Use cases

Cross-system reporting without a warehouse#

A retailer keeps orders in SQL Server, customers in PostgreSQL and product data behind a REST catalogue API. One live SQL query joins all three for a dashboard. There is no nightly ETL job, and the report never shows yesterday's data.

Distributed business transactions#

"Place an order, deduct inventory, notify the shipping partner's API" becomes one saga definition. If the shipping call fails, the inventory deduction and the order are reversed automatically, with no reconciliation script and no half-finished order.

One connection for every analytics tool#

Tableau, Power BI, psql, DBeaver, a Python notebook using Arrow Flight and a Spark job can all point at the same Kiwi Fabric instance. They see the same governed, federated view of the data through the protocol they already speak. See Connect BI and analytics tools.

Governed self-service analytics#

Give an analytics team broad query access to a federated view of sensitive data, while row filters restrict each analyst to the accounts or regions they are authorised for, and column masks hide personal data from anyone without explicit permission. The rules live in one place instead of being copied into every tool.

Natural-language querying over live data#

Point an AI assistant (OpenAI, Azure OpenAI, Anthropic, Google, a local Ollama model or any OpenAI-compatible endpoint) at Kiwi Fabric so business users can ask questions in plain English. Because requests run through the same access layer, the assistant can never surface a row or column the asking user could not already see.

Incremental modernisation#

A team moving off a legacy system does not need a single cutover. Kiwi Fabric can federate queries and writes across the old and new systems during the transition.

Prioritising rollback work#

The built-in compensation gap reports show which write paths lack automatic rollback, so platform teams can prioritise the work that matters most.