atomhub
Book a demo ↗
atomiq

Ask your data what happened, and why.

atomiq's agents answer in plain language, explain why a number moved, and trace every number to the query behind it. Then turn the answer into a dashboard or an app. In your own cloud.

Book a demo ↗Watch the 60-second tour
How atomiq compares
How did revenue do last quarter?

Revenue was in Q3, below Q2. North was the only region that grew, up .

Click a highlighted number to see where it came from.

1Revenue by quarter
SELECT quarter, SUM(amount) AS revenue
FROM orders
WHERE year = 2026
GROUP BY quarter
quarterrevenue
Q2$5.26M
Q3$4.82M

Example data

Every number has a source.

Four things atomiq does with your data, every time.

Ask
Plain-language questions. The number comes first, then the chart and a short summary.
Why
When a metric moves, the RCA agent tests the drivers and ranks the causes.
Trace
Every number links to the query it came from. Click it to see the SQL and the rows.
Ship
Turn any answer into a dashboard, or export it as an app your team can open.

From answer to dashboard to app.

  1. 1

    Ask

    Get the answer with its chart, numbers and sources.

  2. 2

    Pin it to a dashboard

    atomiq lays out the panels and keeps them on the same queries.

  3. 3

    Export it as an app

    Share a working app with your team, built on the same queries.

More about dashboards and apps

Quarterly revenue review

Shared with Finance
Revenue, Q3
$4.82M
vs Q2
−8.4%
Orders
61,204
Revenue change by region
North
+6.1%
East
−9.2%
South
−11.8%
West
−12.5%

Example data

Ask why, and get the causes ranked.

The RCA agent breaks a change down by every dimension that matters, tests the likely drivers, and ranks the causes by how much of the change each one explains. It says what it ruled out, and what the data can't prove.

More about RCA
Why did revenue fall 8.4% last quarter?
Ranked causes
  1. 1West: Fewer repeat orders46%1
  2. 2South: Top products out of stock31%2
  3. 3East: Price rise on three products15%3

Ruled out: seasonality (the same quarter grew 3% last year) and missing data (every day is present).

Can't prove: whether competitor pricing played a part.

Share of the drop each cause explains. Example data.

It learns your data before it answers.

No semantic layer to write first. Pick your tables and atomiq builds the knowledge it needs, then checks it.

More about the knowledge base
  1. 1

    Choose the tables

    Point atomiq at the tables your team actually asks about.

  2. 2

    Build

    Every table is read, profiled and described: columns, values, relationships.

  3. 3

    Check

    The knowledge is checked before anyone can ask. Questions open once it passes.

  4. 4

    Add your rules

    Business rules and metric definitions sit on top, so everyone gets the same answer.

See atomiq in 60 seconds.

Ask a question, get the number first, flip the chart to its SQL, and click a number to see its query.

Better with atomhub.

atomiq runs on its own in your cloud. Run it on atomhub, the data operating system, and atomhub takes care of ingestion, storage, governance and the agents underneath — so atomiq can focus on the answers. On its own, atomiq runs read-only, straight on your existing databases.

Works on the data you already have

  • PostgreSQL
  • MySQL
  • MariaDB
  • SQL Server
  • Oracle
  • ClickHouse
  • Snowflake
  • BigQuery
  • Redshift
  • Databricks
  • Athena
  • Trino
How the atomhub platform works
atompulse

AI-powered observability for every log

Log Analyst
atomiq

Ask your data what happened, and why

Data AnalystRCA
atomhubThe data operating system
  • Data plane
  • Execution runtime
  • Governance kernel
  • Observability & lineage
  • Agentic operations

Agents that run it

atomautomateatombuzzatomtuneatomretry
In your cloud
  • AWS
  • Google Cloud
  • Azure
  • On-premises

Questions

Working with logs instead? See atompulse.
What is atomiq?

atomiq is AI analytics for your own data. Specialist agents answer questions in plain language, explain why a metric moved, and trace every number to the query it came from. Answers turn into dashboards and apps, and it runs in your own cloud.

How do I know an answer is right?

Every number in an answer links to the query that produced it, so you can click it and see the SQL and the rows. The RCA agent also lists what it ruled out and what the data can't prove.

What's the difference between the Data Analyst and RCA agents?

The Data Analyst answers what happened: a number, a chart and a summary. The RCA agent answers why: it breaks a change down, tests the likely drivers and ranks the causes in a report.

Do I have to set up a semantic layer first?

No. Pick the tables you want and atomiq builds a knowledge base: every table is read, profiled, described and checked before questions open. You can add business rules and metrics on top.

Can I share what I find?

Yes. Turn any answer into a dashboard, or export it as an app your team can open.

Where does atomiq run?

Inside your own VPC on AWS, Google Cloud or Azure, or on-premises. Run it on atomhub, or on its own, read-only, straight on the databases you already have. No migration, and your data stays where it is.

atomiq

See it on your own data.

Bring a question you've been meaning to ask. We'll answer it live, with every number traced.

Book a demo ↗How atomiq compares