Data scientists — Sauna

The AI agent for data scientists

An AI coworker that queries the warehouse, runs the analysis, and writes up the finding — it builds the cohort, charts the result, and drafts the readout while you keep thinking about the model.

Sauna is an AI assistant for data scientists that closes the loop from question to write-up. It queries your warehouse and explores the cohort, reads the experiment for lift and significance, writes the analysis up as a stakeholder memo, and monitors the metrics for drift — so more of your day is the modeling, not the plumbing.

What it does

What Sauna does for data scientists

  1. 01

    Query and explore

    Write the SQL against your warehouse, pull the cohort, profile the columns, and hand back the distribution and the first charts — so exploration starts from data, not a blank notebook.

  2. 02

    Read the experiment

    Pull an A/B test's results, compute the lift and the significance, and draft the plain-English readout — what moved, by how much, and whether you can trust it.

  3. 03

    Write the analysis up

    Turn a finished analysis into a stakeholder-readable memo — the question, the method, the result, the caveat — with the charts embedded, ready to share.

  4. 04

    Monitor the metric

    A recurring check on a key metric or model — flag the drift, the anomaly, the day it broke trend — delivered before anyone asks why the number looks off.

Put Sauna to work on this.

Get started for free

In context

Ask in chat. It does the work.

Sauna shows up where you already work — the web app, Slack, email, iMessage, and Superhuman. It reads what it needs, does the task, and comes back with the draft for your approval.

app.sauna.ai/sauna
You
You
@Sauna Query our warehouse for last month's signups by acquisition channel, compute 30-day retention per channel, chart it, and draft a memo on which channel retains best and why.
S
Sauna
Reading Snowflake and PostgreSQL.
Done — here's the draft and what I found. Approve to send.

Try it

Ask in plain words

The literal prompt for this job. Open it in Sauna and it picks up from there.

Query our warehouse for last month's signups by acquisition channel, compute 30-day retention per channel, chart it, and draft a memo on which channel retains best and why.
Example promptRun in Sauna →

Plugs into the tools you already run — and thousands more, or any MCP server.

What you get

A finished result, not a transcript.

Reports, drafts, and dashboards Sauna assembles from your live tools — kept current on its own.

team.sauna.app/analysis
4
experiments live
2
readouts due
1
metric drifting
Retention by signup cohortLast 8 weeks
Onboarding A/B+3.4% · significant
Pricing test readoutDue · Fri
DAU modelDrift flagged · investigate

Get the result, not the busywork.

Get started for free

Good to know

Questions, answered.

Can it run code and query data, not just talk?

Yes. Sauna writes and runs SQL against your warehouse, executes the analysis, builds charts, and exports the memo — it does the work, not just describe it.

Where does it get the data?

Whatever you connect — Snowflake, Postgres, your product analytics, sheets, and thousands of other sources, or your own over an API key or MCP server.

Is my data safe?

Sauna reads only what you connect, and acts only after you approve. Your workspace and its memory are yours, not training data.

Keep exploring

More ways teams use Sauna

Put Sauna to work for data scientists.


Get started for free