The xentropy data platform gives you an isolated data warehouse for every tenant — powered by Delta Lake on Cloudflare R2 and queried via DuckDB, an embedded OLAP engine.
#
- Per-tenant Delta Lake — your data lives in its own isolated catalog on R2. No other tenant can access it.
- Direct SQL access — run read-only SQL queries through the xentropy API. SELECT, aggregate, join, and analyse your data using standard SQL.
- Automatic Supabase sync — connect your Supabase project and the platform automatically ingests your public tables into your Delta Lake.
- No infrastructure to manage — the query engine and storage are fully managed. You get ACID transactions, time-travel, and schema evolution for free with Delta Lake.
#
Your Supabase Project
│
▼ PostgREST REST API (range-paginated)
Sync Worker (managed)
│
▼ Delta Lake writes
Cloudflare R2
s3://xen-platform/tenants/{your-tenant-id}/
events/
users/
subscriptions/
_config.json
│
▼ DuckDB queries (read-only)
Query Engine (managed)
│
▼ JSON over HTTP
xentropy API (/data-platform/*)
#
- Connect your Supabase project via the dashboard or API
- Wait for the initial sync — the ingestion worker auto-detects your public tables and writes them as Delta tables
- Query your data using the SQL API endpoint
- Set your sync schedule — by default connectors sync every 5 minutes
The first sync fetches all rows from every discoverable public table. Subsequent syncs are full overwrites (CDC is coming soon).
#
- Connecting Supabase — setting up your first Supabase connector
- Querying Data — SQL reference, examples, and best practices
- Sync Schedules — understanding and configuring sync frequency
- Data Retention — how data is stored and retained
#
All data-platform requests must include:
- An Authorization header with a valid Supabase JWT (see Authentication)
- An X-Tenant-Id header matching your tenant ID
The API enforces tenant isolation at the guard level — you can only access data belonging to your tenant.