Build a realtime, AI-native data warehouse
resink.ai is an AI-first, cloud-native realtime data warehouse. With a realtime-first architecture, you ship streaming analytics and AI features faster—without sacrificing governance or reliability.Core stack: Flink for compute, Kafka for streams, StarRocks for OLAP, and open
table formats (Iceberg/Paimon). ResinKit SDK + UI provides a Jupyter-friendly
experience for Variables, Data Sources, and Flink SQL.
Get started
Quickstart
Install ResinKit, connect, and run the UI tools in minutes.
ResinKit SDK & UI
SDK Overview
Install, configure, and explore the SDK with Jupyter-friendly UI.
Variables UI
Centrally manage configuration and secrets.
Data Sources UI
Add SQL/Kafka sources, test connections, and crawl schemas.
Flink SQL UI
Execute SQL with live streaming results in notebooks.
RTDW reference workflows
Start from proven blueprints to accelerate implementation.S001: Entity Extraction & Archiving
Extract entities and persist history for downstream enrichment.
S002: Raw Event Ingestion & Storage
Ingest raw events reliably with partitioning and retention.
S005: Event Aggregation (Session/Daily)
Build session windows and daily aggregates for fast analytics.
S007: Fact Order Star Schema
Model a star schema for orders with dimensions and facts.
S008: SCD Type 2
Preserve entity history for accurate point-in-time queries.
S011: CDC Ingest to Iceberg
Stream database changes into open table formats.