Connect your warehouse, shape a semantic model, layer on instructions and examples, then let analysts and domain experts explore through conversation—without drifting from the metrics and guardrails you define.
SELECT r.name AS region, SUM(o.amount) AS revenue FROM orders o JOIN accounts a ON a.id = o.account_id JOIN regions r ON r.id = a.region_id WHERE o.order_date >= DATE '2025-10-01' AND o.order_date < DATE '2026-01-01' AND COALESCE(a.is_test, FALSE) = FALSE GROUP BY 1 ORDER BY 2 DESC;
Ask in plain language on the home screen. The assistant drafts SQL against your semantic model, surfaces assumptions, and supports threaded follow-ups so you refine filters, grain, and time windows without hand-writing queries.
Long generations stream into the UI so you see reasoning and SQL as they arrive—not only after a long wait.
Open suggested charts, pivot views, and summaries next to the result grid for stakeholders and leadership.
Save several connections, then switch the active profile from settings, modeling, or before you ask—always the right catalog and credentials.
Tables, columns, calculated fields, relationships, and lineage-style navigation. Deploy so every NL request compiles against definitions your team approves.
Capture business definitions, metric rules, and exemplar SQL. The AI service retrieves this context during generation so output matches how your org already writes SQL.
Pin recurring questions to dashboards, lean on cache and background jobs for heavier widgets, and revisit past runs from the home experience.
Tune data source parameters, track deploy status to the query engine, and—where enabled—use REST endpoints for ask, SQL generation, and model metadata so other tools can embed the same intelligence on your network.
Register to start a pilot, or sign in if your workspace is already provisioned. The console always opens after authentication.