New KurrentDB Release Delivers Non-disruptive Integration for AI Systems and Eliminates Custom Code in Event-driven Architectures

New connectors enable enterprises to ingest from Kafka, sync to PostgreSQL and SQL Server, and reorganize events without custom development

SAN FRANCISCO--(BUSINESS WIRE)--Event-native data platform innovator Kurrent today released KurrentDB 26, adding native Kafka Source Connector, Relational Sink and Custom Indices capabilities that reduce custom code requirements for event-driven architectures. The release enables operators to ingest events directly from Kafka, automatically synchronize read models in PostgreSQL and SQL Server and reorganize events by custom properties without touching application code. For enterprises building systems that require both audit trails and historical context for AI and analytics, KurrentDB 26 eliminates months of integration work by moving these capabilities from code libraries into configuration. The new version also extends archiving support to Azure and the Google Cloud Platform, complementing existing AWS archiving capabilities and providing enterprises with flexibility in long-term storage provider selection.

Enterprises increasingly need systems that capture complete event history for compliance, AI training and agentic AI operations while maintaining real-time streaming performance. KurrentDB’s event-native architecture stores immutable event logs while providing Kafka-compatible streaming. KurrentDB 26 extends this approach by natively integrating with Kafka for ingestion and relational databases for read model creation, eliminating the middleware layers that have become standard in contemporary data stacks. This builds on KurrentDB’s Multi-Stream Append capability, which allows atomic writes across multiple streams to maintain data integrity and consistency without complex saga, process manager or workflow patterns.

“The value of KurrentDB has always been time and flexibility – a data model capturing state and context, the power to shape data products instantly and the freedom to adapt without migrations. For agentic AI, these aren’t nice-to-haves, they’re requirements,” said Kirk Dunn, CEO of Kurrent. “Our new release extends these capabilities. The Kafka Source Connector ingests without custom code, Custom Indices reorganize without touching source events and the Relational Sink streams to Postgres and SQL Server in real-time. Whereas CRUD architectures are lossy, rigid and slow, KurrentDB keeps your source of truth complete, your architecture fluid and your data where you need it.”

Native connectors and configuration replace custom integration code

Building event-driven systems typically requires substantial custom development. Teams struggle to organize and persist data in Kafka, build microservices to maintain read models and implement orchestration logic for change data capture pipelines. KurrentDB 26 addresses this operational burden through three integrated capabilities:

• The Kafka Source Connector makes it possible for KurrentDB to subscribe directly to Kafka topics as a standard consumer, automatically routing events into KurrentDB streams based on topic names, message keys or header values. Events can be transformed and enriched during ingestion using KurrentDB’s projection engine, providing data preparation capabilities beyond what Kafka alone supports. The resulting system retains complete event history in an immutable append-only log while supporting the same consumer patterns as Kafka itself.
• The Relational Sink automates the creation and maintenance of read models in PostgreSQL and SQL Server databases. Each event automatically triggers SQL statements that update the target database, eliminating the custom code required to maintain real-time materialized views. Unlike traditional ETL pipelines that batch data periodically, the Relational Sink maintains continuous synchronization while preserving complete event history in KurrentDB. This approach addresses the CQRS pattern directly, moving implementation from custom development to declarative configuration.
• Custom Indices allow operators to create virtual views organized by any property without modifying the underlying event log. In practice, this enables use cases where events need reorganization by business date rather than arrival time (common in financial services for handling backdated corrections, eliminating the overhead of walking through entire ledgers twice), or where events must be filtered and reordered by customer ID or foreign key relationships (common in change data capture scenarios). Multiple indices can operate on the same event stream simultaneously, creating different logical views of the same immutable data without duplication overhead.

KurrentDB 26 delivers non-disruptive integration for AI systems and audit trails

Enterprise AI deployment introduces specific data requirements that traditional database and streaming architectures do not meet simultaneously. AI systems require comprehensive audit trails documenting every decision and action for compliance and debugging. They also require rich historical context across multiple systems to make informed predictions. Monolithic architectures designed around current operational needs struggle to provide both capabilities without significant refactoring.

KurrentDB 26 addresses this through non-disruptive integration with existing systems. Using change data capture capabilities, organizations can capture events from transaction logs and application databases without modifying application code. These captured events serve dual purposes: they form the audit trail for AI actions and provide the historical context AI systems require for decision-making. This allows enterprises to establish event-driven architectures gradually from existing monolithic systems, rather than requiring parallel rewrites.

“KurrentDB 26’s architectural flexibility extends beyond integration,” Dunn added. “It transforms one-way-door architectural decisions into two-way doors. Organizations can restructure their stream architecture, create new organizational views and evolve their systems without the massive refactoring traditional databases require.”

Availability

KurrentDB 26 is available immediately for deployment on-premises, on Kubernetes and on AWS, Microsoft Azure and Google Cloud Platform. Organizations can deploy open source KurrentDB or upgrade existing KurrentDB instances. Enterprise Connectors (Kafka Source, Relational Sink, MongoDB, RabbitMQ) are available through Kurrent Enterprise Edition.

About Kurrent

Kurrent provides an event-native data platform that stores and streams real-time, business-critical data with historical context in fine-grained streams, from origination to destination, for optimal AI, analytics and application outcomes. Kurrent is available on AWS, Microsoft Azure and Google Cloud Platform and as an on-premises solution. Headquartered in San Francisco, Kurrent’s event-native technology is deployed in high stakes use cases in finance, SaaS, oil and gas, manufacturing, retail, healthcare, automotive, government and many other industries globally. Illuminate your business™

Resources

• www.kurrent.io
• Kurrent LinkedIn