What is Airbrx?

Airbrx is an intelligent data gateway. Your BI tools point at Airbrx instead of your data warehouse, and queries that have been answered before come back instantly from cache — while everything else flows through to the warehouse, untouched. No client refactor, no SQL rewrites, no new credentials.

Drastically lower your warehouse compute

Databricks and Snowflake both bill for compute time. Every wake of a SQL warehouse, every minute a cluster runs, every metadata round-trip — they all show up on your bill. But the BI traffic that drives most of those bills isn't unique work; it's the same dashboards refreshing, the same reports re-running, the same handful of users asking the same questions. Airbrx serves those repeats from cache at a fraction of a cent each. The warehouse stays asleep. Your bill drops.

What changes for the people using your data

Almost nothing. The connection host changes — a Tableau dashboard, Power BI report, DBeaver session, or Python script that used to connect to your-warehouse.snowflakecomputing.com now connects to your-slug.gateway.airbrx.ai instead. Same SQL dialect, same authentication, same response shape. Users don't need to learn anything new — they just notice that their dashboards load faster.

The trick is that Airbrx doesn't replace the warehouse. It sits in front of it, catching the queries it can answer from cache and forwarding the rest. Your warehouse stays directly accessible for everything that needs to bypass the cache: scheduled jobs, dbt builds, native dashboards, schema management.

What you get

Airbrx is built around four product goals — the Four Pillars. Each one shows up somewhere concrete:

• Platform Savings. Cached queries don't wake your warehouse. For read-heavy workloads — dashboards, recurring reports, BI exploration — the savings compound quickly.
• Consumer Experience. Cached responses return in milliseconds instead of seconds. The same query, the same result, only faster — with no change to the workflows your users already have.
• Granular Control. Fine-grained cache configuration paired with fine-grained rule-busting — a combination most data platforms don't offer. A rules engine decides what gets cached, for how long, how it's keyed, and which writes should bust which cached results. You can express things like "cache aggregate queries on the orders table for one hour, keyed per user, but invalidate immediately when an INSERT lands on that table" — and have the Gateway honor it on the next read.
• Secure & Compliant. Credentials pass through; Airbrx never stores or logs them. Row-Level Security and column masking continue to be enforced by your warehouse. Audit trails stay intact.

How it's put together

Airbrx is three resources working together:

• The App at app.airbrx.ai — where you sign in, configure tenants, author rules, and watch traffic.
• The API at api.airbrx.ai — the source of truth. Stores tenants, rules, and access; issues tokens; serves analytics.
• The Gateway at <slug>.gateway.airbrx.ai — the traffic path your data tools connect to instead of the warehouse.

For a more careful walk through the relationships, see The Gateway, App, and API.

What it supports today

Airbrx works with the warehouses and BI tools most data teams already use:

• Warehouses — Databricks (SQL Warehouses) and Snowflake.
• BI & analytics tools — Tableau, Power BI, Looker, DBeaver, DataGrip, dbt, and anything that connects via JDBC or ODBC.
• Custom code — Python (snowflake-connector-python, databricks-sql-connector, SQLAlchemy), R, JDBC for Java callers.

For per-tool walkthroughs — exact connection-string deltas, verification steps, and gotchas — see the connection recipes.

Where to go next

• The Gateway, App, and API — name App, API, Gateway and what each one owns.
• How the Gateway works — the request lifecycle in detail.
• The Four Pillars — the product goals that shape every feature.