Skip to content

Analysis notebooks for data and development teams

From raw data
to a reasoned answer.

Query with SQL, explore with Python, and explain the result in one notebook. Built for data and development teams who want to see how an answer was reached.

Open Quantum LabWatch the film

SQL, Python, and your connected data in one team notebook.

Queued runs, notebook variables, and file handling are upcoming and not yet deployed.

Follow the question through SQL, Python, notes, and charts.28 sec · English captions

Why Quantum Lab

The analysis and its explanation, together.

01

Work across SQL and Python.

Query your connected data, reshape it with Python, and turn the results into tables and charts.

SQL + Python + DataFrames

02

Leave a trail others can follow.

Keep assumptions, intermediate results, and written findings beside the code that produced them.

An ordered notebook for the team

03

Ask Kole for a hand.

Get help building the analysis. Review configured consequential proposals before the agent continues.

AI assistance with human review

A closer look at Quantum Lab.

Product preview · Sample data
Illustrative interface

Notebook workspace

Weekly account review

No live execution
Notebook3 ordered blocks
Preview only
SQL query

Account activity by week

Connection: warehouse
SELECT
  date_trunc('week', created_at) AS week,
  COUNT(*) AS active_accounts
FROM public.accounts
WHERE created_at >= CURRENT_DATE - INTERVAL '12 weeks'
GROUP BY 1
ORDER BY 1;
Python

Compare the SQL DataFrame

Input: SQL DataFrame
weekly = sql_result.assign(
    change=lambda frame: frame.active_accounts.diff()
)
weekly[["week", "active_accounts", "change"]]
Markdown

Decision note

Written context

Review the weekly change from the SQL and Python DataFrames before deciding whether to prepare a follow-up for the account team.

Decision

Keep the source query and comparison visible with the note.

Practice the underlying skill

Work through a SQL example.

A useful analysis starts with checks you can explain and repeat. Work through a small data audit before bringing the same questions to your notebook.

Runs locally in DuckDB. Fictional data. No signup required.

Build the analysis someone else can follow

Start with a SQL block against the right connection. Use Python to reshape the results, add a table or chart, and record the assumptions in Markdown. Each block carries the investigation forward without separating the explanation from the work.

Kole can help propose the next step. You review configured consequential actions before they run, while keeping the notebook open for direct editing.

Quantum Lab / Ordered notebook

Follow the thought.
From query to chart.

SQL finds the data. Python shapes it. Notes explain it. Keep every step and its output together in one ordered notebook.

01SQL Query

Start with regional revenue

Aggregate orders on the selected connection. Name the output column so the next block can use it.

warehouse / analyticsSQL
SELECT region, SUM(amount) AS amount
FROM orders
GROUP BY region
ORDER BY amount DESC;
Output DataFramesql_result

02Python

Calculate each region’s share

Copy the SQL result and calculate a share of total revenue. Python starts its runtime when needed.

Input: sql_resultPython
# Work from the SQL block's DataFrame
regional = sql_result.copy()
total = regional["amount"].sum()
regional["share"] = regional["amount"] / total
Output DataFrameregional

03Markdown / Text

Give the numbers context

Keep the question and caveats beside the work.

Planning review / Analyst’s note

“Which regions need a closer look?”

North America contributes 48% of this sample’s revenue. Compare with the prior period and the regional plan before drawing a performance conclusion.

One DataFrame. Two ways to inspect it.

04Table

Inspect the regional summary

Input: regional
Illustrative regional revenue and share from the regional DataFrame
RegionAmountShare
North America$120,00048%
EMEA$80,00032%
APAC$50,00020%

3 sample rows · $250,000 total revenue

05Chart

See the revenue mix

Input: regional
0%25%50%North America48%EMEA32%APAC20%

Region → category · Share → value

Illustrative notebook and data; no live execution. SQL retains its connection context. Table and Chart blocks consume the selected SQL or Python DataFrame; rerun the source after changing it.

Give the next person a useful starting point

Keep the intermediate results that explain the conclusion. A teammate can revisit the notebook, inspect its SQL and Python, and extend the investigation as the question changes.

For a conversation and a report without working in a notebook, start with Kole chat.

Before you get started

A few useful answers.

Talk through your use case
How is Quantum Lab different from Kole chat?

Kole chat is a conversational way to ask questions and create reports. Quantum Lab is the notebook workspace for people who want to write or inspect SQL and Python, arrange analysis blocks, and build on intermediate results. Kole can help inside the notebook too.

Where do SQL and Python run?

SQL runs through the selected database connection. Python runs in the browser with Pyodide and bundled data packages. Table and chart blocks use DataFrames produced by the analysis.

Are queued runs, variables, and files available?

Queued notebook runs, notebook variables, and file handling are upcoming and not yet deployed. The current workspace supports SQL, Python, Markdown, tables, and charts.

Does every action need approval?

No. Approval applies to configured consequential actions proposed by the agent. Manual notebook work and read-only discovery do not all pass through that gate.

Make the next analysis easier to follow.

Keep the query, calculation, chart, and explanation together from the first block.

Open Quantum LabOr book a product walkthrough