Build Smarter Knowledge Management Systems with AI Technology
summary
Claims a working Java stack: a Spring Boot app using Spring AI plus OCI Generative AI and Oracle Database 23ai vector types to build a RAG knowledge system. Implements neighbor‑graph and IVF indexes to scale (real pain point at ~500M chunks), supports embeddings from cloud or an AMA local LLM, and runs in Docker/Kubernetes with an Oracle Backend for Spring Boot for production. Demo ingests PDFs, returns grounded answers (embeddings ~100 ms on GPU) and an automated test run reports ~50% accuracy for one config.
AI-written summary · about
chapters (14)
- 0:00Introduction to RAG
- 2:27Vector search fundamentals
- 5:28Vector indexing strategies
- 8:32Implementing vector search
- 10:39Getting started with Oracle
- 12:06Spring AI integration
- 15:25Configuration and ingestion
- 18:28Building a RAG demo
- 21:01Leveraging private LLMs
- 23:51Demo code walkthrough
- 30:33Production deployment
- 35:18AI microservices sandbox
- 40:58Evaluating RAG accuracy
- 49:43Conclusion and resources
related talks
01h 03mBuilding AI powered Applications with Java
48mBuilding AI Powered Applications with Java
44mJava Meets AI: Empowering Spring Developers to Build Intelligent Apps
47mBlazingly Fast GenAI App Development Java and Spring AI
26mJava’s GenAI Era: Getting Spring AI and Neo4j Applications to Production
53mUnleash the Power of AI Agents with Java Building Intelligent Application
49mPractical AI in Java
46mGenerative AI and Java