Best of QuarkusJanuary 2026

  1. 1
    Article
    Avatar of rhdevRed Hat Developer·15w

    The case for building enterprise agentic apps with Java instead of Python

    Java offers significant advantages over Python for enterprise AI applications in regulated industries. The JVM provides architectural stability through compile-time verification, mature security tooling with SAST and CVE management, and unified observability. Java frameworks like Spring AI and Quarkus with LangChain4j enable production-grade agentic applications while maintaining existing governance standards, reducing cognitive load on teams, and leveraging established Java expertise. Code-specialized LLMs also generate higher quality Java code due to strong typing and abundant well-structured training data.

  2. 2
    Article
    Avatar of shaafShaaf·17w

    Keycloak MCP Server: Manage Identity with Natural Language

    An MCP server enables natural language management of Keycloak identity and access management through AI assistants. The open-source project supports 45+ operations including user management, realm configuration, client setup, and authentication flows. Version 0.3 introduces JWT authentication, SSE transport, Kubernetes deployment support, containerization, and production-ready features like health checks and Prometheus metrics. Developers can manage Keycloak by conversing with AI assistants instead of using the admin console or CLI commands.

  3. 3
    Article
    Avatar of quarkusQuarkus·16w

    New Dev Services Api

    Quarkus 3.25 introduces a new Dev Services API that resolves port conflicts and excessive resource usage in test suites by deferring service startup until after augmentation. The new model eliminates the problem where all Dev Services started during JUnit discovery, causing multiple containers to run simultaneously. Extension authors benefit from a simpler programming model using builders, with Quarkus core handling discovery and container reuse. Migration requires implementing the Startable interface, using builder patterns instead of direct construction, and removing static variables from extension processors. Several extensions including Redis, Lambda, Narayana, and Kafka have already been converted, with ongoing work tracked for others.