Building full stack AI chatbots with Java - livecoding
summary
Live-coded a working Java full‑stack chatbot: Spring Boot backend, Telegram client, Dockerized Solr search index and MariaDB persistence, plus Metabase for visuals. Uses Stanford NLP for sentence-level sentiment, Lingua for language detection, Jsoup to scrape Wikipedia when Solr has no hit, and auto‑indexes extracted passages. Also wires Spring AI to OpenAI for generative answers. Demo shows the bot answering greetings, querying Solr for “Devoxx”, scraping and indexing Wikipedia, then replying—end‑to‑end functional.
AI-written summary · about
chapters (10)
- 0:00Speaker introduction
- 2:01Project goal and overview
- 4:05Architecture and tools
- 7:00Environment setup
- 15:57Architectural design
- 19:37Telegram integration
- 26:08Agent decision logic
- 39:20Knowledge base implementation
- 51:24Persistent storage
- 53:04Generative AI integration
related talks
58mBuilding Your Own Private GenAI Chatbot With Java Tools – A Step-by-Step Adventure
27mBuilding AI Agents with Spring & MCP
34mBuilding a Custom AI Chatbot with Spring Boot, React, and LangChain4j
28mPaving a Painless Path to Production AI in Your Java Applications
46mA LAMP Stack Professional’s Journey to AI
51mIntelligent Applications with Spring AI
46mBuilding a custom AI chatbot with Spring Boot, React, and LangChain4j
01h 00mSpring AI