RAG works… until it doesn’t. Most failures aren’t crashes.
They’re silent: weak grounding, partial answers, confident hallucinations.

I wrote a hands-on guide showing how to measure RAG quality in Java using Quarkus + LangChain4j:
– faithfulness
– relevance
– retrieval quality
– reference-based checks

Local models. No hype. Production thinking.

https://www.the-main-thread.com/p/measuring-rag-quarkus-langchain4j-evaluation

#Java #Quarkus #RAG #LangChain4j #LLM #AIEngineering #Fediverse #OpenSource

LangChain4j has crossed an important line:
from “interesting library” → production Java AI infrastructure.

I wrote a long-form guide with 50 LangChain4j interview questions, covering:
• AI Services
• RAG & embeddings
• Tools & agents
• Memory & context limits
• Observability, cost, and security

Written from a production Java perspective — not Python demos.

https://www.the-main-thread.com/p/langchain4j-interview-questions-java

#Java #LangChain4j #AIinJava #EnterpriseJava #RAG #LLM #SoftwareArchitecture #DevExperience

Prompt injection gets a lot harder once users stop writing in English.

Regex-based guardrails fail quietly the moment prompts cross language boundaries. In this article, I walk through how to build semantic, multilingual prompt injection guardrails in Java using Quarkus, LangChain4j, and ONNX embeddings—fast enough for real systems.

https://www.the-main-thread.com/p/multilingual-prompt-injection-guardrails-quarkus-langchain4j

#Java #Quarkus #AI #LangChain4j #AISecurity #PromptInjection #EnterpriseAI

Open source is reshaping how AI gets built. ⚙️

At #ArcofAI, Cedric Clyburn & Legare Kerrison deliver a crash course on building, testing & deploying open AI models — with real demos using LangChain4j & open tooling.

🔗 https://www.arcofai.com/speaker/254a9c51f2b34a8c8fb6b3fb428fc948

🎟️ Secure tickets - https://arcofai.com

#OpenSource #AI #GenerativeAI #LangChain4j #AIEngineering #Java #SoftwareEngineering #DevCommunity #TechConference

Build smarter apps with AI! 🤖

In this full-day LangChain4j workshop, @venkats shows you how to safely integrate AI, tap into global & org data, & create practical solutions through hands-on exercises.

https://www.arcofai.com/speaker/7e0794827e8242ef8464e383e1209fad

🎟️ Tickets https://arcofai.com

#AI #LangChain4j #SoftwareEngineering #DeveloperWorkshop #MachineLearning #PracticalAI #FullDayWorkshop

Au #SnowCamp j'ai eu le plaisir de donner une conférence avec
@emmanuelbernard sur les 𝗽𝗿𝗼𝘁𝗼𝗰𝗼𝗹𝗲𝘀 𝗲𝘁 𝘀𝘁𝗮𝗻𝗱𝗮𝗿𝗱𝘀 𝗽𝗼𝘂𝗿 𝗹𝗲𝘀 𝗮𝗴𝗲𝗻𝘁𝘀 𝗜𝗔, avec un focus sur #MCP et #A2A

Tous les liens vers les slides, les repos GitHub avec nos démos, sont disponibles ici :

https://glaforge.dev/talks/2026/01/16/on-ai-standards-and-protocols-focus-on-mcp-and-a2a/

En plus, avec du ☕️ #java dedans, avec #micronaut, #quarkus, #langchain4j et #ADKjava.

Aucun 🐍 Python en vue !

As a #Java developer, there’s no need to learn a new language to start building AI-infused apps!

With the #opensource project #LangChain4j, you can manage interactions with #LLMs - handling chat memory to keep requests efficient, focused & cost-effective.

This #InfoQ article covers the essential concepts you need to begin building intelligent, responsive Enterprise #JavaApps with #AI.

Read more ⇨ https://bit.ly/4oNmLqH

#AI #Quarkus #SoftwareDevelopment

Most AI configuration problems I see are not model problems.
They’re configuration problems.

Once prompts get longer, examples creep in, and feature flags multiply, application.properties stops being a good fit. That’s when I reach for a typed TOML file and wire it cleanly into #Quarkus and #LangChain4j.

This post shows how I do that with local models via Ollama and Dev Services, keeping AI configuration readable, reviewable, and boring in the best way.

https://www.the-main-thread.com/p/toml-ai-configuration-quarkus-langchain4j-ollama

#Java