dbt Summit 2026: AI-Ready Data, Faster Engines, and Open Infrastructure Take Center Stage
A chief data officer at a health insurance administrator recently told dbt Labs a familiar story: he has two data engineers he wants to move onto AI work, he has executive backing and budget, yet...
A chief data officer at a health insurance administrator recently told dbt Labs a familiar story: he has two data engineers he wants to move onto AI work, he has executive backing and budget, yet he still can't start. His reason? "Your data is still not at the level where you can put an AI agent on top of that." That gap is exactly what this year's dbt Summit aims to address.
The event runs September 15-18 at The Cosmopolitan in Las Vegas. Organizers say AI is shifting data consumers from analysts in BI tools to agents working continuously at machine speed, without human checks on every answer. The summit's keynote is built on four pillars: leveling up the engine for faster parsing and cost control, leveling up for AI with governed context, leveling up with AI via the dbt Wizard and agentic workflows, and leveling up the stack with open, flexible infrastructure. Thursday's Community Keynote, featuring Grace Goheen and Jeremy Cohen, celebrates the dbt community and the open work on dbt Core.
Product sessions go deeper. One session covers dbt State, which checks warehouse metadata and model SQL to avoid rebuilding unchanged models, saving an average of 30% on compute. Another highlights dbt Wizard, a coding agent for data with three modes: develop, analyze, and discover. A session on rebuilding dbt Core details a shift to a Rust-based Apache 2 engine, a cleaner adapter model on ADBC and Arrow, and parquet-backed artifacts, with side-by-side parse speed demos.
Other sessions tackle semantic layers - now described as the most load-bearing piece of an AI strategy - and Apache Ossie, a vendor-neutral semantic interchange spec co-led by dbt Labs, Snowflake, Salesforce, BlackRock, and RelationalAI. Additional talks cover open data infrastructure with Iceberg, unified lineage from Fivetran and dbt, and giving agents accurate context through governed semantic models. More sessions will be announced before September. The summit runs September 15-18, 2026, at The Cosmopolitan in Las Vegas.
Source: dbt Labs Blog
Ready to Modernize Your Business?
Get your AI automation roadmap in minutes, not months.
Analyze Your Workflows →