Enterprise Architecture
AI-assisted 8-layer enterprise architecture using a purpose-built meta-model. Richer, always current, lower cost than traditional EA.
Overview
Traditional EA approaches produce models that are costly to create, almost impossible to keep current, and rarely used for real strategic reasoning. They tend to focus on application and infrastructure layers, with little connection upward to the organisation’s motivation model. The result is architecture that describes what systems exist but cannot answer the questions that matter: which systems implement our regulatory obligations? Which capabilities support our strategic goals? Which critical services depend on a single ageing platform?
GoSource takes a different approach. We combine senior consulting expertise with an AI-assisted methodology grounded in a purpose-built meta-model. The meta-model blends concepts from the Business Motivation Model, BIZBOK, TOGAF, and ArchiMate into a single minimal schema — 32 entities across 8 layers — small enough for an LLM to reason over coherently, but complete across the full strategy-to-infrastructure chain. The result is enterprise architecture that is richer, always current, and dramatically cheaper to sustain.
We offer two modes of engagement: architecture assessment (extracting the current state from existing documentation into a knowledge graph, running diagnostic tests, and producing an evidence-based assessment) and architecture maintenance (establishing an ongoing low-cost capability to keep the model current as the organisation evolves).
Both modes use our DOGRAG methodology, which constrains AI outputs to the 32 entity types and approved relationships. The AI cannot invent new entity types or arbitrary relationships. For organisations that prohibit cloud-hosted AI, the methodology runs entirely on locally deployed open-weight models within the organisation’s security boundary.
The 8-Layer Meta-Model
| Layer | Question | Key Entities |
|---|---|---|
| 1. Motivation | WHY does the organisation exist? | Vision, Mission, Goal, Objective, Strategy, Initiative, Policy, Regulation, Risk |
| 2. Value Delivery | WHY do stakeholders care? | Stakeholder, Value Proposition, Service, Value Stream, Value Stage, Outcome |
| 3. Capability | WHAT must the organisation do? | Capability, Capability Level, Capability Owner, KPI |
| 4. Operating Model | HOW is work performed? | Process, Activity, Business Rule, Case, Event |
| 5. Organisation | WHO performs work? | Organisation, Organisation Unit, Role, Actor |
| 6. Information | WHAT does the organisation manage? | Business Object, Data Entity, Record, Identifier |
| 7. Application | WHAT systems are used? | Application, Application Service, Component, Interface |
| 8. Technology | WHAT infrastructure supports them? | Platform, Node, Technology, Network |
The primary traceability chain — Regulation → Policy → Strategy → Initiative → Capability → Value Stage → Process → Application → Technology — enables questions that cross all layers. Traditional EA tools stop at the application-infrastructure boundary. This model connects systems all the way up to the regulatory obligations they exist to serve.
The Assessment Method
- Ingest artefacts. Strategy papers, architecture diagrams, capability maps, system inventories, policy documents, org charts, process documentation.
- Extract entities. The LLM maps content to the 32 meta-model entity types, chunking documents into 2–4 page sections. Each extraction carries provenance, confidence, and state attributes.
- Normalise. Merge synonyms and near-duplicates. Embedding-based pre-filtering identifies candidates; the LLM makes the semantic judgement; uncertain merges are flagged for human review.
- Map relationships. Only relationships from the approved set. No free-form “relates to” edges.
- Run diagnostic tests. Coverage, alignment, duplication, and fragility tests across the knowledge graph. Most tests are deterministic graph queries requiring zero LLM reasoning.
- Produce the assessment. Executive summary, layer-by-layer assessment, findings register with evidence and risk ratings, and heat maps for capability maturity, application duplication, traceability coverage, and dependency concentration.
Every entity carries a state attribute (current, target, transitional, retired, unknown). This forces the distinction between what exists and what is hoped for.
Why This Approach
| Dimension | Traditional EA | GoSource AI-Assisted EA |
|---|---|---|
| Scope | Application + infrastructure layers | Full 8-layer model, regulation to infrastructure |
| Creation cost | Months of skilled modeller effort | Days of AI-assisted extraction + senior review |
| Maintenance | Expensive periodic refresh projects | Continuous low-cost incremental updates |
| Currency | Stale within months | Always current from living organisational artefacts |
| Tooling | Expensive commercial tools (Sparx, Mega, BiZZdesign) | Open-weight LLMs + graph database + Python |
| Strategic value | Describes what systems exist | Answers cross-layer strategic questions with evidence |
Principles
- Evidence-traceable, not opinion-based. Every entity, relationship, and finding points back to a source document. Every extraction carries a confidence tag (explicit, strongly implied, or inferred). Claims without provenance are flagged.
- Senior judgement, not automation. The AI performs extraction and diagnostic queries at scale. Senior architects provide the judgement — interpreting findings, assessing risk, advising on remediation.
- Data sovereignty by design. Sensitive architecture documentation never leaves the organisation’s network. Locally deployed open-weight models with constrained decoding guarantee structured output compliance.
Policy Alignment
- Australian Government Architecture Framework (AGA) — Aligns with AGA principles while providing richer traceability.
- TOGAF / ArchiMate — Entity types and relationships draw selectively from both frameworks. Organisations already using them will find the meta-model familiar.
- Australian Government Policy for Responsible Use of AI — Human oversight, transparency, and provenance tracking throughout.
Evidence
- Reference Framework: LLM-Optimised Enterprise Architecture Framework — 32 entities, 8 layers, assessment methodology, local model deployment guide.
- Case Study: Home Affairs Integration Strategy — Enterprise architecture principles applied to reduce 700 ESB integrations to 16 RESTful resources across 35 domains.
- Case Study: ABF Trade Modernisation — Modular future-state architecture with full business-to-technology traceability.
- Case Study: ABF Trade Modernisation — Enterprise strategy applying DDD, value metrics, and generative AI.
- Staff: Steven Capell — SFIA Level 6 enterprise and business architecture; 30+ years; Vice-Chair UN/CEFACT; architect of the meta-model.
Tools & Technologies
- Meta-model & Knowledge Graph: Custom 32-entity ontology, Neo4j or NetworkX, JSON entity/relationship stores
- LLM Deployment: Locally deployed open-weight models (Llama 3.3 70B, Qwen 2.5 72B, DeepSeek-R1 70B); constrained decoding via Outlines or llama.cpp grammars
- Diagnostic Queries: Python scripts, Cypher (Neo4j), deterministic graph traversal
- Architecture Documentation: C4 model, Mermaid diagrams, Architecture Decision Records