Documentation

Build context like a file system

Everything you need to connect your data, engineer a reliable context tree, and deploy a chat UI your whole team can trust.

Quickstart

01

Connect a warehouse

Point nao at Postgres, Snowflake, or BigQuery with a read-only role. nao introspects your schema to bootstrap a starter context tree.

02

Engineer your context

Define metrics, business rules, and schemas as discrete, versioned files. Organize them into folders the same way you would a codebase.

03

Deploy the chat UI

Publish a chat interface where anyone on your team can ask questions in plain English and get answers grounded in your context.

example context tree
context/
├── metrics/
│   ├── revenue.md
│   ├── active_users.md
│   └── churn_rate.md
├── schemas/
│   ├── orders.sql
│   └── subscriptions.sql
├── rules/
│   ├── fiscal_calendar.md
│   └── data_freshness.md
└── glossary.md

Browse the docs

Guides

Step-by-step walkthroughs for defining trustworthy metrics, scoping context by role, and rolling out a shared chat UI to non-technical teammates.

Context versioning

Track every change to your context tree. Roll back definitions, review diffs, and keep answers consistent as your business evolves.

Integrations

Connect Postgres, Snowflake, BigQuery, and dbt. Sync semantic models and reuse the metric definitions you already maintain.

Chat UI

Embed or host the chat experience. Configure permissions so each viewer only sees the context they are allowed to query.

API reference

Programmatic access to context files, queries, and answer history. Build nao into your own products and workflows.

Examples

Reference context trees for SaaS, ecommerce, and fintech teams that you can clone and adapt to your own data.