{"id":4466,"date":"2024-02-06T17:42:18","date_gmt":"2024-02-06T17:42:18","guid":{"rendered":"https:\/\/kindgeek.com\/blog\/?p=4466"},"modified":"2025-02-21T08:11:33","modified_gmt":"2025-02-21T08:11:33","slug":"experiments-with-langchain4j-or-java-way-to-llm-powered-applications","status":"publish","type":"post","link":"https:\/\/www.kindgeek.com\/blog\/post\/experiments-with-langchain4j-or-java-way-to-llm-powered-applications","title":{"rendered":"Experiments with Langchain4j or Java way to LLM-powered applications"},"content":{"rendered":"

Recently updated on February 21, 2025<\/p><\/div>\n

In this article, we are discussing with Michael Kramarenko, Kindgeek CTO, how to incorporate LM\/LLM-based features into Java projects using Langchain4j. This framework streamlines the development of LLM-powered Java applications, drawing inspiration from Langchain, a popular Java Langchain<\/em> framework that is designed to simplify the process of building applications utilizing large language models.<\/em><\/p>\n\n\n\n

Join us with a cup of a great coffee, and enjoy our info-packed ten-minute read on Java way to building an LLM app with Langchain4j\u2615.<\/em><\/p>\n\n\n\n

Brief intro or a quick reality check<\/h2>\n\n\n\n

Integrating LM\/LLM-based functionality in existing and new products is definitely proving to be more than a mere trend. Today, embracing AI innovation is more about keeping up with the curve.<\/p>\n\n\n\n

Indeed, as per Gartner<\/a> analysis, Generative AI, GPT, and LM\/LLM applications are becoming key driving forces in shaping IT products for the upcoming 5\u201310 years. And no wonder, artificial intelligence is the future. Such solutions will use features like chat but will be integrated with a corporate knowledge base, internal and external data via API, and acquired from web crawling. This article provides insights on how to use LM\/LLM in Java applications effectively.<\/p>\n\n\n\n

Content<\/h2>\n\n\n\n
    \n
  1. Setting the Scene or \u201cWhat problem are we solving?\u201d<\/a><\/li>\n\n\n\n
  2. Starting from business needs, not from technological aspects<\/a><\/li>\n\n\n\n
  3. Typical building blocks of LLM-powered application<\/a><\/li>\n\n\n\n
  4. Starting our Langchain4j for llm journey <\/a><\/li>\n\n\n\n
  5. User Experience aspect: delving into nuances<\/a><\/li>\n\n\n\n
  6. Memory and state management<\/a><\/li>\n\n\n\n
  7. Corporate data aspect or extending LLM’s knowledge<\/a><\/li>\n\n\n\n
  8. LLM Functions and Langchain4j java @Tools<\/a><\/li>\n\n\n\n
  9. Summing up<\/a><\/li>\n<\/ol>\n\n\n\n<\/a>\n\n\n\n

    Setting the Scene or \u201cWhat problem are we solving?\u201d<\/h2>\n\n\n\n

    At this point of the story let us meet our virtual character<\/strong> \u2013 John Geek<\/strong>.<\/p>\n\n\n\n

    He’s a tech lead of a mission-critical Java Spring-based project that has been in development for years. <\/p>\n\n\n\n

    Now, amidst AI advancements, the customer asks John to explore LLM integration into Java fully integrated with the essential corporate data (structured, semi-structured, and unstructured), internal and external data points accessible via REST API. <\/p>\n\n\n\n

    \"\"<\/figure>\n\n\n\n

    So what are the options for John?<\/p>\n\n\n\n

    He perfectly knows that Python is the lingua franca of AI development, not Java.<\/p>\n\n\n\n

    Theoretically, he could choose the \u201cPolyglot architecture\u201d way. However, such a way would increase the Learning Curve for John’s development team, maintenance complexity, complicate dependency management, CI\/CD pipeline, testing, and deployment strategies. <\/p>\n\n\n\n

    It’s clear to him that the microservices architecture is technology-neutral. However, his current architecture is a modular monolith. Given this, transitioning to a microservices architecture presents significant implementation challenges. It\u2019s an additional complication compared to the \u201cPolyglot architecture\u201d way.<\/p>\n\n\n\n

    On the other hand, John has heard about different pure Java options like<\/p>\n\n\n\n