Genie Agents

What it does

Genie Agents takes plain-English questions from business users and answers them using data registered in Databricks Unity Catalog, including structured tables and unstructured files in Unity Catalog volumes. Responses can include analysis and visualizations, and an Agent mode handles more exploratory questions with multi-step reasoning and hypothesis testing.

Subject matter experts set up each agent for a specific domain by adding example queries, text snippets, sample questions, a knowledge store of semantic definitions, and registered SQL functions. Teams can test agents with accuracy benchmarks, collect thumbs-up/down feedback, and monitor responses. Users can chat with an agent directly, ask follow-up questions inside AI/BI Dashboards, or reach agents through the Genie Agents API, which can be added to apps such as Microsoft Teams, Slack or Glean.

It suits business and nontechnical staff who want answers without waiting on analysts, plus the data experts who curate the agents. It is available to existing Databricks Lakehouse (Pro and Serverless) customers and requires data managed in Unity Catalog. Pricing follows Databricks DBU rates. Answer quality depends on how well an agent has been curated and benchmarked, and answers respect each user's existing data permissions.

What it can do

• Natural-language Q&A over Unity Catalog tables and volume files
• Expert-curated instructions, sample questions and knowledge store
• Agent mode for multi-step reasoning and hypothesis testing
• Accuracy benchmarks, feedback and monitoring
• Follow-up questions inside AI/BI Dashboards
• Genie Agents API and Agent mode APIs for embedding in apps

Works with

• Unity Catalog
• AI/BI Dashboards
• Microsoft Teams
• Slack
• Glean

Example workflows

Self-serve sales question

Input: a curated sales agent over Unity Catalog tables. Steps: a user asks for top sales reps by pipeline value, reviews the answer and chart, then asks follow-ups. Output: an insight within the user's data permissions.

Embed an agent in a chat app

Input: a configured Genie agent and API access. Steps: connect the Genie Agents API to an app such as Slack or Teams; users ask questions there. Output: answers returned inside the business app.

Before you choose

• Requires data in Unity Catalog and a Databricks Lakehouse workspace.
• Accuracy depends on expert curation; benchmarks and feedback are provided to check it.
• Cost is based on Databricks DBU usage, not a flat per-seat price.

Listed under

• Data

Get started

Express-setup free trial with a sample bakery dataset

1. Confirm your data is managed in Unity Catalog on a Databricks Lakehouse (Pro or Serverless) workspace, or start the express free trial.
2. Create a Genie agent and add example queries, sample questions and a knowledge store.
3. Run accuracy benchmarks, then share the agent or connect it via the API.

Pay-as-you-go by DBU; see the Databricks rate tables for current rates. A free trial exists but its length and credits are not confirmed.

Source: databricks.com/product/genie/agents, checked 28 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Genie Agents: Conversational Analytics for Business Teams ↗www.databricks.com
• Databricks Pricing: Flexible Plans for Data and AI Solutions | Databricks ↗www.databricks.com
• Databricks documentation | Databricks ↗www.databricks.com

Company website

databricks.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.