{"slug": "quarkus-newsletter-74-november", "title": "Quarkus Newsletter #74 - November", "summary": "Quarkus Newsletter #74 for November highlights an agentic GitHub pull request reviewer built with Quarkus LangChain4j that combines a decision router, debates, voting, reflection loops and a human in the loop, with a per-step model choice and a cost cap, as described by Mario Fusco in \"Debates, voting, and reflection loops: composing an agentic PR reviewer that costs less than a dollar.\" The issue also covers Fouad Almalki's Quarkus Almanac, which shows configuration, dependency, extension, class and API changes between two Quarkus releases, plus Gaelle Fournier's guide to extending a Quarkus REST app with Apache Camel for Kafka events, HTTP calls with retries and file ingestion. Markus Eisele's \"Quarkus Already Does That: 7 More Features Worth Knowing\" and Emil Lefkof's account of Trax streaming Oracle change data through Debezium and Kafka Streams to React and Swift iOS clients over WebSockets round out the issue.", "body_md": "See how an agentic GitHub pull request reviewer, built with Quarkus LangChain4j, combines a decision router, debates, voting, reflection loops and a human in the loop, with the right model for each step and a cost cap by reading \"Debates, voting, and reflection loops: composing an agentic PR reviewer that costs less than a dollar\" by Mario Fusco. Fouad Almalki introduces \"Quarkus Almanac\". The Quarkus Almanac shows what changed between two Quarkus releases: configuration, dependencies, extensions, classes, and their API. Read \"Extending Your Quarkus Application with Apache Camel\" by Gaelle Fournier to get a practical guide showing how to extend an existing Quarkus REST app with Camel for Kafka events, HTTP service calls with retries, and file ingestion. Markus Eisele demonstrates a few smaller Quarkus features to check before you write more code around your application in \"Quarkus Already Does That: 7 More Features Worth Knowing\". Learn how Trax uses Quarkus to stream Oracle change data through Debezium and Kafka Streams and push it live to React and Swift iOS clients over WebSockets in \"How Trax keeps aircraft maintenance real-time with Quarkus\" by Emil Lefkof.\n\nYou will also see the latest Quarkus Insights episodes, top social posts and upcoming Quarkus attended events.\n\nCheck out [Newsletter #74: November](https://quarkus.io/newsletter/74/)!\n\nWant to get newsletters in your inbox? [Sign up for the newsletter](https://quarkus.io/newsletter) using the on page form.", "url": "https://wpnews.pro/news/quarkus-newsletter-74-november", "canonical_source": "https://quarkus.io/blog/quarkus-newsletter-74/", "published_at": "2026-11-19 00:00:00+00:00", "updated_at": "2026-11-19 23:14:39.968535+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "large-language-models"], "entities": ["Quarkus", "Quarkus LangChain4j", "Mario Fusco", "Fouad Almalki", "Quarkus Almanac", "Gaelle Fournier", "Apache Camel", "Markus Eisele"], "also_reported_by": [], "takeaway": "Composing a PR-review agent from a decision router, debate, voting and reflection loops with a human in the loop and a per-step model choice keeps the whole review under one dollar.", "recommendation": {"advice": "Route each step of an agentic PR reviewer to the cheapest adequate model and cap total spend; a router-plus-debate-and-voting design kept the review under $1.", "playbook": "agent-architecture", "playbook_url": "https://wpnews.pro/playbooks/agent-architecture"}, "alternates": {"html": "https://wpnews.pro/news/quarkus-newsletter-74-november", "markdown": "https://wpnews.pro/news/quarkus-newsletter-74-november.md", "text": "https://wpnews.pro/news/quarkus-newsletter-74-november.txt", "jsonld": "https://wpnews.pro/news/quarkus-newsletter-74-november.jsonld"}}