Tukun.ai Documentation

Tukun.ai is built for teams that want faster analysis without giving up reviewability. The product helps you move from a business question to a checked result, then turn that result into reusable operational assets.

Read by goal

I need to understand the product quickly

• Start with Overview.
• Then read How to Evaluate Tukun.ai.

I want to get from sign-in to a trustworthy answer

• Read Quickstart.
• Then follow First Trusted Answer.
• Use Review Results before sharing anything externally.

I want to onboard real business data

• Read Connect Data.
• Check Supported Sources.
• Review Security Overview.

I want consistent answers across repeated questions

• Read Core Concepts.
• Then read Modeling Guide and When to Model.

I want to operationalize analysis for the team

• Read From Question to Dashboard.
• Then use Create Dashboards.

The core product loop

Most teams adopt Tukun.ai through the same loop:

1. Connect one approved data source.
2. Ask one high-value business question in the Workbench.
3. Review the generated result and its evidence path.
4. Fix ambiguity by clarifying the question or modeling the metric.
5. Save stable outputs as cards and dashboards.
6. Reuse the same definitions for the next decision cycle.

What these docs optimize for

These docs are written for customer onboarding and everyday product use. They focus on product logic, workflow decisions, and practical operating guidance rather than low-level APIs or SDK references.

If you are comparing tools, the most important difference to understand is this: Tukun.ai is not a generic chat UI over data. It is a governed analysis workflow with an explicit review step and reusable semantic assets.