Architecture Reference Diagrams

Overview

The Knowledge Architecture. How institutional knowledge is modeled as durable, versioned objects and typed relationships — a living graph rather than documents in folders.

Integration Architecture. How the Intelligence Layer connects to ERP, CRM, PLM, MES, collaboration, and operational systems without becoming another silo.

Deployment Models. Cloud, hybrid, on-premise, air-gapped, and sovereign deployments — with the trade-offs, industries, and controls that shape each choice.

The layered architecture

The platform is organized as four cooperating layers. Each depends on the one beneath it, and reasoning flows upward from raw sources to institutional judgment.

1. Business & Decisions. The decisions, judgments, and outcomes the organization is accountable for.
2. Institutional Intelligence Layer. Captures why decisions are made — judgment, rationale, and context — for people and agents.
3. Knowledge Architecture. Durable, versioned knowledge objects and typed relationships — a living graph.
4. Application & Agents. Workspaces, assistants, and autonomous agents that read and write the model.
5. Technology & Runtime. Model routing, local inference, compute, storage, and networking.
6. Deployment & Infrastructure. Cloud, hybrid, on-premise, air-gapped, and sovereign environments.

Core knowledge concepts

• Knowledge Objects. Durable, versioned units of institutional knowledge — an asset, a supplier, a decision, a standard — that persist independently of any document that describes them.
• Typed Relationships. Explicit, meaningful edges (depends-on, supersedes, owned-by, derived-from) that turn isolated facts into a graph a machine can reason over.
• Ontology. The shared vocabulary that normalizes every source, so the same entity referenced in a contract, a filing, and a sensor stream resolves to one object.
• Knowledge Fabric. The connective layer that unifies knowledge objects across systems of record into a single queryable model without copying them into a new silo.
• Evidence. Every assertion carries its provenance — the source, timestamp, and derivation path — so conclusions remain auditable and defensible.
• Institutional Memory. The accumulated, retained knowledge of the organization that survives reorganizations, turnover, and system migrations.
• Decision Graph. A structured record of decisions and their rationale, linking options considered, evidence weighed, and outcomes observed.
• Identity Graph. A resolved map of people, roles, teams, and agents that grounds access, accountability, and attribution across the model.
• Confidence. A first-class signal on assertions and derivations, letting the system distinguish established fact from working hypothesis.

Deployment models

Managed Cloud. Fastest to value, elastic scale, fully managed operations. Best fit: Rapid rollout, Elastic analytical workloads, Distributed teams.

Hybrid. Sensitive data and inference stay local; scale bursts to cloud. Best fit: Mixed sensitivity, Gradual migration, Cost/performance balance.

On-Premise. Runs entirely inside your data centers and network. Best fit: Strict residency, Existing data-center investment, OT proximity.

Air-Gapped. No external network connectivity; models and updates delivered offline. Best fit: Classified programs, Critical infrastructure, Highest-assurance sites.

Sovereign. Operated within a national or jurisdictional boundary you define. Best fit: National programs, Regulated data, Public-sector mandates.

Connected systems

• ERP. Financials, supply chain, and resource planning (SAP, Oracle).
• CRM. Customer, account, and pipeline systems (Salesforce, Dynamics).
• PLM. Product lifecycle and engineering bills of material.
• MES. Manufacturing execution and shop-floor systems.
• SCADA. Supervisory control and data acquisition for industrial processes.
• OPC-UA. Standardized machine and sensor telemetry from the plant floor.
• SharePoint. Document repositories and institutional content.
• Teams / Slack. Conversation and decision context from where work happens.
• Data Warehouse. Analytical stores and lakehouses (Snowflake, Databricks).
• Identity Provider. Federated identity and SSO (Entra ID, Okta).

Learn more

This reference summarizes material covered in full across the Architecture Center. For the complete treatment, see The Knowledge Architecture, Integration Architecture, and Deployment Models.