← back
Memory Masterclass: Make Your AI Agents Remember What They Do! — Mark Bain, AIUS
Takeaway
Graph-structured memory preserving causal relationships, not just embeddings, is the missing primitive for trustworthy agentic recall and reasoning.
Summary
- Workshop on AI memory featuring Vasilije Markovic (Cognee), Daniel Chalef (Graphiti/Zep), Alex Gilmore (Neo4j), introducing a Graph Chat Arena
- Argues memory = any data including code/algorithms/hardware plus causal changes; LLM weights+biases are themselves memory structures
- Frames hallucinations as inevitable consequences of insufficient memory/compute for combinatorial problem space — fuzziness as a feature
- Core claim: graph-based memory preserves causal links via relationships, distinguishing GraphRAG from naive/hybrid RAG and enabling hallucination mitigation
memorygraph-ragagents
Original description
Are you ready to give your AI agents a memory upgrade? Join us for a fast-paced workshop exploring how memory can transform your agents. What You'll Do: Learn Leading Memory Solutions: Gain practical experience with open-source tools like Neo4j, Cognee, Graphiti, and Mem0. Explore Memory Types: Understand the theory behind long-term, short-term, episodic, semantic, and other memory types. Discover Memory Benefits: Learn how memory improves recall, contextual awareness, and reasoning in autonomous agents. Compare Implementations: Get a snapshot of how different solutions implement memory—what’s built-in, flexible, and experimental. We'll also demonstrate GraphRAG memory solutions and a GraphRAG chat implemented with Google ADK. Whether you’re working on AI copilots, agentic workflows, or research prototypes, this workshop will help you embed real memory into your AI stack. About Mark Bain I'm a deep tech founder building a private research lab: AIUS Technologies. We're on a mission to develop artificial life through R&D of long-term memory. Previously I led a cybersecurity lab and an edtech startup. Worked with defense and national security clients, banks and Fortune 500 clients. Recorded at the AI Engineer World's Fair in San Francisco. Stay up to date on our upcoming events and content by joining our newsletter here: https://www.ai.engineer/newsletter