AI/BI solution

AI business intelligence grounded in business context.

Ask a business question, explore what changed, and inspect the evidence behind the answer. Datus Agent uses the context your team maintains, including reviewed metrics in DOSI where configured, to help analyze data in your supported sources.

Try Datus Studio See how it works

Datus + DOSI · 1-minute film

What you can do

One shared definition. Different ways to analyze.

Ask a new question, watch a recurring measure, or share a finished report. Each example uses the same reviewed Northstar Premier definition and the same recorded sample data.

Reviewed DOSI definitionCustomer margin ≥ $3,000 · delivered orders ≥ 6 · return rate ≤ 20%

01 / 03

Explore: why does a customer qualify?

See the question, context lookup, and conclusion in one conversation.

Choose an agent

Datus AgentPremier analyst · published Subagent

You

Which customers are Premier, and why does Lucas Martin miss the cut?

Datus AgentAnalysis complete

◕Datus Agent

Maya, Daniel, and Sofia qualify. Lucas meets the margin and order thresholds, but his 33.3% return rate exceeds the 20% limit.

Premier rule reviewedSQL available to inspectNorthstar · DuckDB

Review the rule, source rows, and compiled SQL before sharing the answer.

Ask a follow-up question…

read-onlyConfigured model

Illustrative Datus Agent flow using the recorded Northstar result; no live query runs on this page.

Northstar fictional demo · verified fixture figures · presentation mockups, not live product screens. Agent conversations and third-party client views are illustrative.

Read about agent-built dashboards

The difference · See the product

Follow business context into a reviewable answer.

Datus brings existing business knowledge into the Agent workflow. Select a step to see what someone can inspect in Studio. Where DOSI is configured, reviewed Apache Ossie definitions give supported semantic queries a path to native SQL.

Datus StudioCommerce workspaceD

business-rules.mdUTF-8

1# Business definitions
2 
3Revenue = SUM(orders.amount).
4Group by order status for comparison.
5Use this definition in generated SQL.

status	revenue
completed	350.00
cancelled	100.00

Datus Agent

Show revenue by order status.

business-rules.mdShared with Agent

I found the revenue definition in your shared context. Here is the SQL and its sample result.

context_search_tools

db_tools

revenue.sql
SUM(amount) AS revenue

Ask a follow-up…

Datus Studio

Datus StudioCommerce workspaceD

business-rules.mdUTF-8

1# Business definitions
2 
3Revenue = SUM(orders.amount).
4Group by order status for comparison.
5Use this definition in generated SQL.

status	revenue
completed	350.00
cancelled	100.00

Datus Agent

Show revenue by order status.

business-rules.mdShared with Agent

I found the revenue definition in your shared context. Here is the SQL and its sample result.

context_search_tools

db_tools

revenue.sql
SUM(amount) AS revenue

Ask a follow-up…

Datus StudioCommerce workspace

commerce business-rules.md

# Business definitions

Revenue = SUM(orders.amount).
Group by order status for comparison.
Use this definition in generated SQL.

Shared context: orders · customers · revenue

Illustrative Datus Studio workspace and sample commerce data. This preview does not run a live DOSI query.

Apache Ossie is an open, vendor-neutral specification. DOSI is licensed under Elastic License 2.0; the Datus Agent is open-source. Reusing definitions across supported sources still requires checking mappings and each source’s capabilities.

Explore the semantic layer Explore Datus Studio Read the Dashboard Copilot guide See SQL results and AI charts in VS Code

FAQ

Questions about AI/BI with Datus

What data can Datus analyze?

Datus works with data sources you connect and configure through its supported database adapters. The available analysis also depends on the context, permissions, and tools you give the Agent. Connect a source, review its schema and business definitions, and test the workflow against questions your team actually asks.

Can I inspect the SQL and metric definition behind an answer?

Yes, in supported workflows you can review the generated or executed SQL and its result. Where a DOSI semantic model is configured, you can inspect the compatible metric definition and how a supported semantic query compiles to native SQL. The evidence available in a particular answer depends on the interface and configuration you use.

How do teams check and improve an AI-generated analysis?

Treat the answer as a draft to verify. Review the metric definition, source mapping, SQL, and result, then correct any underlying business definition or query before reusing it. Datus supports this review workflow; it does not guarantee that every generated answer or automatically drafted model is correct.

Does this replace our existing BI dashboards?

It depends on how your team wants to work. You can keep your existing dashboards and use DOSI as a headless, centralized semantic layer to share reviewed metric definitions with supported BI tools. Or you can migrate selected dashboards to HTML reports and dashboards created with Datus, reviewing each result before replacing an established view. Both approaches can work side by side. Interactive dashboards still need a connected data source and backend.

Go deeper

What is agentic analytics?What is conversational BI?Composable analytics and agent-built dashboards

How it works

The Datus workflow, from sources to answers.

Datus brings business knowledge and data sources into a shared context. Studio and Agent use that context; DOSI and Apache Ossie provide reviewable semantic definitions where configured.

Existing knowledge

Documents

SQL & ETL

Warehouses

Datus

Datus Studio

A shared workspace for Agent-driven data engineering.

Datus Agent

A complete data Agent for engineering and analysis.

Ontology + semantic model

DOSI Engine

Execute compatible Ossie definitions as native SQL.

Apache Ossie Spec

An open, vendor-neutral semantic modeling specification.

Put context to work

Teams

AI agents

Applications & BI

Conceptual architecture. The exact steps and available evidence depend on your configured workflow and data source.

Start with a question your team already cares about.

Bring your data and business definitions together, then inspect each step of the analysis.

Try Datus Studio