T
traeai
登录
返回首页
Databricks

Introducing Genie Agent Mode

5.5Score
Introducing Genie Agent Mode
AI 深度提炼
  • Genie Agent Mode 允许 AI 智能体理解并执行自然语言数据任务
  • 该模式集成于 Databricks Lakehouse 平台,强调安全与治理
  • 目标是降低非技术用户使用高级数据分析和 AI 的门槛
#Databricks#Genie#AI Agent#Lakehouse#数据分析
打开原文

Introducing Genie Agent Mode | Databricks Blog

Skip to main content

[![Image 9](blob:http://localhost/c3d26385bd032c882a09c45135533626)](http://www.databricks.com/)

[![Image 10](blob:http://localhost/c3d26385bd032c882a09c45135533626)](http://www.databricks.com/)

  • Why Databricks
  • * Discover
  • Customers
  • Partners
  • Product
  • * Databricks Platform
  • Integrations and Data
  • Pricing
  • Open Source
  • Solutions
  • * Databricks for Industries
  • Cross Industry Solutions
  • Migration & Deployment
  • Solution Accelerators
  • Resources
  • * Learning
  • Events
  • Blog and Podcasts
  • Get Help
  • Dive Deep
  • About
  • * Company
  • Careers
  • Press
  • Security and Trust
  • DATA + AI SUMMIT ![Image 11: Data+ai summit promo JUNE 15–18|SAN FRANCISCO Last chance to save 50% — ends April 30. Register](http://www.databricks.com/dataaisummit?itm_source=www&itm_category=home&itm_page=home&itm_location=navigation&itm_component=navigation&itm_offer=dataaisummit)

1. Blog 2. / Product 3. / Article

  • * *

Contents in this story

How Agent Mode Works

Introducing Genie Agent Mode

Image 12: Databricks "Week of AI Agents" graphic featuring Genie Agent Mode with a branching diagram showing three connected task nodes.

Published: April 17, 2026

Product3 min read

by Hanlin Sun

Share this post

  • [](https://www.linkedin.com/shareArticle?mini=true&url=https://www.databricks.com/blog/introducing-genie-agent-mode&summary=&source=)
  • [](https://twitter.com/intent/tweet?text=https://www.databricks.com/blog/introducing-genie-agent-mode)
  • [](https://www.facebook.com/sharer/sharer.php?u=https://www.databricks.com/blog/introducing-genie-agent-mode)

Keep up with us

Subscribe

#### Summary

  • Agent mode enables users to ask Genie a more advanced class of questions (Why? What if? How could we improve?) and get meaningful answers.
  • Behind the scenes, Agent mode investigates like a data analyst: planning, testing hypotheses, and reasoning across queries to answer your business questions.
  • Agent mode has been tuned to dynamically scale its reasoning to your question’s complexity—this results in faster answers for everyday questions and more rigorous analysis when tackling complex topics.

We are excited to introduce Agent mode in Genie spaces. Our team has developed a powerful agentic process that iteratively plans, explores, and reasons over your data to answer your business questions. As part of Databricks’ Week of Agents, this blog highlights another way we’re changing how organizations interact with their data using agentic AI.

This experience unlocks a much more actionable level of data analysis for everyone in your organization. Now, anyone can get real-time insights into complex business questions like:

  • Why did our churn rate spike in Q3?
  • How can we optimize our campaign spend?
  • What revenue impact should I anticipate if these two supply lines are interrupted?
Image 14: image8

How Agent Mode Works

When you ask a question in Agent mode, Genie doesn’t just return a single query. It investigates the problem like a real data analyst: planning an approach, testing hypotheses, and iterating toward an explanation.

For example, imagine a Genie space for customer support. You notice a spike in reopened cases in December 2025 and ask: “What’s contributing to the spike in reopened support cases?”

Agent mode first confirms the spike, then explores possible contributors such as customers, products, categories, or teams. It uses the business context in your Genie space, including Unity Catalog metadata and author-defined semantics, to focus on the most relevant potential contributing factors.

Image 15:  Genie Agent reasoning trace showing 8 queries executed to analyze the Dec 2025 reopened cases spike.

Agent mode then evaluates these hypotheses by executing multiple queries against the underlying data. The semantics defined in Genie’s knowledge store teach it how to produce accurate queries. We also made Agent mode’s work transparent so users can always verify its accuracy.

During its analysis, Genie continuously reflects on the results of each query and decides what to explore next. In this example, after testing several potential drivers of the spike, the agent decides that it should further investigate if seasonal patterns are contributing to the spike. This iterative cycle of hypothesis generation, querying, and reflection allows Genie to explore data more thoroughly and arrive at a well-supported explanation.

Image 16: image7

After completing its analysis, Genie generates a report of its findings. Following this example, the report first quantifies the increase in reopened cases and then identifies the primary contributors—namely a rise in bug-related cases and performance from the L2 regional team. To support these conclusions, the report also includes visualizations and references to the underlying SQL for users to review.

Image 17: Line chart showing reopened cases trend throughout 2025, peaking at 76 in December.
Image 18: Bar chart showing reopened support cases by category in Dec 2025, with Bug as the top driver at 27 cases.
Image 19: Grouped bar chart comparing reopened cases by team across Oct–Dec 2025, with L1 Global and L2 Regional leading.
Image 20: Genie Agent analysis conclusion with additional context and a recommendation to review case closure procedures.

Depending on the type of question, Genie also provides actionable recommendations on what teams should focus on to improve performance. Users can then share these reports directly in the platform or download them as PDFs to easily distribute insights and collaborate.

**Built for Questions of Any Complexity**

Agent mode not only unlocks advanced business investigations—it improves accuracy across all question types, from simple analytics to multi-step analysis.

Even for straightforward questions, it takes small validation steps to ensure it understands the data before responding. We’ve also tuned the agent to dynamically scale its reasoning to the complexity of the task—moving quickly for simple prompts, and spending more time planning and evaluating for deeper investigations.

The result is faster answers for everyday questions and more rigorous analysis when tackling complex problems.

Users interacting with Genie inside AI/BI Dashboards can also take advantage of this new experience. When you ask questions to Genie from a Dashboard, it leverages Agent mode by default.

GUIDE

Your compact guide to modern analytics

Read now

Image 21: Your compact guide to modern analytics

Get Started Today

Workspace admins can now confirm Agent mode is enabled in the Workspace Previews page. Once enabled, simply turn on the Agent toggle in your Genie spaces and ask your business questions.

Image 23:  Databricks Genie interface showing a tooltip explaining the Agent mode for multi-step data reasoning.

Agent mode is now available, with much more to come including API support and unstructured document analysis. Give Agent mode a try today — we can’t wait to hear what you think.

Keep up with us

Subscribe

Contents in this story

How Agent Mode Works

Recommended for you

Image 24: How to Build Production-Ready Genie Spaces, and Build Trust Along the Way

Product

February 6, 2026/14 min read

#### How to Build Production-Ready Genie Spaces, and Build Trust Along the Way

Image 26: Use Genie Everywhere with Enterprise OAuth

Product

February 19, 2026/13 min read

#### Use Genie Everywhere with Enterprise OAuth

Image 28: Introducing Genie Code

Announcements

March 11, 2026/10 min read

#### Introducing Genie Code

Share this post

  • [](https://www.linkedin.com/shareArticle?mini=true&url=https://www.databricks.com/blog/introducing-genie-agent-mode&summary=&source=)
  • [](https://twitter.com/intent/tweet?text=https://www.databricks.com/blog/introducing-genie-agent-mode)
  • [](https://www.facebook.com/sharer/sharer.php?u=https://www.databricks.com/blog/introducing-genie-agent-mode)

Never miss a Databricks post

Subscribe to our blog and get the latest posts delivered to your inbox

Sign up

*

*

Country*

By clicking “Subscribe” I understand that I will receive Databricks communications, and I agree to Databricks processing my personal data in accordance with its Privacy Policy.

Subscribe

What's next?

More from the Authors

Image 30: Introducing Predictive Optimization for Statistics

Product

November 20, 2024/4 min read

#### Introducing Predictive Optimization for Statistics

Image 32: How to present and share your Notebook insights in AI/BI Dashboards

Product

November 21, 2024/3 min read

#### How to present and share your Notebook insights in AI/BI Dashboards

![Image 34: databricks logo](https://www.databricks.com/)

Why Databricks

Discover

Customers

Partners

Why Databricks

Discover

Customers

Partners

Product

Databricks Platform

Pricing

Open Source

Integrations and Data

Product

Databricks Platform

Pricing

Open Source

Integrations and Data

Solutions

Databricks For Industries

Cross Industry Solutions

Data Migration

Professional Services

Solution Accelerators

Solutions

Databricks For Industries

Cross Industry Solutions

Data Migration

Professional Services

Solution Accelerators

Resources

Documentation

Customer Support

Community

Learning

Events

Blog and Podcasts

Resources

Documentation

Customer Support

Community

Learning

Events

Blog and Podcasts

About

Company

Careers

Press

Security and Trust

About

Company

Careers

Press

Security and Trust

![Image 36: databricks logo](https://www.databricks.com/)

Databricks Inc.

160 Spear Street, 15th Floor

San Francisco, CA 94105

1-866-330-0121

  • [](https://www.linkedin.com/company/databricks)
  • [](https://www.facebook.com/pages/Databricks/560203607379694)
  • [](https://twitter.com/databricks)
  • [](https://www.databricks.com/feed)
  • [](https://www.glassdoor.com/Overview/Working-at-Databricks-EI_IE954734.11,21.htm)
  • [](https://www.youtube.com/@Databricks)
Image 38

See Careers

at Databricks

  • [](https://www.linkedin.com/company/databricks)
  • [](https://www.facebook.com/pages/Databricks/560203607379694)
  • [](https://twitter.com/databricks)
  • [](https://www.databricks.com/feed)
  • [](https://www.glassdoor.com/Overview/Working-at-Databricks-EI_IE954734.11,21.htm)
  • [](https://www.youtube.com/@Databricks)

© Databricks 2026. All rights reserved. Apache, Apache Spark, Spark, the Spark Logo, Apache Iceberg, Iceberg, and the Apache Iceberg logo are trademarks of the Apache Software Foundation.

We Care About Your Privacy

Databricks uses cookies and similar technologies to enhance site navigation, analyze site usage, personalize content and ads, and as further described in our Cookie Notice. To disable non-essential cookies, click “Reject All”. You can also manage your cookie settings by clicking “Manage Preferences.”

Manage Preferences

Reject All Accept All

Image 42: Databricks Company Logo

Privacy Preference Center

Opt-Out Preference Signal Honored

Privacy Preference Center

  • ### Your Privacy
  • ### Strictly Necessary Cookies
  • ### Performance Cookies
  • ### Functional Cookies
  • ### Targeting Cookies
  • ### TOTHR

#### Your Privacy

When you visit any website, it may store or retrieve information on your browser, mostly in the form of cookies. This information might be about you, your preferences or your device and is mostly used to make the site work as you expect it to. The information does not usually directly identify you, but it can give you a more personalized web experience. Because we respect your right to privacy, you can choose not to allow some types of cookies. Click on the different category headings to find out more and change our default settings. However, blocking some types of cookies may impact your experience of the site and the services we are able to offer.

#### Opting out of sales, sharing, and targeted advertising

Depending on your location, you may have the right to opt out of the “sale” or “sharing” of your personal information or the processing of your personal information for purposes of online “targeted advertising.” You can opt out based on cookies and similar identifiers by disabling optional cookies here. To opt out based on other identifiers (such as your email address), submit a request in our Privacy Request Center.

More information

#### Strictly Necessary Cookies

Always Active

These cookies are necessary for the website to function and cannot be switched off in our systems. They assist with essential site functionality such as setting your privacy preferences, logging in or filling in forms. You can set your browser to block or alert you about these cookies, but some parts of the site will no longer work.

#### Performance Cookies

  • [x] Performance Cookies

These cookies allow us to count visits and traffic sources so we can measure and improve the performance of our site. They help us to know which pages are the most and least popular and see how visitors move around the site.

#### Functional Cookies

  • [x] Functional Cookies

These cookies enable the website to provide enhanced functionality and personalization. They may be set by us or by third party providers whose services we have added to our pages. If you do not allow these cookies then some or all of these services may not function properly.

#### Targeting Cookies

  • [x] Targeting Cookies

These cookies may be set through our site by our advertising partners. They may be used by those companies to build a profile of your interests and show you relevant advertisements on other sites. If you do not allow these cookies, you will experience less targeted advertising.

#### TOTHR

  • [x] TOTHR

Cookie List

Consent Leg.Interest

  • [x] checkbox label label
  • [x] checkbox label label
  • [x] checkbox label label

Clear

  • - [x] checkbox label label

Apply Cancel

Confirm My Choices

Allow All

![Image 43: Powered by Onetrust](https://www.onetrust.com/products/cookie-consent/)

Image 44

!Image 45!Image 46

Image 47

问问这篇内容

回答仅基于本篇材料
    0 / 500

    Skill 包

    领域模板,一键产出结构化笔记
    • 论文精读包

      把一篇论文 / 技术博客精读成结构化笔记:问题、方法、实验、批判、延伸阅读。

      • · TL;DR(1 段)
      • · 研究问题与动机
      • · 方法概览
    • 投融资雷达包

      把一条融资 / 创投新闻整理成投资人视角的雷达卡:交易要点、判断、竞争格局、风险、尽调清单。

      • · 交易要点(公司 / 轮次 / 金额 / 投资人 / 估值,材料未明示则写 “未披露”)
      • · 投资 thesis(这家公司为什么值得关注)
      • · 竞争格局与替代方案

    导出到第二大脑

    支持 Notion / Obsidian / Readwise
    下载 Markdown(Obsidian 直接拖入)