Latest insights, best practices and expertise from software development, AI and digital transformation.
27 articles
Page 1 of 3
AI agents need a declarative contract – just like APIs need OpenAPI. Discover how ADL (Agent Definition Language), the A2A protocol, the Inference Gateway, and Spring Boot enable production-ready agent infrastructure – from code generation through OCI images to Kubernetes and GitOps.
AI agents are more than a clever chatbot: they plan, call tools and work through multi-step tasks. Turning that into reliable software takes orchestration — and it doesn't have to happen in Python. With Spring AI, LangChain4j and Quarkus, the Java ecosystem has reached a point in 2025/2026 where agent systems can be cleanly integrated into existing enterprise landscapes. This article shows what matters when it comes to architecture.
Adding AI capabilities to a Java backend used to mean building infrastructure from scratch. Spring AI changes the game — with familiar Spring principles and provider-agnostic abstractions.
We show how we secure our Spring Boot services as an OAuth2 Resource Server with Keycloak -- from the security configuration and JWT validation to method-level security and testing.
In distributed systems, network failures and retries are inevitable. Idempotent APIs ensure that repeated requests do not produce unwanted side effects -- a pattern we use in every microservice.
Why we chose Nix Flakes and direnv over Docker Compose for local development environments -- and how this leads to faster, lighter, and more reproducible dev setups.
After four years of experience with GraalVM Native Images, we know the typical pitfalls -- from reflection issues to missing resources to debugging without JVM tools. A field report.
With a single property, Spring Boot 3.2+ makes Virtual Threads available -- we show the migration, compare benchmarks with platform threads, and explain when switching is worthwhile.
Java 21 is arguably the most significant LTS release in years -- Virtual Threads are finally GA and fundamentally change the concurrency model. We look at the most important features.
With Java 19, Virtual Threads have landed as a preview feature -- Project Loom is becoming tangible after years of development. We examine how Virtual Threads work, when they help, and what to consider during adoption.
Dockerfiles are error-prone and hard to reproduce. We show why we rely on Google Jib for our Spring Boot services and how it integrates with Gradle.
Spring Boot 3.0 has arrived, bringing fundamental changes with Jakarta EE, Java 17 as the baseline, and GraalVM support. We show what's changing and how to successfully migrate from 2.x.