From JVM evolution to Spring AI patterns and developer tooling

The Java Story: A Film About All of Us

🎯 Featured Article

🎬 The Java Story: A Film About All of Us The upcoming Java documentary celebrates nearly three decades of innovation and the global developer community that shaped one of the industry’s most influential platforms.

  • Explores Java’s technical and community evolution
  • Premieres live on YouTube on July 17

⏩ TL;DR (Quick Recap)

  • Java tooling continues advancing across CI, performance and AI
  • Kotlin gains momentum through productivity and AI benchmarks
  • Spring focuses on AI integration and startup optimization
  • Developer workflows increasingly prioritize reliability and automation

☕ JVM Corner

📊 Comparative Analysis of Development Cycle Speed in Java and Kotlin Based on IDE Telemetry Data JetBrains telemetry suggests Kotlin projects generally sustain shorter development cycles than comparable Java projects.

  • Uses real IDE telemetry instead of surveys
  • Highlights slower productivity degradation over time

🗄️ Demystifying Exposed: The Intelligent SQL Library for Kotlin An introduction to JetBrains Exposed and its DSL-based approach to database access for Kotlin applications.

  • Explains Kotlin-first ORM concepts
  • Demonstrates Spring Boot integration

🤖 Introducing the Kotlin Benchmark for AI Coding Agents JetBrains introduces a public benchmark for measuring AI coding agents on realistic Kotlin engineering tasks.

  • Evaluates end-to-end development tasks
  • Enables comparable agent benchmarking

Why Low-Latency Java Still Requires Discipline Modern JVM improvements help performance, but predictable latency still depends on disciplined engineering practices.

  • Reviews Loom, ZGC and Valhalla impacts
  • Focuses on deterministic trading workloads

🌱 The (Petty) Reason We Didn’t End Up Using jj Gradle explains why a seemingly minor repository detail prevented adoption of Jujutsu in JVM projects.
🔐 What’s New in actions/setup-java 5.4 and 5.5 Recent releases introduce JDK signature verification, Kona JDK support, and several Maven workflow improvements.
📽️ Forking the JVM to Save JINI A look at reviving JINI through JGDMS and the engineering decisions behind maintaining distributed Java infrastructure.

🍃 Spring Updates

🔄 A New Home for Spring Cloud Contract: Transitioning to Stubborn.sh Spring Cloud Contract transitions to independent stewardship while maintaining continued development.

  • Transfers project ownership
  • Preserves ecosystem continuity

🧰 Spring AI’s Dynamic Tool Discovery Dynamic tool discovery reduces unnecessary token usage by exposing only relevant tools to language models.

  • Optimizes LLM context size
  • Improves AI request efficiency

🚀 Bean Background Initialization in Spring Framework Background bean initialization improves application startup by asynchronously creating selected beans.

  • Reduces startup latency
  • Supports expensive initialization workloads

🧠 Spring AI Recipes: Using More Than One LLM Configure multiple ChatClient instances to route requests across different AI providers.
📨 Kafka Streams Dead Letter Queue in Spring Kafka 4.1 Spring Kafka 4.1 introduces native support for Kafka Streams dead-letter queues and processing exceptions.

🔍 Extra Reads

Web & Performance
🌐 CORS: What is it Protecting? — Clarifies why CORS is fundamentally a browser security mechanism.
💾 Data Access Patterns That Makes Your CPU Really Angry — Demonstrates how memory access patterns dominate CPU performance.

AI Engineering
🦙 Qwen 3.6 27B Is the Sweet Spot for Local Development — Evaluates Qwen 3.6 as a practical local coding model.
📝 Markdown Is Becoming the Memory Layer for AI Agents — Explores why Markdown is emerging as AI agents’ shared knowledge format.
🛠️ Better Models: Worse Tools — Examines why newer language models can regress in structured tool usage.

Engineering in Practice
🏠 Two Nasty Surprises in Home Assistant’s Config — Shares practical lessons from Home Assistant automation pitfalls.
🏢 How Successful Companies Go Blind — Discusses how organizations gradually lose the ability to recognize competence.
💼 LinkedIn, a Mass Grave of Ghost Jobs, Is Now Becoming a Dating App — Looks at LinkedIn’s expanding role beyond professional networking.

Originally posted on marconak-matej.medium.com.