Two leaders from StubHub have spent their careers building with data and building data teams. Over the last few years, they've constantly reevaluated roles amid AI innovations and obsessed over what comes next for data careers. In this talk, Meghana and Nova walk through how they navigated the new normal as AI entered every aspect of their profession. They'll share how their thinking evolved: Not by clinging to old job definitions, but by re-contextualizing what their roles were for all along. Equal parts group therapy and field report, this session is for any data professional who's wondered where their role is headed.

When Chime®'s data team sought to scale AI across its organization, Jordan Farrer, Director of Data Science, put humans at the center of the work. He resisted pressure to mass-ingest the "valuable" context sitting in Confluence, Slack, and Google Drive. In this talk, Jordan, Dori, and Connor share where human judgment is non-negotiable, where AI belongs, and how they manage context at scale.

There are a lot of stories out there about AI on data, and many concerns about whether it can really work in a complex org. At the NHL, Oliver Gomes built his data team for the exact purpose of enabling everyone to self-serve using modern AI tools. In this session, you’ll learn how Oliver re-architected his team, what they learned, and how it resulted in a repeatable operating model that will scale across the league.

AI can build a chart, write a query, even frame a story. But you’re the one who knows your business. As data teams take on new technology, the real work shifts to judgment: curating context, catching the moments AI gets the framing wrong, and creating the conditions for people across the business to use AI well. Andy Cotgreave draws on real examples of AI-driven data storytelling to lay out what leadership looks like when the technology moves faster than any policy can.

Welcome to the post-AI data stack era. Now data teams have two jobs: enable everyone to confidently build with data and create a singular reality for their company to operate. The differentiator between an okay and a great data scientist is their ability to encode judgment, influence, and consensus into infrastructure. Join Ian Macomber, VP of Data at Ramp, as he shares the story of how we got here and why the scarce resource becomes company-wide truth.

With AI, analysis can happen anywhere. Devon Mittow, who leads data platform work at Lyft, keeps metrics and context in a tool-agnostic layer so analysis can start in Claude, pass through an MCP, and land in Hex. He and Hex co-founder and CTO Caitlin Colgrove dig into the next wave of data engineering: less about moving data between tools, more about structuring context for whatever surface the question comes from.

When the digital editor-in-chief can build their own Replit app in minutes, their expectations of waiting on the data team shrink to near zero. As data teams meet AI-speed expectations, the work shifts to judgment: knowing when fast-and-directional wins, when numbers must be exact, and owning the meaning people take from the data. Sean Hunter, SVP of Core Engineering at Fox, uses real examples to show how to deliver both when everyone expects data instantly.

Data teams finally got what they've wanted for a decade: self-service. Stakeholders no longer wait in the queue—but that shift changed how data teams work. Four leaders who rebuilt their teams this year join Katie Bauer: Emily (LangChain), Nick (Ro Health), Evan (Whatnot), and Firdevs Abacioglu (Airwallex). They'll cover the trade-offs of self-service and how they rebuilt workflows now that everyone's a data person.

For years, the deal between data teams and business stakeholders was simple: you ask, we build. Then AI put a prompt bar in front of everyone, and suddenly business users could explore data themselves. Today both sides are figuring out new rules of engagement in real time. In this roundtable, John Choi and Nick Pasquinzo will deliberately and unapologetically force you to break down the barriers between business users and data folks. Come prepared for an interactive, lively discussion.

How good are AI models at the work we actually pay analysts to do? There’s lots of SQL benchmarks, but that's a small fraction of “analytics”. A skilled analyst has to figure out what's really being asked, make sense of messy and conflicting data, and know whether a result is strong enough to base a decision on. We should expect the same of AI! At Hex, we built DataBench to test that standard across 100 realistic business questions. It changed how we view which ones are actually good at analytics. Izzy Miller shares the mistakes behind those results, what it took to catch them, and where models shine.

AI has made it possible for anyone to build whatever they want: a dashboard, an app, a slide deck. The how isn’t the constraint anymore; it’s deciding what you actually want to create. In this 30-minute workshop, we’ll teach you how to prompt, design, and build generative apps. No prior coding required, and no matter where you sit — marketing, sales, CS, ops, or the data team — if you can describe what you want, you can build it.