I built a few conversational agents that are privacy-focused, and we realized that we need a memory.

It is a compilation of my Medium articles on graph and memory topics. We need a memory to make the agent smart, and this book is not a solution, nor does it guide you on how to build a memory. Still, it explains how knowledge graphs and metagraphs are essential for memory and how to build a better memory for AI agents based on the graph approach. So it's a graph book, actually, but with the artificial intelligence flavour.

We will touch:

• Why is memory not RAG
• how to model AI memory
• graphs
• graph evolution for a memory
• meta and hypergraphs
• temporal aspect of memory
• event graphs
• casual graph
• semantic space time 

This book introduces a revolutionary framework for knowledge representation and AI agent memory: Semantic Spacetime. Drawing from theoretical physics and graph theory, this framework offers a new way to understand how meaning, relationships, and causality can be structured in intelligent systems.

Why This Book Matters

Current approaches to AI memory and knowledge representation face fundamental limitations. Vector embeddings, while popular, create opaque high-dimensional spaces where relationships lack clear semantic meaning. Traditional graph databases often rely on arbitrary relationship types that don't generalize across domains. Most critically, existing systems struggle with the dynamic, contextual nature of how humans actually understand and use knowledge.

Semantic Spacetime addresses these challenges by proposing four fundamental relationship types—NEAR/SIMILAR TO, LEADS TO, CONTAINS, and EXPRESSES PROPERTY—that can represent virtually any knowledge domain while maintaining semantic clarity and computational tractability.

What You'll Discover

This book explores how spatial and temporal concepts from physics can be adapted to create semantic spaces where meaning emerges from relationships. You'll learn how causality graphs can form the backbone of AI agent memory, enabling systems that don't just store information but understand the "why" behind events and decisions.

The framework presented here moves beyond static knowledge representation to embrace the dynamic, contextual nature of understanding. By focusing on causal relationships and pragmatic proximity, AI systems can adapt their knowledge structures to different contexts and purposes, much like human cognition.

For Whom This Book Is Written

This book is intended for researchers and practitioners working in AI, knowledge representation, graph databases, and semantic technologies. While the concepts are rigorous, they are presented with practical applications and implementation considerations in mind.

Whether you're building recommendation systems, developing AI agents for personal assistance, creating knowledge management platforms, or exploring the foundations of machine reasoning, the principles in this book provide both theoretical grounding and practical guidance.

The Journey Ahead

The framework presented here represents a synthesis of ideas from multiple disciplines: graph theory, category theory, physics, cognitive science, and computer science. By bringing these perspectives together, we can build AI systems that not only process information but truly understand the structured nature of knowledge and experience.

This is not just another approach to knowledge representation—it's a fundamental rethinking of how intelligent systems can model the world in ways that align with how humans actually think and reason about complex relationships and causality.

If you're following my work, I have a couple of deep research pieces about agentic memory and the application of spacetime concepts. Right now, I'm also working on a book about agentic protocols. This particular book focuses on the question of better context for agents—following the ideas of context graphs and exploring how to actually build something beyond context graphs to manage decision traces.

The goal is to create an architecture for enterprise-level agents that follow rules, make their own decisions, and—most importantly—make explainable decisions. I'll also explore the ability for agents to learn and apply that learning based on past experience.

We'll talk extensively about concepts like agentic memory: why we need memory, how to build memory, and why memory is not just another RAG system. We'll cover how to apply decision traces, how they work, and why cognitive processes—just like memory structures—contribute to learning capabilities.

Beyond this, we'll explore promise theory and promise graphs as an extension of agentic action logs. This creates a rich trace from data signals to promises to actions, making the architecture multi-agent ready. We'll examine this in the scope of agent cooperation, where agents don't just act in isolation but coordinate through explicit promises and commitments.

My core assumption with this book is that we need to build sophisticated agentic memory that goes beyond context graphs, beyond knowledge graphs, and applies advanced topics like causality, temporal causality research, and a deep focus on time itself. This is extremely relevant to my memory book, but here the focus is specifically on agents—and how they work together.

Empowering AI Agents with Graph-Based Memory on User Devices

This book addresses a critical gap in modern AI development: building truly autonomous agents with sophisticated memory systems that run entirely on user devices. While most AI applications rely on cloud-based processing and centralized knowledge stores, this book demonstrates how to architect intelligent agents with rich semantic memory using knowledge graphs, hypergraphs, and metagraphs—all operating locally on user devices.

What You'll Learn

From LLMs to Autonomous Agents: Move beyond simple prompt engineering to build AI systems that can remember, reason, and act autonomously. You'll discover why traditional language models fall short and how the three pillars of agent autonomy—tools, memory, and reasoning—work together to create truly intelligent systems.

Graph-Empowered Memory Architecture: Master the implementation of personal knowledge graphs as the foundation for AI agent memory. Learn practical approaches to modeling complex relationships, temporal data, and multi-modal information using relational databases, making sophisticated memory systems accessible without specialized graph databases.

Edge AI Implementation: Build AI agents that respect user privacy and data sovereignty by running entirely on personal devices. Discover how to implement vector search, graph queries, and complex reasoning using embeddable databases like LibSQL, enabling powerful AI capabilities without compromising user data.

Advanced Graph Structures: Progress from simple directed graphs to hypergraphs and metagraphs, understanding when and how to use each structure for maximum effectiveness. Learn practical strategies for handling temporal relationships, multi-context memory, and hierarchical knowledge representation.

Real-World Applications: Bridge the gap between theoretical knowledge representation and practical software development. Understand how to map ontological concepts to domain objects, implement Graph-to-Object Mapping (GOM), and integrate semantic reasoning with modern application architectures.

Who This Book Is For

This book is designed for software engineers, AI researchers, and technical architects who want to build the next generation of AI applications with sophisticated on-device capabilities. Whether you're developing personal AI assistants, knowledge management systems, or autonomous agents, this book provides the practical knowledge needed to implement graph-based memory systems that scale.

You should have basic familiarity with databases, software architecture, and AI concepts, though the book builds from foundational principles to advanced implementations.

Why This Matters Now

As AI regulation evolves and privacy concerns grow, the future belongs to systems that empower users with sovereign control over their data and AI capabilities. This book shows you how to build that future today, creating AI agents that are both powerful and privacy-preserving, sophisticated yet deployable on personal devices.

If you're following my work, I have a couple of deep research pieces about agentic memory and the application of spacetime concepts. Right now, I'm also working on a book about agentic protocols. This particular book focuses on the question of better context for agents—following the ideas of context graphs and exploring how to actually build something beyond context graphs to manage decision traces.

The goal is to create an architecture for enterprise-level agents that follow rules, make their own decisions, and—most importantly—make explainable decisions. I'll also explore the ability for agents to learn and apply that learning based on past experience.

We'll talk extensively about concepts like agentic memory: why we need memory, how to build memory, and why memory is not just another RAG system. We'll cover how to apply decision traces, how they work, and why cognitive processes—just like memory structures—contribute to learning capabilities.

Beyond this, we'll explore promise theory and promise graphs as an extension of agentic action logs. This creates a rich trace from data signals to promises to actions, making the architecture multi-agent ready. We'll examine this in the scope of agent cooperation, where agents don't just act in isolation but coordinate through explicit promises and commitments.

My core assumption with this book is that we need to build sophisticated agentic memory that goes beyond context graphs, beyond knowledge graphs, and applies advanced topics like causality, temporal causality research, and a deep focus on time itself. This is extremely relevant to my memory book, but here the focus is specifically on agents—and how they work together.

If you're following my work, I have a couple of deep research pieces about agentic memory and the application of spacetime concepts. Right now, I'm also working on a book about agentic protocols. This particular book focuses on the question of better context for agents—following the ideas of context graphs and exploring how to actually build something beyond context graphs to manage decision traces.

The goal is to create an architecture for enterprise-level agents that follow rules, make their own decisions, and—most importantly—make explainable decisions. I'll also explore the ability for agents to learn and apply that learning based on past experience.

We'll talk extensively about concepts like agentic memory: why we need memory, how to build memory, and why memory is not just another RAG system. We'll cover how to apply decision traces, how they work, and why cognitive processes—just like memory structures—contribute to learning capabilities.

Beyond this, we'll explore promise theory and promise graphs as an extension of agentic action logs. This creates a rich trace from data signals to promises to actions, making the architecture multi-agent ready. We'll examine this in the scope of agent cooperation, where agents don't just act in isolation but coordinate through explicit promises and commitments.

My core assumption with this book is that we need to build sophisticated agentic memory that goes beyond context graphs, beyond knowledge graphs, and applies advanced topics like causality, temporal causality research, and a deep focus on time itself. This is extremely relevant to my memory book, but here the focus is specifically on agents—and how they work together.

Empowering AI Agents with Graph-Based Memory on User Devices

This book addresses a critical gap in modern AI development: building truly autonomous agents with sophisticated memory systems that run entirely on user devices. While most AI applications rely on cloud-based processing and centralized knowledge stores, this book demonstrates how to architect intelligent agents with rich semantic memory using knowledge graphs, hypergraphs, and metagraphs—all operating locally on user devices.

What You'll Learn

From LLMs to Autonomous Agents: Move beyond simple prompt engineering to build AI systems that can remember, reason, and act autonomously. You'll discover why traditional language models fall short and how the three pillars of agent autonomy—tools, memory, and reasoning—work together to create truly intelligent systems.

Graph-Empowered Memory Architecture: Master the implementation of personal knowledge graphs as the foundation for AI agent memory. Learn practical approaches to modeling complex relationships, temporal data, and multi-modal information using relational databases, making sophisticated memory systems accessible without specialized graph databases.

Edge AI Implementation: Build AI agents that respect user privacy and data sovereignty by running entirely on personal devices. Discover how to implement vector search, graph queries, and complex reasoning using embeddable databases like LibSQL, enabling powerful AI capabilities without compromising user data.

Advanced Graph Structures: Progress from simple directed graphs to hypergraphs and metagraphs, understanding when and how to use each structure for maximum effectiveness. Learn practical strategies for handling temporal relationships, multi-context memory, and hierarchical knowledge representation.

Real-World Applications: Bridge the gap between theoretical knowledge representation and practical software development. Understand how to map ontological concepts to domain objects, implement Graph-to-Object Mapping (GOM), and integrate semantic reasoning with modern application architectures.

Who This Book Is For

This book is designed for software engineers, AI researchers, and technical architects who want to build the next generation of AI applications with sophisticated on-device capabilities. Whether you're developing personal AI assistants, knowledge management systems, or autonomous agents, this book provides the practical knowledge needed to implement graph-based memory systems that scale.

You should have basic familiarity with databases, software architecture, and AI concepts, though the book builds from foundational principles to advanced implementations.

Why This Matters Now

As AI regulation evolves and privacy concerns grow, the future belongs to systems that empower users with sovereign control over their data and AI capabilities. This book shows you how to build that future today, creating AI agents that are both powerful and privacy-preserving, sophisticated yet deployable on personal devices.