> ## Documentation Index
> Fetch the complete documentation index at: https://honeydew.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Building with AI

Honeydew provides tools for integrating AI into your data workflows.
You can use AI coding agents to build and maintain your semantic model,
query data with natural language, and embed AI-powered analytics into your applications.

## Use Cases

### Build a Semantic Model with AI

Use an AI coding agent to create and manage your semantic model
through natural conversation — all without leaving your IDE.

For example:

* Bootstrap a new semantic model in minutes by importing warehouse tables
  and defining entities, relationships, and metrics
* Export semantics from an existing semantic model, query, or data source
* Quickly add and validate new metrics and attributes
* Ask data questions and iteratively add and refine missing semantics

For setup instructions, see the
[MCP Server](/integration/mcp) page.

### Deep Analysis

Use the deep analysis capability for multi-step analytical workflows.
The AI agent breaks down complex questions, runs multiple queries,
and synthesizes the results into a coherent answer.

For example:

* "Why did revenue drop last quarter? Break down by region and product category."
* "Compare customer retention rates across acquisition channels over the past year."
* "Which product lines are underperforming relative to their growth targets, and why?"
* "There was a spike in returns last week. Identify the root cause and affected segments."

See [Deep Analysis](/integration/context-layer/deep-analysis) for how it works
and available interfaces.

### Embed AI Analytics in Applications

Build AI-powered data applications
using the Honeydew
[AI API](/integration/context-layer/api).
The API supports natural language questions,
multi-step deep analysis,
and follow-up conversations —
all grounded in your semantic model.

You can integrate Honeydew AI into:

* Custom chatbots and assistants
* Internal data portals
* Slack and Teams workflows
  ([Slack App](/integration/context-layer/slack-app),
  [Teams App](/integration/context-layer/teams-app))
* Streamlit dashboards ([Streamlit](/integration/context-layer/streamlit))

## Getting Started

To start building with AI:

1. **Connect an AI coding agent** —
   Set up the
   [MCP Server](/integration/mcp#supported-clients)
   in your preferred client (Claude Code, Cursor, VS Code, and more)
2. **Install plugins** —
   Add the
   [Honeydew plugins](https://github.com/honeydew-ai/honeydew-ai-coding-agents-plugins)
   for guided semantic modeling
   and data analysis workflows
3. **Explore your data** —
   Ask questions, build your model,
   and query data through natural conversation