OntologiesFeatured
Ontologies for AI Agents
How to define the entities, relationships, and rules an agent reasons over, and how to hand that definition to the agent at runtime.
Guides for practitioners
Understand the architecture behind intelligent systems.
Clear, practical guides to AI stacks, ontologies, knowledge graphs, and the technologies connecting them.
- 03 · AgentsModels and agents acting on the system.
- 02 · KnowledgeOntologies and graphs that define meaning.
- 01 · DataStores, documents, and tables.
Topics
Six areas, one map
Each topic covers a layer of the same system. Start wherever your current problem lives.
AI Architecture
How models, retrieval, orchestration, and interfaces fit together into one system.
1 guideOntologies
Formal vocabularies that define the entities, relationships, and rules a system reasons over.
2 guidesKnowledge Graphs
Connected facts about entities that ground retrieval, reasoning, and explanation.
1 guideAI Infrastructure
Serving, scaling, observability, and cost for systems built on large models.
1 guideData Modeling
Structuring data so models, pipelines, and people can rely on the same definitions.
1 guideIntelligent Agents
Systems that plan, call tools, keep state, and act within defined boundaries.
2 guides
Featured guides
Start with the foundations
Long-form, diagram-first guides. Each one answers a specific set of questions and tells you where it stops.
- Knowledge GraphsFeatured
Knowledge Graphs for Grounded Retrieval
Designing, populating, and querying a knowledge graph that makes retrieval traceable and lets a model reason across more than one hop.
- AI ArchitectureFeatured
The AI Stack, Layer by Layer
A reference map of the modern AI stack, from data sources through the knowledge layer to models, orchestration, and interfaces.
How the guides are written
Information science, applied to AI.
The hard problems in AI systems are problems of definition, structure, and provenance. Those are old problems with good answers. The guides bring them to the current stack.
- 01
One architecture, drawn once
Every guide places its subject on the same layered reference diagram, so what you learn about ontologies connects to what you learn about agents.
- 02
Definitions before tools
We explain what a thing is and what question it answers before naming a product. Vendors change; the information architecture does not.
- 03
Written to be reused
Diagrams, checklists, and decision tables are designed to be lifted straight into your own design documents and reviews.
Know what your system knows.
Begin with the reference architecture, then go deep on the layer that is causing you trouble.