🤖AI Assistant
The Euno AI Assistant provides an intuitive way to interact with your data model using natural language. It can help you explore your data assets, answer questions about your metadata, and perform complex queries without having to learn EQL syntax.
What you'll learn
How to use the AI Assistant to explore your data model
Understanding the assistant's capabilities and limitations
Examples of queries you can ask
How the assistant interprets your data model
Getting Started
Accessing the AI Assistant
The AI Assistant is available in the main navigation sidebar under "AI assistant" (marked with a Beta badge). Click on it to open the assistant interface where you can start asking questions about your data assets.
Note: The AI Assistant feature must be enabled for your account. If you don't see it in the navigation, contact your administrator.
Note: The AI Assistant can also be used as a tool for your existing agent (Claude, goose, Cursor, Snowflake Intelligence, etc.). See MCP Setup Guide for integration instructions.
What the AI Assistant Can Do
🔍 Resource Discovery
Find tables, views, dashboards, and other data assets by name or description
Discover resources based on properties like owner, tags, or usage patterns
Locate resources within specific projects, schemas, or folders
📊 Data Lineage and Relationships
Understand upstream and downstream dependencies between resources
Find what tables feed into a specific dashboard
Discover which dbt models are used by Tableau workbooks
Trace data flow through your analytics stack
📈 Usage and Performance Analysis
Identify most and least used resources
Find resources with high query costs or long build times
Discover abandoned or underutilized assets
Analyze resource performance metrics
🏷️ Metadata Exploration
Search by custom properties and tags
Find resources by business domain or data governance status
Explore resource descriptions and documentation
🔗 Cross-Platform Analysis
Understand connections between dbt, Tableau, Looker, and Snowflake resources
Find resources that span multiple platforms in your data stack
How to Use the Assistant
Natural Language Queries
Simply ask questions in natural language. The assistant understands various query patterns:
Finding Resources:
"Show me all tables in the sales schema"
"Find dashboards created by John Smith"
"What are the most popular dbt models?"
Understanding Relationships:
"What tables feed into the Revenue dashboard?"
"Show me all dbt models that use the customers table"
"Which Tableau dashboards depend on the orders model?"
Performance and Usage:
"Which tables have the highest query costs?"
"Show me unused dashboards from last month"
"Find dbt models with the longest build times"
Governance and Quality:
"Show me tables without descriptions"
"Find resources tagged as 'critical'"
"What data sources are not connected to dbt?"
Resource Context
The assistant maintains context during your conversation:
When you mention "this table" or "these dashboards," it understands which resources you're referring to
You can build upon previous queries without repeating resource names
The assistant remembers resources you've discussed throughout the conversation
Capabilities and Limitations
✅ What the Assistant Excels At
Converting natural language to precise EQL queries
Finding resources across your entire data stack
Understanding complex relationships between resources
Providing detailed property information
Maintaining conversation context
⚠️ Current Limitations
Beta Feature: The assistant is in beta and capabilities are evolving
🎯 Best Practices
Be Specific: Include resource types, names, or properties when possible
Use Exact Names: Quote exact resource names for more precise results
Ask Follow-ups: Build on previous queries to explore related resources
Check Results: Review the generated EQL to understand how your query was interpreted
Iterate: If results aren't what you expected, rephrase your question
Example Conversations
Finding High-Impact Resources
You: "Show me the most used Tableau dashboards"
Assistant: I'll search for Tableau dashboards ordered by usage metrics...
[Returns list of dashboards with total_impressions_60d data]
You: "Which dbt models feed into the top dashboard?"
Assistant: I'll find the upstream dbt models for [Top Dashboard Name]...
[Shows data lineage from dbt models to the dashboard]Performance Analysis
You: "Find dbt models with long build times"
Assistant: I'll look for dbt models with high average_build_time_60d values...
[Returns models sorted by build time]
You: "Do any of these models have downstream Tableau usage?"
Assistant: I'll check which of these models are connected to Tableau resources...
[Shows which slow models are actually being used in BI]Data Governance
You: "Show me tables that aren't governed by dbt"
Assistant: I'll find tables that don't have upstream dbt model definitions...
[Returns tables not defined by dbt resources]
You: "Which of these are being used in dashboards?"
Assistant: I'll check which ungoverned tables have downstream dashboard usage...
[Shows governance gaps in actively used resources]Getting Help
If you encounter issues or have questions:
Try rephrasing your question with more specific terms
Check the generated EQL query to understand how your request was interpreted
Use the conversation context by referring to previously discussed resources
Contact your Euno administrator for account-specific questions
The AI Assistant is designed to make data discovery and analysis more accessible. Experiment with different types of questions to discover insights about your data ecosystem!
Last updated