Loading…
Plain-English questions on time-series data — built on Postgres.
TimescaleDB is Postgres with a powerful time-series extension, and iDBQuery treats it as both. Paste the connection string from Timescale Cloud or your self-hosted instance, and the schema — including hypertables, regular tables, continuous aggregates, and any PostGIS layers — is introspected in seconds. The AI is told which tables are hypertables and which continuous aggregates already exist, so it tends to query the right rollup at the right granularity instead of scanning raw chunks. Every chat question becomes a safe parametrized SQL statement using `time_bucket`, window functions, and gap-filling where appropriate, and the result is rendered as a time-series chart, table, or dashboard. Compression, data retention, and chunk pruning all work transparently — they're a property of how Timescale stores the data, not something the SQL has to know about. Connection strings are encrypted at rest with strong symmetric encryption, used only at query time, and never returned by any API. The free tier — 1M tokens per month, three sources, five reports, no credit card — is enough to wire up a real Timescale instance and validate the workflow. iDBQuery turns time-series telemetry into a conversation.
Everything iDBQuery's TimescaleDB connector supports out of the box.
Project → Add source → PostgreSQL. TimescaleDB is a Postgres extension, so the Postgres connector handles it.
Timescale Cloud → service → Connection info, or use your self-hosted host/port/db/user/password.
iDBQuery reads `_timescaledb_catalog` to flag hypertables and continuous aggregates so the AI plans queries against the cheapest source.
Ask 'Show CPU usage bucketed by 5 minutes for host-7 last 24 hours' — the AI writes the `time_bucket` SQL and plots it.
Free tier covers most teams. No credit card.
No credit card required · 1M tokens / month free