Ten years ago, this community rewrote how the world works with data. Analytics engineers turned data teams into strategic drivers of the business.

We're at another one of those moments. AI changes what it means to work with data. The consumers of your models are shifting from analysts in a BI tool to agents acting on their own, at machine speed and scale. Four out of five data leaders say their data isn't ready for enterprise AI. The teams who win this era are the ones who build the foundation everything else runs on. That's you.

dbt Summit 2026 is where this community levels up together. September 15-18 at The Cosmopolitan in Las Vegas, with 100+ sessions across keynotes, breakouts, hands-on labs, and peer exchanges. We've pulled out just a sample of the sessions worth planning your week around and organized them by role, so you can jump straight to what fits your work. Register here.

For analytics and data engineers

You're the one shipping models. These sessions are about getting hands-on with what's new and hearing from peers who have already put agents to work.

Hands-on learning

Scaling trusted self-service for dbt stakeholders. Scale dbt beyond the build team by helping stakeholders find, understand, and reuse trusted data products without turning everyone into a developer. dbt Labs Resident Architect Kyle Tuft and Solutions Architect Matteo Dijoux walk you through documentation patterns, ownership, and a stakeholder access model that expands governed consumption while protecting your development workflow. Bring a laptop; this is a working session.

Breakout sessions

YAML doesn't know why: Building the business context your agents are missing. The dbt MCP server, dbt skills, and dbt Semantic Layer give agents technical fluency. The context agents keep missing is organizational. Pedro Heyerdahl of Kilo Code shows how to build a living context layer that multiple agents can read from and contribute to.

From AI experiment to production: How Okta governs context for agents at scale. Okta found that production AI depended less on a better model and more on a governed, discoverable semantic layer any agent could reason over from day one, built on dbt as the source of truth. Pooja Crahen shares how they got there.

Peer exchanges

Peer exchanges are small-group, discussion-first sessions. You bring your experience and your notepad.

Agents, MCPs, and buzzword fatigue: What AI actually changes for analytics engineers. New AI tooling launches weekly and the terminology multiplies faster than the problems it solves. XiaoHan Li of Xebia hosts a hype-free conversation about which tools actually stuck, who owns the logic when AI writes your models, and the skills worth investing in as more of the boilerplate gets automated.

How to build a successful data career. Three practitioners, Millie Symns of Justworks, Silja Märdla of Bolt, and Bruno Lima of phData, trade practical patterns for building career momentum as AI shifts what's expected of the role. Expect honest talk about durable skills, cross-industry moves, and making your impact visible without defaulting to "just become a manager."

For data team leaders

You're deciding how your team scales, standardizes, and stays ahead. These sessions are about rollout patterns, governance, and positioning your team for what's next.

Breakout sessions

From selection to scale: How ING is operationalizing dbt across a global bank. Jarno Boeijink shares how ING drives governed enterprise adoption inside a regulated bank, and how the dbt Semantic Layer and the dbt MCP server are opening new ground for natural-language analytics and AI-assisted development.

Governed by default: How data teams at Nordstrom turn dbt governance into an AI advantage. Nadine Bruxel makes the case for dbt as the control surface for safe AI: freshness as an AI SLA, tests and contracts and lineage as guardrails, and a conversational agent built on top of all of it. Governance-first is how Nordstrom gets to AI readiness.

Multi-agent dbt orchestration at Riot Games: Redefining the analytics engineering SDLC. Jessica Zhang shows how Riot Games safely coordinates multiple agents to read metadata, translate legacy logic, and generate pull requests, all behind read-only guardrails. If you want to know how far multi-agent workflows can go in production, this is the session.

Peer exchanges

Beyond the bottleneck: Position your analytics engineering team as a strategic force. Kasey Mazza of HubSpot leads a discussion on moving your analytics engineering team from a service desk to a strategic driver, with the framing and language to make that shift stick with leadership.

How to build a successful data career. Worth the crossover for leaders too. Millie Symns, Silja Märdla, and Bruno Lima swap patterns on durable skills and career growth, useful for anyone coaching a team through the AI shift.

For business leaders and executives

You're weighing where data investment turns into measurable outcomes. These sessions lead with results, governance, and the business case for a strong data foundation.

Breakout sessions

Real-time analytics at Bilt: Architecture and approach. James Dorado and the Bilt team walk through the architecture behind their real-time analytics, powering audience targeting and offer execution on fresh data.

An AlphaSense case study: Scaling AI on enterprise data with dbt-first governance and context from Euno. Sarah Levy of Euno and Brad Levy of AlphaSense show how dbt-first governance, paired with automated context, keeps AI decisions explainable and traceable back to governed source data.

Automating the impossible: Migrating 40,000+ objects to dbt in 9 months. Rafal Guziak shares how Philip Morris International automated the migration of 40,000+ legacy procedures to dbt, with SQL conversion, dependency mapping, and CI/CD, landing 40+ data products.

Governed by default: How data teams at Nordstrom turn dbt governance into an AI advantage. The executive read on the Nordstrom story: governance is the thing that makes AI on your data safe to trust and safe to scale. Nadine Bruxel shows how a governance-first foundation turns into a real advantage in the AI era.

Peer exchanges

Empowering stakeholders in the age of AI. Lexi Galantino of Zipline hosts a conversation on what "talk to your data" actually takes: which models make it work, how you keep the answers correct, and what the role of the data team becomes when stakeholders can propose their own changes.

Build your week around it

These are a fraction of the 100+ sessions on the agenda. Anchor your schedule around the two keynotes, Level Up on Wednesday and the Community Keynote on Thursday, then fill in the breakouts, labs, and peer exchanges that map to what you're building next.

Registration is open now, and the $1,695 registration includes a free training and certification while spots last. Bring home the templates, playbooks, and patterns you can put to work immediately.

Your next level starts here. Register for dbt Summit.