Agents, Tools, and Mcp, Oh My! Next Level AI Concepts for Developers
summary
A Spring Boot demo wires an LLM to a Neo4j-backed vector store and executes vector search plus a graph RAG traversal to return context-aware book recommendations; tools are exposed via Model Context Protocol (MCP) with get_schema and read_cypher calls so the agent validates queries before running them. The app counts a user’s five books, returns author rankings (Agatha Christie 95 in the dataset), persists short- and long-term memory in the graph, and shows MCP makes tool swapping and deterministic tool results practical.
AI-written summary · about
chapters (13)
- 0:00Introduction and speaker background
- 2:14Complexity in modern AI stacks
- 6:05Limitations of plain LLMs
- 11:20What are AI agents?
- 13:45Evolution of structured tool calling
- 19:39Demonstration of agentic tool usage
- 29:39Context and memory in AI systems
- 36:58Short-term and long-term memory
- 40:26Challenges in agent scaling
- 41:10Model Context Protocol (MCP) explained
- 44:21MCP in the demo application
- 51:39Composable layers of AI systems
- 52:41Q&A and closing remarks
related talks
45mAgents, Tools, and MCP, oh my! Next-level AI concepts for developers
29mBuilding Modular AI Agents with LangGraph, MCP, and
54mFrom Single-Shot LLMs to Intelligent Agents: Building Scalable AI Systems with Spring AI and MCP
46mMCP for LLM Agents, Infrastructure APIs & Data Services
30mThe Fastest Path To Building An Agent For Your Knowledge Graph Is By Using MCP
50mBeyond local tools: Dive into the Model Context Protocol (MCP)
52mBeyond local tools: Deep dive into MCP with Spring AI
40mBuilding AI Agents with Spring & MCP