Features
True streaming data processing engine in Rust
Same code for streaming and batch workloads
Incremental joins, filters, group-by, reduce
Temporal joins, windows, ranges
Custom stateful reducers and pointer-based data structures
User Defined Functions and async API/LLM calls
Built-in vector indexing (HNSW), BM24, hybrid search
LLM extension pack for RAG pipelines
Time-series libraries, clustering, classification
Connectors for Kafka, PostgreSQL, Redpanda, S3, Google PubSub, Slack, Logstash
Data persistence with S3 and schema validation at compile time
REST API endpoint for serving features with sub-millisecond latency
Python and SQL programming APIs
Monitoring and traces with OpenTelemetry and Grafana
Visual Explorer with live dashboards and geospatial viz
High-fidelity synthetic data generation via TabularARGN
Agentic data science automation for training/sampling
Natural language AI Assistant executing Python code
Mock data generation for safe experimentation and testing
Simulated data for edge-case and what-if modeling
Multi-table synthesis with referential integrity
Time-series support and data rebalancing
Differential privacy with temperature control
Deployment on Kubernetes or Red Hat OpenShift
Open-source Synthetic Data SDK under Apache v2 license
Data connectors for Databricks, AWS, Snowflake, BigQuery
Real-time data access from production systems (e.g., Databricks)
Star schema and nested sequences support
REST API and Python Client for programmatic access
Conditional simulation and seeded generation
