Welcome to the Data Intelligence Era

With over 20,000 in-person attendees, Data + AI Summit 2025 was our biggest one yet. And whether they were watching one of the over 700+ sessions or test-riding a Trek bike, we’re confident everyone left feeling energized and informed.

As our CEO Ali Ghodsi proclaimed during his keynote: data intelligence is going to disrupt every industry. We’re already seeing this in motion. Industry leaders such as Jamie Dimon (CEO, J.P. Morgan Chase) and Dario Amodei (CEO and Co-founder, Anthropic) discussed the transformative impact of humans and AI agents working together, while Microsoft CEO Satya Nadella explored the rapid speed of AI progress.

At Data + AI Summit, we announced new products and capabilities that build on our vision for data intelligence. These updates are designed to help organizations:

• Get their AI agents into production, easily
• Put data and AI insights into the hands of everyone, including business users
• Simplify application development without database management headaches
• Empower data analysts to build reliable data pipelines without coding
• Build scalable, reliable data pipelines (with a fraction of the work typically required) and much more

Below are some of the highlights of what we announced.

Lakebase: Fully-managed Postgres for data apps and AI agents

The future of databases: unified analytical, operational and AI workloads and integrated into the lakehouse. Lakebase is built on open source Postgres, the de facto database for modern applications. It also leverages Neon’s architecture that separates compute and storage for independent scaling.

We believe Lakebase is how every OLTP database system should be built in the future.

Read more: Announcing Lakebase Public Preview

Agent Bricks: Production AI agents optimized on your data

We’re solving the biggest pain points in building and scaling high-quality, domain-specific AI agents. One of the biggest challenges is evaluating their performance on specific business problems.

With Agent Bricks, enterprises can perform custom evaluations, as well as auto-optimize their AI agents around the benchmark they set.

Read more: Introducing Agent Bricks: Auto-Optimized Agents Using Your Data

Lakeflow Designer: No-code production quality ETL

A new way for business analysts and data engineers to collaborate on shared, governed ETL pipelines. The no-code pipeline builder empowers users with drag-and-drop and natural language-based tooling. And because Designer outputs are Lakeflow Declarative Pipelines, there are no messy handoffs or rewrites.

Read more: Announcing Lakeflow Designer: No-Code ETL, Powered by the Databricks Intelligence Platform

Databricks Free Edition: Learn professional data and AI tools for free

What better way to learn data and AI than on the platform millions of professionals use every day? Databricks Free Edition offers our industry-leading platform for free, empowering students and enthusiasts to learn professional tools and stay competitive in the AI market. We’re also providing free self-paced training through Databricks Academy.

Read more: Introducing Databricks Free Edition

Databricks One: a new experience designed for business users

We’re bringing the power of data intelligence to everyone. Now, business users can access AI/BI Genie, Dashboards and Databricks Apps with minimal friction to reap the benefits of data and AI.

Whether it’s asking natural language inquiries, interacting with dashboards or using custom-built apps, Databricks One is the secure, intuitive interface non-technical employees need to make smarter, data-driven decisions.

Read more: Introducing Databricks One

Spark Declarative Pipelines: A proven standard for building reliable, scalable data pipelines

Declarative Pipelines, a proven standard for building reliable, scalable data pipelines, is now available as open source. By donating Declarative Pipelines to Apache Spark™, we’re giving users the power to simply define what their pipeline should do, and then let Apache Spark do the rest. This battle-tested, open standard will make it easy to build and operate reliable, scalable data pipelines.

Read more: Bringing Declarative Pipelines to the Apache Spark™ Open Source Project

Lakebridge: Fast, predictable migrations

Lakebridge makes moving from legacy data warehouses to Databricks easy and cost-effective. Features like profiling, assessment, SQL conversion, validation, and reconciliation give users the end-to-end support they need to ditch the old and embrace the open, unified future.

Automate 80% of migration tasks to accelerate implementation speeds by 2x – without any additional cost.

Read more: Introducing Lakebridge: Free, Open Data Migration to Databricks SQL

Managed Iceberg Tables: A fully open Apache Iceberg™ catalog

Databricks users can now write open Managed Iceberg Tables via Unity Catalog’s Iceberg REST Catalog API, as well as access and govern Iceberg tables managed by foreign catalogs. Connect to Unity Catalog from any engine. And break down data silos with access to data, across catalogs and regardless of format.

Read more: Announcing full Apache Iceberg™ support in Databricks

Other product updates and enhancements:

• Announcing GA of Databricks Apps
• Unity Catalog updates: UC Metrics
• AI/BI Genie is now Generally Available
• Lakeflow GA
• What’s New with Data Sharing and Collaboration

You can explore these releases and more by watching the Data + AI Summit keynotes and select sessions on demand.