A real notebook, with an agent inside.
An agent lives inside a real notebook. It plans the analysis, writes the SQL and Python, checks the numbers before it answers — and leaves you a notebook you can edit, rerun and keep.
- Free to start
- SQL, Python and charts
- Nothing to install
An agent that works the way you do.
Say what you want to know. The agent lays out a plan, builds the cells and answers. Follow up and the notebook grows — nothing starts over.
SQL and Python, in the same breath.
Query it in SQL, hand it to Python by name, chart it. No exports, no copy-paste between tools.
SQL
Query any table in your workspace.
- Tables and columns complete as you type
- Every result gets a name
- Rerun a cell; what depends on it follows
Python
Pick up any result by name and keep going in pandas. Nothing to load, nothing to write out.
Charts that belong to the analysis.
Shaped from the data in the notebook, not pasted in. Every chart opens as a table.
Bring the data you have.
Upload a file, connect a database, or start with the sample dataset. Everything in your workspace is ready to query the moment you open a notebook.
CSV
Drop a file in; it is a table you can query right away.
PostgreSQL
Connect a database and work with its tables directly.
MySQL
Same as PostgreSQL: connect once, query its tables in place.
Sample dataset
Start with Telco Churn and see the whole loop in a minute.
Start your first notebook.
Free to start. Bring a file or try the sample dataset.