Memory Masterclass: Make Your AI Agents Remember What They Do!
summary
Workshop demos show concrete tooling and tradeoffs: four live demos (some fallback to video because of Wi‑Fi) that ingest real data and write stateful graphs, a GitHub hiring demo that hired Laszlo with a PhD after agents built semantic links, a Neo4j setup where Neo4j MCP server creates entities and observations during chat sessions, and Zep/Graffiti temporal graphs where Zep models domain‑specific entities for a finance coach. Claim: preserving causal relationships in graph memory reduces hallucinations; a GraphRAG/GraphRack Chat Arena prototype lets teams switch agents and compare implementations.
AI-written summary · about
chapters (6)
- 0:00Introduction and memory talk
- 22:17Neo4j MCP server demo
- 27:33Temporal graphs for memory
- 35:39Agentic firewall use case
- 40:32GraphRAG chat arena demo
- 45:36Q&A session
related talks
29mThe AI Agent Memory Landscape
15mBuilding Evolving AI Agents Via Dynamic Memory Representations Using Temporal Knowledge Graphs
01h 34mHands-on AI workshop: Graph RAG, Memory & Multimodal Agents
01h 49mBuilding Memory Augmented Agents with MemoRizz
21mBeyond Recall: Creating Agentic Personas with Persistent, Evolving Memory
14mUsing Graphs to Fine-Tune Agents
58mAdvanced Autonomous AI Agents: Memory, Collaboration and Evolution in Action
26mStore Your AI Agent's Memory and Context in a Knowledge Graph