home.social

#langchain4j — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #langchain4j, aggregated by home.social.

  1. Our next #JCON2026 session is live: '#AI driven Unstructured Data Extraction Using #Apache Camel and #LangChain4J' with Alexandre Gallice

    A big chunk of the #data generated today is unstructured and it's growing at a #crazy fast pace. However ...

    Grab your coffee and hit play: youtu.be/tcCUTs0S_ZA

  2. Policy-aware moderation works best when the policy travels with the request.

    I built Shieldstral into a Quarkus LangChain4j service as scored input and output guardrails. The tutorial covers the one-token classifier contract, vLLM with Podman, fail-closed outages, deterministic HTTP stubs, and real-model threshold calibration.

    the-main-thread.com/p/shieldst

    #Quarkus #LangChain4j #AIGuardrails #MistralAI

  3. New hands-on tutorial on The Main Thread 🚀

    I built a real-time emotion detector in Java using #Quarkus, #JavaCV, #WebSocketsNext, and #LangChain4j.

    The pipeline streams webcam frames → detects faces locally → sends only the cropped face to a multimodal LLM (GPT-4o/Gemini) for emotion classification.

    High-throughput + low latency = fun and surprisingly practical.

    👉 the-main-thread.com/p/real-tim

    #JavaDevelopers #AIPipelines #MultimodalAI #FaceDetection

  4. Someone recently accused me of writing with AI.
    So I did what any Java dev would do — I built a detector.

    A hands-on Quarkus + LangChain4j project that mixes OpenNLP, embeddings, and curiosity.
    It doesn’t prove who wrote something, but it reveals why it feels human or machine-made.

    🔗 the-main-thread.com/p/ai-writi

    #Java #Quarkus #LangChain4j #AI #OpenNLP #DevLife

  5. LangChain4j 1.7.1 is out and now integrates GPU-accelerated large language model (LLM) via GPULlama3.java and TornadoVM

    #ai #gpu #java #langchain4j #llama #llm #tornadovm

    github.com/langchain4j/langcha

  6. Psst… your forms could talk. They could remember. Validate. Help. Loïc Magnette shows how to give forms a brain using #LangChain4j + #Quarkus. It’s not a dream. It’s a pattern.

    Learn it now: javapro.io/2025/04/15/ai-power

    #AIUX #ConversationalUI #JAVAPRO #Java

  7. RAG, memory, multimodality... in a form? Yep. Loïc Magnette built a talking form that can read files, validate data, and reply like a human.
    This isn’t a demo - it’s a dev-ready guide.

    Dive in: javapro.io/2025/04/15/ai-power

    #LangChain4j #Quarkus #AIUX #JAVAPRO #Java @langchain4j

  8. Forget next-gen #UX. This is next-human UX. Loïc Magnette turns boring forms into helpful convos - with validation, memory, even file upload.
    Your form, but smarter.

    Try the blueprint: javapro.io/2025/04/15/ai-power

    #LangChain4j #Quarkus #AIUX #ConversationalUI #Java

  9. Why should users adapt to your form… when the form can adapt to them? #AI flips the #UX script. Loïc Magnette shares how with memory, validation & RAG, forms become assistants.

    Read it here: javapro.io/2025/04/15/ai-power

    #LangChain4j #Quarkus #AIUX #ConversationalUI #Java @langchain4j