What’s New in Azure Databricks at FabCon 2026: Lakebase, Lakeflow, and Genie

Accelerating data and AI on Azure

This week thousands of data professionals are gathering in Atlanta for Microsoft Fabric Community Conference (FabCon) 2026 — co-located with SQLCon for the first time — bringing together the Microsoft data and SQL communities to explore the future of analytics, BI and AI on Azure.

For Azure Databricks, FabCon highlights the continued momentum of our partnership with Microsoft and how customers are using Azure Databricks to unify data engineering, analytics, BI and AI.

Since 2017, Azure Databricks has been a first-party Azure service, deeply integrated with Microsoft services including Power BI, Excel, Microsoft Teams, Data Factory, Azure OpenAI, Microsoft Foundry, Copilot Studio, and Power Platform.

Enterprises across industries rely on Azure Databricks to build modern data platforms on open lakehouse architecture by combining the flexibility of open data with the performance and scale required for analytics and AI.

At FabCon this week, we’re introducing several new capabilities designed to make it even easier to build intelligent applications on Azure.

Lakeflow Connect Free Tier 

Reliable data ingestion is foundational to modern analytics and AI.

Today, we’re excited to introduce the Lakeflow Connect Free Tier so organizations can easily bring their enterprise data into their lakehouse to build analytics and AI applications.

Lakeflow Connect lets you mirror data from enterprise SaaS applications and databases directly into the lakehouse. Each workspace includes 100 free DBUs per day, allowing you to ingest approximately 100 million records per workspace, per day into the lakehouse at no charge before standard Lakeflow Connect pricing applies. This free tier includes all the benefits of Lakeflow Connect, including a simple UI, efficient ingestion, and unified governance through Unity Catalog.

Key capabilities include:

• Database mirroring for nine of the most used databases (SQL Server, Oracle, Teradata, PostgreSQL, MySQL, Snowflake, Redshift, Synapse, and BigQuery)
• Full support for many of the most popular SaaS applications (including Dynamics 365, Salesforce, ServiceNow, Workday, and Google Analytics)

Lakeflow Connect writes directly to open storage on Azure Data Lake Storage (ADLS), governed by Unity Catalog. Your ingested data is secure, discoverable and accessible from any engine the moment it lands.

Combined with Lakeflow’s orchestration and transformation capabilities, Azure Databricks provides a complete platform for building production data pipelines — enabling teams to build pipelines up to 25x faster while reducing ETL costs by up to 83%.

Learn more about how Lakeflow on Azure Databricks unifies ingestion, transformation, and orchestration on a single governed platform.

Azure Databricks Lakebase: the database for AI agents, now generally available

Modern applications increasingly require operational databases that integrate seamlessly with analytics and AI. In the era of AI agents, this is even more critical: agents need a transactional system of record to manage state, actions, and application workflows.

Azure Databricks Lakebase, now generally available, is a managed, serverless Postgres service that brings production-grade operational capabilities directly to your lakehouse foundation on Azure. Lakebase is the operational database for the agentic era that enables AI agents and applications to read, write, and reason over operational data directly in the lakehouse.

Lakebase combines familiar Postgres with the scalability and economics of open lakehouse storage. As organizations adopt agentic workflows powered by tools like Genie and Agent Bricks, Lakebase provides the operational database layer agents rely on to manage state and application workflows. This foundation also enables a new wave of agentic data engineering and agentic data science with tools like Genie Code.

Key capabilities include:

• Open Postgres foundation with community extension support such as pgvector and PostGIS
• Separated compute and storage enabling high throughput and efficient scaling
• Sub-second startup with autoscaling and scale-to-zero pricing
• Branching and instant restore for development workflows
• High availability with automatic failover across availability zones

With this launch, Lakebase is now available in 14 Azure regions worldwide, enabling organizations to run operational workloads directly on the Databricks platform.

Common use cases include:

• Operational applications and workflow systems
• AI agent state management
• Customer personalization and feature serving
• Transactional analytics on operational data

Extending Azure Databricks into Microsoft 365

Many business decisions still happen in familiar tools like Excel and Teams. A key focus of this week’s announcements is extending Azure Databricks into Microsoft 365 so governed data and AI insights are available where users already work.

This builds on previously announced integrations between Azure Databricks and Microsoft 365, including support for Genie-enabled Copilot Studio agents — allowing employees to receive trusted insights from Genie directly in Teams or M365 Copilot – as well as upcoming initiatives such as a Databricks app in Teams that enables direct access to Genie.

Azure Databricks Excel Add-in (Public Preview)

The Azure Databricks Excel Add-in connects Excel directly to governed lakehouse data.

Users can:

• Browse Unity Catalog tables and Metric Views directly from Excel
• Build pivot tables using Metric Views and governed semantic definitions
• Filter and analyze data without writing SQL

The add-in works across Excel for Windows, macOS and the web, helping organizations replace fragile exports with direct access to trusted lakehouse data.

To learn more, review the documentation.

Genie: AI that knows your business

AI-powered analytics adoption on Azure Databricks has grown rapidly, with 98% of Databricks SQL warehouse customers using AI/BI, and Genie monthly active users increasing >300% YoY. 

At the center of this experience is Genie, which enables users to ask questions about their data and receive answers as tables, charts or natural-language explanations.

Genie is the conversational AI experience for data in Databricks, while Genie Code extends these capabilities to developers building pipelines, ML models, BI dashboards and applications.

Genie Agent Mode 

For complex analytical questions, Genie Agent mode introduces an agentic approach to business analysis.

Agent mode uses multi-step reasoning and hypothesis testing to investigate complex questions and uncover deeper insights from enterprise data. Instead of returning a single query result, Genie can explore a problem iteratively and refine its approach as it learns from intermediate results. Genie Agent Mode enables users to go beyond basic “what happened” questions, to understand “why” and “what’s next”.

With Genie Agent mode, users can:

• Automatically generate and execute a research plan across multiple queries
• Test hypotheses and refine analysis based on intermediate findings
• Combine structured data exploration with narrative explanations and visualizations
• Deliver comprehensive answers supported by tables, charts and evidence

This transforms Genie from a simple conversational query interface into an AI analyst capable of investigating complex business problems.

Genie Code

For data practitioners, Genie Code enables agentic data engineering, data science, and analytics workflows directly in the Databricks workspace.

Genie Code is an AI agent built specifically for data teams. It understands enterprise data context through Unity Catalog, enabling it to reason about datasets, lineage, governance policies, and business semantics while working directly inside notebooks, SQL editors, and Lakeflow pipelines.

Genie Code provides a unified agentic development experience for building and operating data pipelines, analytics and AI applications.

With Genie Code, teams can:

• Build and extend data pipelines, dashboards, and ML workflows from natural language prompts
• Debug failures and investigate anomalies across Lakeflow pipelines and models
• Generate notebooks, SQL queries, and visualizations grounded in enterprise data context
• Automate routine operational work such as pipeline monitoring and troubleshooting

By combining deep platform integration with multi-step reasoning, Genie Code allows data teams to move beyond assisted coding toward delegating complex data tasks to an AI partner.

Genie in Databricks One 

Databricks One now includes a unified, multi-agent chat experience powered by Genie, giving business users a simple way to ask questions across their entire data estate. Users can seamlessly access and combine insights from multiple Genie spaces, without needing to know where data lives or which space to choose. When a question goes beyond existing Genie spaces, Databricks One can engage additional agents to explore the data and generate new answers. This allows users to handle both well-defined and on-the-fly questions in a single experience.

Alongside chat, users can search, explore AI/BI dashboards, and interact with Databricks Apps, all within a streamlined interface designed to make data and AI accessible to everyone.

Databricks One Mobile 

Databricks One Mobile brings the new Genie multi-agent chat experience to iOS and Android, enabling business users to securely access and interact with their data from anywhere.

With Databricks One Mobile, users can ask Genie questions, explore AI/BI Dashboards, and access Databricks Apps from their phone. It gives business users a simple way to analyze data and make decisions while on the go.

Why Azure Databricks is the best data and AI platform on Azure

These announcements build on the core strengths that make Azure Databricks the platform of choice for data and AI on Azure.

Unified governance

Unity Catalog centralizes governance across tables, files, models, dashboards and AI assets.

Deep Microsoft integrations

Azure Databricks integrates natively with Power BI, Excel, Teams, Azure OpenAI and other Microsoft services.

Lakehouse-native analytics

Databricks SQL delivers high-performance analytics directly on open lakehouse storage.

AI development

Genie and Agent Bricks provide a unified platform for building and deploying AI applications.

Lower total cost of ownership

Serverless compute, Lakebase scale-to-zero and simplified ingestion reduce infrastructure complexity and cost.

See these innovations at FabCon

If you're attending FabCon 2026 in Atlanta, stop by to see these innovations in action. The Databricks team will be on-site throughout the week demonstrating how organizations are building modern data and AI applications with Azure Databricks. You can also join our session Accelerating Data and AI with Azure Databricks (Thursday, March 19th, 8:00–9:00 AM, room C302) to see how these capabilities come together to accelerate performance, simplify architecture, and maximize value on Azure.

• At FabCon, you can:
  See live demos of Azure Databricks Lakebase, Lakeflow Connect, Genie, AI/BI, and much more.
• Connect with Databricks SMEs to discuss how to unify data engineering, BI and AI on Azure
• Learn how customers are integrating Azure Databricks, Power BI and Azure OpenAI

We’ll also be joining partners across the Microsoft ecosystem for community events during the conference week. Join us and Slalom for a FabCon networking happy hour on Wednesday 3/18 to connect with data leaders and practitioners across the Microsoft and Databricks communities:

FabCon Happy Hour with Slalom → https://go.slalom.com/MSFT-FabCon26

And mark your calendar for Databricks Data + AI Summit, June 15–18, 2026 in San Francisco — the world’s largest conference dedicated to data, analytics and AI, featuring 25,000+ attendees, 800+ sessions and hands-on training across the Databricks Platform.

The future of data and AI on Azure is here — and we’re just getting started!

Get started with Azure Databricks for free →