bring that context to life? How do you make it easy for others like Viral Nation to build agents on top of all of the data that is increasingly more and more complex? So the kind of products that we're launching like Cortex Sense, for instance, makes that job a lot easier so that our customers can create high quality agents much faster.

Dave Vellante

>> We love, in this industry, we love stacks, pictures of stacks in our industry, but you got the data piece nailed at the bottom of this mini stack I'm going to develop. And then at the top you've got the execution piece and you're clearly going after that. You're moving into the sort of KPIs and metrics and dimensions allowing with CoWork to interact with that. And then I see Sense as sort of the meat in the middle.

Baris Gultekin

>> It's the glue, yes.

Dave Vellante

>> And that's where the intelligence layer-

Baris Gultekin

>> Exactly....

Dave Vellante

>> will emerge over the next several years. For a customer like you guys, it's probably going to take the better part of, big part of a decade to sort of play this out to get to that end state and then who knows where that'll take us. But is that how you're thinking about it or are you trying to bite an elephant one big bite at a time?

Nicolas Doyen

>> Yeah. I'll say we're trying to move a lot faster than 10 years, that's for sure. I think right now, everything is changing so quickly that organizations have to be on top of the ball. And we've actually built out an entire team inside of the business that is focused on trying to apply AI in novel ways inside of the business that's outside of our traditional product organization, and we're seeing great success with that. One of the things that we really were focusing on at the beginning when AI was coming through was a concept of should we have a dashboard? And I can definitely say right now probably we don't need one. There's a lot of ways in which now the agents are able to answer these questions and provide it at the speed that you need it to and a lot of that's built on top of Snowflake. And because of the size of the data that we're working with, things like Cortex Sense are going to be really critical for us to get those answers to be correct every time, which is definitely still a challenge here today. Yeah.

Dave Vellante

>> You say you don't need the dashboard because you can now talk to the data and you say, "Give me this. This is what I want to see." And Baris, for you, I see going well beyond talk to the data, talk to the data, interact with it, make it now do things

Baris Gultekin

>> Exactly. Take action.

Dave Vellante

>> Do new things for me and then take the action. That is the future, which is maybe coming faster than we realize.

Baris Gultekin

>> Yeah.

Nicolas Doyen

>> Yeah.

Rebecca Knight

>> Well, Baris and Nick, thank you both so much for coming on the show. Really interesting conversation.

Nicolas Doyen

>> Thank you.

Baris Gultekin

>> Thanks guys.

Nicolas Doyen

>> Yeah, thanks so much for having us.

Baris Gultekin

>> Appreciate it.

Rebecca Knight

>> I'm Rebecca Knight for Dave Vellante. Stay tuned for more of theCUBE's live coverage at the Snowflake Summit. You're watching theCUBE, the leader in enterprise tech news and analysis.