Data Aqmen
Data Aqmen
Sign in

Getting started

Ask an AI to analyze your files and it works — once. The cleaned data, the queries, the reasoning: all of it evaporates with the chat, and the numbers come without receipts. The files it started from are no better — column meanings living in someone's head, copies drifting apart. Data Aqmen gives that work somewhere to live: a workspace where the data, what it means, and what you concluded from it stay together — and stay connected.

You (or an agent acting for you) load clean, cited data, transform it with SQL, chart it, and record what it means. Every number traces back to its source document, every column carries its meaning and units, and every conclusion is pinned to the query that demonstrates it — so when the data changes, you know exactly what to re-check. Work compounds instead of starting from zero.

1. Sign in and create a workspace

Sign in and create your first workspace. Check "Start with demo data" to get a complete worked example — a real market-research project on European HR-tech, built from public sources with every row cited. It is the fastest way to see how the pieces fit, and you can delete it anytime.

2. The pieces

  • Datasets — your raw tables, imported from CSV/JSON/Parquet files. Every dataset carries sources: citations for where the data came from.
  • Transformations — SQL that materializes new tables from your datasets. Each transformation and its output are one artifact: open one to see the result, the schema, and the SQL that produces it.
  • Queries — saved SELECTs with a chart. They re-run on open, so they are always current.
  • Views — custom dashboards, usually built by an agent you describe them to. See Views.
  • Insights — short conclusions ("Core HR captures 62% of tracked funding") pinned to the query or dataset that demonstrates them. When the underlying data changes, the insight is flagged so you can re-check it.
  • Collections — folders that keep a growing workspace organized.

3. Import your own data

Datasets → Import data: drop a CSV, preview how it parses, and import. Add sources so the data stays defensible — where each file came from, and any judgement applied while preparing it. The full flow is in Import & sources.

4. Explore

Write SQL in the SQL editor (the schema browser shows every table, ⌘⏎ runs), save useful results as queries, and watch the lineage graph connect sources → datasets → transformations → charts.

5. Connect an agent

The recommended way to work: connect an agent over MCP and let it do the heavy lifting while every result lands in your workspace, cited and auditable. See MCP integration for the two-minute setup.

Faster with an agent — once connected, your first prompt can be as simple as "describe my workspace and tell me what's worth looking at." The agent reads the schemas, the sources, and the recorded insights, and comes back oriented — every guide page ends with a prompt like this one.