Using Generative AI with Java: Beyond Simple Chatbots
summary
Shows concrete Java examples using LangChain4J to call Google Gemini and to run open Gemma models locally via Olama/LocalAI. Demonstrates retrieval-augmented generation by encoding documents into embeddings, storing them in an Apache Cassandra/Astra vector store with a Cassandra vector index, and wiring retrieval into chat flows with prompt templates; also presents a custom Java ANN (jvector) claimed to give higher QPS than Pinecone. Covers running local Gemma, integrating LangChain4J + Gemini, and mentions FLARE and LMM techniques to reduce hallucinations with end-to-end demos.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 2:33Generative AI and LLMs explained
- 1:03:14Demos with Vertex AI and Gemma
- 1:33:1415-minute break
- 1:48:11Vector databases and embeddings
- 2:00:53Cassandra and vectors with Java
- 2:21:01RAG: Retrieval Augmented Generation
- 2:33:04Tools and function calling
- 2:43:33Q&A
related talks
44mHands-on-lab: It's all about going hands-on with Generative AI!
55mAI in Java? Meet LangChain4j and Build Intelligent Services
02h 49mJava meets AI: Build LLM-Powered Apps with LangChain4j
46mGenerative AI and Java
53mJava meets AI: Build LLM-Powered Apps with LangChain4j
42mBuilding AI Powered Applications with Java
48mBuilding AI Powered Applications with Java
58mBuilding Your Own Private GenAI Chatbot With Java Tools – A Step-by-Step Adventure