The Missing Piece of AI Agents: Knowledge Infrastructure
The main reason AI agents fail is the lack of knowledge infrastructure in existing systems.
The main reason AI agents fail is the lack of knowledge infrastructure in existing systems.
I attempt to build a framework-agnostic conceptual model that covers the core components of LLM-based agent systems and the relationships between them.
Discussing Metadata Management and the MOF Concept.
In this post, we discuss how to make RAG pipelines more effective and production-ready.
A deep dive into Data Mesh: decentralized data ownership, data as a product, self-serve data platforms, and federated governance.
How does the human brain think and express thoughts? Can we simulate it with today’s technology?
In this post, I explain why memory structures are critical in agent architectures, detailing types from sensory to semantic memory and how they are practically applied.
How to make agent-based systems more scalable using MCP, semantic tool search and dynamic tool injection — with a real implementation example.
Where did my motivation to write come from? What will my posts be about? A small introduction.