Managing Small Knowledge Graphs for Multi-agent Systems - MLOps Podcast #236

Explore a comprehensive podcast episode featuring Tom Smoker, Technical Founder of whyhow.ai, discussing the management of small knowledge graphs for multi-agent systems. Delve into the challenges of using RAG (Retrieval-Augmented Generation) in generative models, focusing on repeatability and accuracy issues in multi-agent pipelines. Learn how knowledge graphs can structurally ground agents and improve system reliability. Discover insights on integrating graphs and vectors, data integration for robust knowledge graphs, and strategies for scoped knowledge retrieval. Examine RAG limitations and potential solutions, the complexities of working with multi-agent systems, and the future of agent-based systems with modular processes. Gain valuable perspectives on balancing risk tolerance in company operations and leveraging AI for high-quality content generation.

[] Tom's preferred coffee
[] Takeaways
[] Please like, share, leave a review, and subscribe to our MLOps channels!
[] Academic Curiosity and Knowledge Graphs
[] Logician
[] Knowledge graphs incorporated into RAGs
[] Graphs & Vectors Integration
[] "Exactly wrong"
[] Data Integration for Robust Knowledge Graph
[] Structured and Dynamic Data
[] Scoped Knowledge Retrieval Strategies
[28:01 - ] LatticeFlow Ad
[] RAG Limitations and Solutions
[] Working on multi agents, questioning agent definition
[] Concerns about performance of agent information transfer
[] Anticipating agent-based systems with modular processes
[] Balancing risk tolerance in company operations and control
[] Using AI to generate high-quality, efficient content
[] Wrap up