Cube is an agentic analytics platform built on a semantic layer that provides AI-native business intelligence and embedded analytics. By creating a single, governed semantic model, Cube ensures that all data points, calculations, and metrics remain consistent across different surfaces, including chat interfaces, dashboards, workbooks, and external tools like Slack or LLM-based agents. Created to help teams provide trusted answers, the platform addresses the challenges of fragmented data and manual reporting by offering a central source of truth for all analytical needs.

The platform enables organizations to build AI-powered data experiences that are governed, accurate, and scalable. It provides the necessary infrastructure to integrate analytics directly into applications, allowing developers to create custom AI agents, iframes, or programmatic interfaces for data access. Cube simplifies the complex tasks of data modeling, caching, and security enforcement, ensuring that insights are not just visualized but also explained through natural language capabilities.

Some of the key features are:

• Semantic Layer: A unified model that grounds AI agents and BI tools with consistent definitions for metrics and dimensions.
• Analytics Chat: A conversational interface that allows users to ask questions in plain English and receive accurate, governed answers.
• Embedded Analytics: Flexible deployment options including iframes, APIs, and MCP connectors for seamless integration into customer-facing products.
• Multi-tenancy & Governance: Built-in row-level security and access controls that ensure data isolation and consistent permissions across all tenants.
• Caching & Query Performance: Advanced caching mechanisms and pre-aggregations that ensure high performance even under heavy load and with large datasets.
• AI Agentic Workflows: Support for multi-step reasoning, agent skills, and contextual insights powered by a robust AI context layer.
• dbt Integration: Seamless connectivity with existing dbt projects, allowing teams to leverage their current data modeling investments directly within Cube.

Cube operates by acting as a semantic bridge between your data sources and the presentation layer. You define your data models within the platform—specifying metrics, dimensions, and access policies—which are then utilized by the AI agent to translate natural language queries into efficient, governed SQL queries. The platform manages the entire lifecycle of data exploration, from authentication and security enforcement to result visualization and export, ensuring that every user, whether an internal analyst or an external customer, receives the same validated output.

Some common use cases include:

• Customer-facing Analytics: SaaS companies use Cube to embed AI-powered reporting and dashboards directly into their products, providing customers with self-serve insights without exposing raw database complexity.
• Automated QBR Generation: Customer Success teams use Cube to automate the creation of data-driven quarterly business reviews, shifting from manual spreadsheet work to intelligent, AI-generated storytelling.
• AI-Native Business Intelligence: Data teams use Cube to ground LLMs in their corporate data, allowing employees to query data via Slack or chat interfaces while maintaining strict governance and data consistency.
• Unified Metric Definitions: Large organizations use Cube to ensure that core KPIs like 'Monthly Active Users' or 'Lifetime Value' are defined once and accurately reported everywhere, from BI tools to executive dashboards.