Agents, Tools, and MCP, oh my! Next-level AI concepts for developers
summary
Spring Boot app uses OpenAI for reasoning, Neoforj as the Neoforj vector store, and five hard‑coded tools (three repository queries + RAG + graph RAG) so an agent selects from five tools to answer natural language queries. MCP is added as a client/server layer exposing MCP read_cipher tool and get_schema so the agent can run text->Cypher without bespoke code. Demos run: count books = 5, top reader = 486, Agatha Christie = 95 books; memory is planned but not yet implemented.
AI-written summary · about
chapters (13)
- 0:00Welcome
- 1:58Tech systems analogies
- 5:33Agentic workflows focus
- 8:59Traditional systems queries
- 11:06RAG limitations
- 16:18AI agents purpose
- 21:36Agent system components
- 24:06Building agents with tools
- 29:04Agent challenges
- 30:27Model Context Protocol MCP
- 33:59MCP architecture
- 37:11MCP integration demo
- 42:52Next steps
related talks
53mAgents, Tools, and Mcp, Oh My! Next Level AI Concepts for Developers
01h 01mAI Agents In-Depth – Function Calling, MCP and Tool Use Under the Hood
01h 28mAI Coding Agents and how to code them
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP
30mComposable AI: Building Next-Gen AI Agents with MCP
01h 03mAI Coding Agents and how to code them
27mBuilding AI Agents with Spring & MCP
52mBeyond local tools: Deep dive into MCP with Spring AI