The #1 open-source
analytics agent

Your data team owns the context layer.

Everyone in the company can build on data.

Sign Up

Backed byYCombinator

Sarah, what do you want to analyze?
Ask anything about your data...
Claude Opus 5.5
Suggested ideas based on your data set
Which customers churned last quarter?

They use nao every day

Create your own Agentic BI

Build your data context layer

Sync your warehouse, dbt and docs into files your agent reads. Context work takes an agent from 15% to 85% reliability.

15%
No context
30%
nao sync
45%
Add your docs
65%
Rework data model
85%
Rework context
Source →

Let your whole company analyze data

From anywhere: nao UI, MCP, Slack, Teams, WhatsApp.

app.getnao.io
New chat Search chats Stories FeedChats Starred Usage report1h agoChats Weekly business review1h agoRevenue trend2h agoChurn by plan5h agoTop accounts Q31d agoSignups by country1d agoSupport tickets trend2d agoSSarahsarah@acme.com
Revenue trend Share chat
Show me monthly revenue and the running total
Explored 1 file
SQLMonthly revenue10 rows
Revenue trend by monthAll data
0K0K20K125K40K250K60K375K80K500K2026-01-012026-03-012026-05-012026-07-012026-09-01
Revenue Cumulative revenue
Ask anything about your data...
Claude Opus 5.5
9:41
naoonline
Today
Churn this month?09:12
2.1%, down from 2.6% in September. Fewer cancellations on annual plans.09:12
Send me a chart of daily signups09:13
Daily signups, last 10 days
Up 24% on the previous 10 days.09:13
Message
Teams
top 5 customers last month?
nao 2:50 PM
Explored 1 query
Summary of findings
Northwind · Globex · Initech · Umbrella · Stark Ind.
Claude · Used 5 tools ›
Widget from nao display_chart ⌄
Weekly signups Open in nao
nao app
Ask anything about your data...
Claude Opus 5.5
Thread
claire @nao chart of weekly signups
nao APP
Explored 2 files, 1 query.
Weekly signups

Let anyone build their own Stories

Revenue Trend Open chat
Jan - Oct 2026 All countries
Revenue YTD
$4.57M
↑ +12%
Customers
1,284
↑ +6%
Net retention
118%
↑ +3 pts
Monthly revenue and running total
JanFebMarAprMayJunJulAugSepOct0350K700K02.5M5M
Monthly revenue Running total
countryrevenue_usdgrowth_yoy
1France1,210,400+18.2%
2United States1,043,900+9.4%
3Germany718,250+12.1%

Monitor that your agent always gets it right

Evaluate offline

nao test runs your key questions against the SQL you expect and scores every context change before it ships.

$ nao test
✓ last_month_mrr
✓ revenue_by_country
✓ active_users_28d
✗ churned_mrr
...
Summary
TestsSuccess %Always %CostTime (s)
2491.7%87.5%$0.41212.4

Observe online

Chat replay shows every conversation. Context recommendations turn errors and downvotes into fixes, opened as pull requests.

nao · Context Recommendations
Context Recommendations
Diagnostic suggestions for improving this project's context, mined from real usage.
Run now
ConfigRecommendations 12Applied 30Dismissed 8
MRR excludes annual plans paid upfrontSemantic missing

Three chats summed monthly invoices only. The agent read amount_usd as monthly while annual plans are billed once.

432
Detected 2d ago 1 suggested change +3 -1 Create PR
Orders lookup filters on a column that does not existTool errors

The whole core product is Apache 2.0. Self-host it, read the code, and shape the roadmap with the community.

Stack agnostic

Contribute →

nao works with all warehouses, data stacks, and LLMs. Anything missing? Contribute to build it!

BigQuerySnowflakeDatabricksPostgreSQLRedshiftAthenaDuckDBMotherDuckClickHousedbtCubeAirbytedltDagsterAirflowNotionConfluenceGoogle DriveGitHub

The most secure BI option

Your models, your infrastructure, your permissions.

Trust Center →Enterprise features →

Bring your own LLM key.

Any provider at your own price, with budgets per user and per team.

AnthropicOpenAIGoogle GeminiMistralAmazon BedrockOllama

Deploy in your own infrastructure.

Self-host on your own cloud, or run on nao Cloud.

Your SSO and warehouse permissions.

Okta, Google, Microsoft Entra ID or any OIDC. Row-level security and identity passthrough.

OktaGoogleMicrosoft

SOC 2 Type II.

Audited controls. Your data is never used to train models.

SOC 2 Type II

Get started

Sign Up
Free playbook

Learn Agentic Analytics

Take an analytics agent from POC to production, with a tool benchmark, a context engineering method and examples from 7 real companies.

Agentic Analytics Playbook