arcodash
Start free
PlatformConnectorsPricingSecurityDocsBlog
Sign inStart free

What is arcodash?

arcodash is BI with data observability built in. It connects to the databases and warehouses you already run, lets your team write SQL against them or ask a question in plain English and review the SQL it drafts, turns results into charts and dashboards, and watches the data itself for signs of trouble.

What you can do with it

  • Connect to 19 connectors - PostgreSQL, MySQL, BigQuery, Snowflake, Amazon Athena, Redshift, ClickHouse, MongoDB, and more - plus a Query Results meta-source for joining across the results of your other saved queries.
  • Query with a SQL editor that supports parameters, reusable snippets, scheduled refresh, and full version history with draft/publish so you can iterate without breaking what’s live.
  • Ask with AI: describe what you want in plain language and an agent drafts SQL, runs a read-only preview, checks the result, and retries on errors before handing you the query to review. AI can also pick a visualization for a result, propose data-quality checks, and explain why an alert fired. See AI-assisted SQL generation.
  • Visualize results as tables, line/bar/area/pie/scatter charts, counters, and funnels.
  • Dashboard those visualizations onto a grid layout, with dashboard-level parameters that filter every widget at once.
  • Alert on thresholds - a query result crossing a value notifies email, Slack, webhook, PagerDuty, Mattermost, Teams, Discord, Google Chat, or Datadog.
  • Monitor data observability directly: freshness, row count, null rate, and duplicate checks that catch a broken pipeline before a stakeholder does, either against a threshold you set or against a baseline arcodash learns from the monitor’s own history.
  • Share dashboards and queries with per-object access control, public secret-token links (org-only, or anonymous when an admin allows it), or iframe embeds.
  • Administer with users, groups (each scoped to specific connectors), and an audit log.

[!NOTE] The AI features are optional and off until an admin configures a provider key (Gemini, OpenAI, or Anthropic - see AI provider configuration). Most of what they send is schema-level metadata. Two features can also send data values: sampled column values (switchable off in AI Settings) and, for alert explanations, investigation query results (the model is asked to aggregate them). Data handling & privacy states exactly what each feature sends.

Where to go next