SpotterModel
AI agent · Data & analytics agent · Data & analytics by ThoughtSpot
Automate governed data modeling with AI-driven recommendations and logic.
SpotterModel is an AI data modeling agent for data teams that automates the creation of governed, AI-ready data foundations. It recommends columns and explains table joins to turn raw data into trusted models for conversational analytics.
Agent specs
The full spec card (context, pricing per token, modalities and more) on AntSpace. Sign up free to see the full spec card.
- Languages
- English
Category: Data Governance
- Integrations
- Databricks Metric View, Snowflake, BigQuery, dbt, Git
- Data sources
- Snowflake, Databricks, BigQuery, Redshift
- Deployment
- Cloud, Multi-cloud, On-premises
- Languages
- English, German, Japanese, French
- Primary Function
- Automated data modeling and relationship mapping
- Supported Platforms
- Snowflake, Databricks, BigQuery, Redshift
- Automates raw table to governed model conversion in minutes
- Recommends relevant columns based on intended business questions
- Explains the logic behind every table connection and join
- Maintains row and column-level security at the source
Overview
SpotterModel is an AI-powered data modeling agent designed for data engineers and analysts to accelerate the creation of governed data foundations. It automates the process of turning raw data tables from warehouses like Snowflake and Databricks into structured, AI-ready models. The agent assists users by recommending relevant columns and explaining the logic behind table joins based on the business questions the model must answer. By defining security and logic at this layer, SpotterModel ensures that all downstream conversational analytics and dashboards remain accurate and compliant. It integrates with the ThoughtSpot semantic layer and supports importing definitions from external tools like Databricks Metric View.
Data Modeling Automation
- Column Recommendations: SpotterModel recommends the most relevant columns for a data model based on the specific business questions the model is intended to answer.
- Relationship Explanation: The agent explains the rationale behind every table connection and join, ensuring data engineers understand the underlying logic.
- Automated Modeling: SpotterModel automates the transition from raw data tables to structured, governed data models in minutes.
- AI-Ready Optimization: The agent ensures that data models are optimized with the necessary metadata and logic to power natural language querying.
Governance and Logic
- Semantic Layer Integration: SpotterModel works with the ThoughtSpot semantic layer to define metrics, joins, and business logic once for use across all agents.
- External Metric Integration: Users can import semantic definitions from external sources like Databricks Metric View to maintain a single source of truth.
- Governed Foundations: The agent maintains row-level and column-level security at the modeling layer, which is inherited by all downstream analytics.
- Modeling as Code: Data models can be exported as TML files and version-controlled in Git for deployment through existing pipelines.
What it does
- Recommend relevant columns for data models
- Explain table connections and joins
- Automate raw table to model conversion
- Import semantic definitions from external sources
- Maintain row and column-level security
- Export models as TML files for version control
Use cases
- Data engineers use SpotterModel to quickly turn raw Snowflake tables into governed models for business teams.
- Analysts use the agent to understand complex table relationships through natural language explanations of joins.
- Data leaders import existing semantic definitions from Databricks to maintain a single source of truth in ThoughtSpot.
- Developers export AI-generated models as TML files to version-control their analytics infrastructure in Git.
Specifications
- Integrations
- Databricks Metric View, Snowflake, BigQuery, dbt, Git
- Data sources
- Snowflake, Databricks, BigQuery, Redshift
- Deployment
- Cloud, Multi-cloud, On-premises
- Languages
- English, German, Japanese, French
- Primary Function
- Automated data modeling and relationship mapping
- Supported Platforms
- Snowflake, Databricks, BigQuery, Redshift
- Governance Layer
- ThoughtSpot Semantic Layer
- Security Model
- Row and column-level security
- Integration Support
- Databricks Metric View, dbt, Git
- Output Format
- TML (ThoughtSpot Modeling Language) files
- User Interface
- Natural language prompt and visual model editor
How it works
- Connect Data Sources: Connect the agent to your cloud data platform such as Snowflake, Databricks, or BigQuery.
- Define Model Goals: Describe the business questions the model needs to answer in natural language.
- Review and Refine: Review AI-recommended columns and join logic, then approve the generated data model.
- Deploy to Production: Deploy the model to the semantic layer where it powers conversational search and Liveboards.
FAQ
What is SpotterModel?
SpotterModel is an AI data modeling agent that automates the creation of governed data foundations by recommending columns and explaining table relationships.
How does SpotterModel simplify data modeling?
It recommends columns and explains joins based on the questions the model must answer, turning raw tables into structured models in minutes.
Does SpotterModel support data security?
Yes, it maintains row-level and column-level security at the modeling layer, ensuring users only see data they are authorized to access.
How does it work with the semantic layer?
SpotterModel integrates with the ThoughtSpot semantic layer, allowing you to define metrics once and deliver them everywhere.
Added by AntSpace from public sources (thoughtspot.com).