Building a custom AI chatbot with Spring Boot, React, and LangChain4j
summary
Hands-on build of an airline customer-service chatbot that integrates Spring Boot, React, LangChain4j, Hilla with OpenAI (GPT‑4o) streaming, an embedding model, and a vector store; app ingests a terms‑of‑service text split into 50‑token segments and stores embeddings in an in‑memory embedding store with a 1,000‑token chat memory window. Implemented callable tools getBookingDetails, changeBooking and cancelBooking that update an in‑memory bookings list; demo runs live and answers grounded queries, with minor latency and occasional “holding” behavior but overall successful.
AI-written summary · about
chapters (12)
- 0:00Introduction to AI app dev
- 2:24Bridging generic AI to context
- 3:45Architectural components of AI
- 5:03AI autonomy levels
- 7:42Understanding context and tokens
- 9:57Methods for teaching LLMs
- 11:40Implementing RAG with embeddings
- 14:47Adding tools to co-pilots
- 16:01Full-stack UI with Hilla
- 17:24Coding the AI assistant
- 35:07Implementing tools and demo
- 42:14Q&A and concluding remarks
related talks
34mBuilding a Custom AI Chatbot with Spring Boot, React, and LangChain4j
44mBuilding AI Chatbot with Spring Boot, React and ChatGPT (Live Code)
01h 12mBuild Your Own AI Chatbot with OpenAI & LangChain
55mAI in Java? Meet LangChain4j and Build Intelligent Services
48mBuilding AI Powered Applications with Java
01h 02mBuilding full stack AI chatbots with Java - livecoding
46mCodepocalypse Now Lang Chain4j vs Spring Al
01h 00mSpring AI