# What is Flax?

> A warehouse-native BI platform — a git-backed semantic layer, exploration, spreadsheet modelling, dashboards, and an AI assistant, all governed as code.

*[View this page in the Flax docs](https://flax-analytics.com/docs/getting-started/what-is-flax)*

Flax is a **warehouse-native business-intelligence platform**. You point it at the
data warehouse you already have — Postgres, Snowflake, Databricks, or BigQuery —
define your metrics once in a git-backed semantic layer, and everyone in your
organization explores, charts, and shares from that single source of truth.

The design goal is to combine the *consistency* of a modeled semantic layer with
the *agility* of ad-hoc SQL and spreadsheets, without the per-viewer pricing that
makes traditional BI expensive to roll out.

## How Flax fits together

Flax is a pipeline from raw tables to governed answers. Each stage builds on the
one before it:

1. **Connections** point Flax at your warehouse (or a CSV upload). Credentials are
   encrypted at rest and every outbound connection passes an SSRF egress guard.
2. **The semantic model** is dbt-compatible YAML — dimensions, measures, and joins
   defined once. Because it lives in git, every change is reviewed and versioned.
3. **Explore** turns the model into point-and-click queries. Flax compiles them to
   dialect-correct SQL and pushes the work down to your warehouse.
4. **Sheets** let you model on top of query results with a familiar spreadsheet,
   including real-time multiplayer editing.
5. **Charts and dashboards** visualize results with cross-filtering, drill-down,
   and a bespoke chart renderer.
6. **Actions, deliveries, and the AI assistant** automate and accelerate the work.

## What makes Flax different

> [!NOTE]
> Every metric is **defined once** as version-controlled YAML and then explored,
> charted, and shared everywhere — governed by row-level security in every surface.

- **Governed as code.** Semantic changes flow through GitHub pull requests, so BI
  logic gets the same review as application code.
- **Warehouse-native.** Queries compile per-dialect and run in your warehouse; Flax
  adds an embedded DuckDB engine only for spreadsheet modelling and blending.
- **Programmable.** A REST API, an MCP server, and embeddable charts mean Flax is a
  building block, not a walled garden. See the [Developer guide](/docs/developers/overview).

## Where to go next

- New to Flax? Follow the [analyst quickstart](/docs/getting-started/quickstart-analysts).
- Just need to read dashboards? See the [viewer quickstart](/docs/getting-started/quickstart-viewers).
- Setting up a team? Start with the [admin quickstart](/docs/getting-started/quickstart-admins).
- Building on the API? Jump to the [developer quickstart](/docs/getting-started/quickstart-developers).
- Unsure what a term means? The [core concepts glossary](/docs/getting-started/core-concepts) defines every noun used in these docs.