01Foundation
Governance
Principles, decision rights, controls, accountability, evidence, and editorial integrity.
Enterprise outcomeMake responsibility and risk decisions visible.Inspect component
Framework
One connected system for putting enterprise AI into practice.
EAINE brings governance, knowledge, engineering, architecture, delivery, controls, reuse, and adoption together so teams can see and improve the whole system.
Cross-cutting assuranceEvidence + human accountability across every layerEnterprise intent remains traceable to engineering decisions
- Layer 01Foundation
Clear authority and a shared language
01GovernanceFoundation02EBOKFoundation - Layer 02Operating core
Turn strategy into controlled, practical work
03Engineering ModelOperating model - Layer 03Delivery system
Manage delivery, platforms, and controls
04AI SDLCLifecycle05Reference ArchitecturePlatform06StandardsAssurance - Layer 04Scale + learning
Reuse what works and keep improving
07PatternsReuse08PlaybooksAdoption
Operational signalsControlled learningUpdates governance, knowledge, standards, patterns, and playbooks↺
Eight components
Each part solves a clear problem, and all eight work together.
Organisations can adopt EAINE in stages. The connected design keeps governance, architecture, delivery, and operations aligned.
02Foundation
EBOK
A shared body of knowledge, concepts, domains, taxonomy, and language.
Enterprise outcomeGive every role a common engineering vocabulary.Inspect component
03Operating model
Engineering Model
Capabilities connecting strategy, delivery, platform, governance, and learning.
Enterprise outcomeTurn AI ambition into an enterprise operating system of work.Inspect component
04Lifecycle
AI SDLC
Nine governed stages from discovery through operations and continuous improvement.
Enterprise outcomeEngineer AI behaviour through traceable lifecycle evidence.Inspect component
05Platform
Reference Architecture
Control points for knowledge, models, agents, tools, identity, evidence, and observability.
Enterprise outcomeCreate reusable platform paths without hiding accountability.Inspect component
06Assurance
Standards
Minimum expectations for AI SDLC, governance, evaluation, security, prompts, agents, and operations.
Enterprise outcomeSet a consistent floor while allowing risk-based depth.Inspect component
07Reuse
Patterns
Repeatable approaches, tradeoffs, failure modes, and anti-patterns.
Enterprise outcomeConvert repeated experience into reusable engineering knowledge.Inspect component
08Adoption
Playbooks
Practical implementation paths, review questions, templates, and 90-day guidance.
Enterprise outcomeHelp teams start with control and learn through delivery.Inspect component
EAINE does not add governance after engineering. It makes accountability, evidence, and learning part of how engineering is performed.
One operating record
Five traces connect the eight framework components.
The components provide reusable guidance. The five traces keep one consequential outcome reconstructable from intent through production learning.
Intent
What outcome matters, for whom, and within which constraints?
Authority
Who may decide, approve, act, stop, and answer?
Behaviour
Which exact configuration produced the outcome?
Evidence
What justified release, and what remains unknown?
Learning
What did production change for the next decision?
Applicability before control depth
Record where AI is used and how much risk it creates.
Software does not become AI-native simply because AI helped build it. EAINE records AI use and risk separately, so teams apply the right controls.
Engineering-use axis
E0–E4 records whether AI assists, produces, or executes engineering work.
Runtime-use axis
R0–R4 records whether delivered AI informs, recommends, or executes material actions.
Composable profiles
Seven profiles add controls for engineering use, internal or embedded AI, high impact, agents, custom models, and third parties.
Inherent risk
Consequence and exposure set a base tier; mandatory floors prevent critical factors from being averaged away.
System connections
The real value comes from connecting the parts.
Strategy to delivery
Business outcomes and risk appetite become use-case decisions, requirements, architecture, and measurable release criteria.
Governance to evidence
Policies become engineering controls with owners, artefacts, gates, and operational signals.
Platform to practice
Shared capabilities support approved delivery paths without removing team or human accountability.
Operations to learning
Real behaviour, incidents, cost, and feedback update tests, systems, standards, patterns, and investment.
Lockstep alignment map
Every framework component maps directly to book coverage and architecture control.
If this map changes, it should be mirrored in the book chapters, RA control table, and the relevant website entry pages.
Framework as operating model
Book anchor: Part I + Chapter 13, Architecture as a Control System
Website anchor: /framework
Control proof: DOM-002, DOM-007, DOM-010, and lifecycle stage linkage in controls JSON.
AI engineering platform and governance controls
Book anchor: Chapter 14, 15, 18
Website anchor: /ai-sdlc
Control proof: CMP-002, CMP-005, CMP-006, CMP-007, CMP-008, IF-004 to IF-007.
Agent, tool, and decision boundaries
Book anchor: Chapters 9, 20
Website anchor: /evidence
Control proof: CMP-006, CMP-010, CMP-011, IF-005, DEC-014, DEC-018.
Knowledge and context supply chain
Book anchor: Chapter 19, Knowledge and Context Architecture
Website anchor: /library
Control proof: DOM-004, CMP-007, IF-004, REC-004, REC-005.
Operations and continuous improvement
Book anchor: Chapters 16, 17, 22
Website anchor: /adoption
Control proof: DOM-009, DOM-010, CMP-012, CMP-013, CMP-015, REC-015, REC-016.
Role alignment
One shared model, with a clear view for every role.
Everyone works from the same system while focusing on the decisions they own.
| Role | Primary question | EAINE starting point |
|---|---|---|
| CTO / CIO | How do we scale AI value without creating invisible enterprise risk? | Executive brief, maturity profile, 90-day adoption path |
| Engineering Leader | How do teams deliver faster while preserving quality and accountability? | AI SDLC, minimum standards, team evidence model |
| Enterprise Architect | Which shared capabilities and control points should the platform provide? | Reference architecture, traceability matrix, engineering model |
| Security / Risk Leader | Where are AI risks controlled and how can the organisation prove it? | Applicability and risk standards, security standard, evidence traceability |
| AI Engineer | What does good engineering look like beyond a working demonstration? | Lifecycle stages, evaluation, prompt/context, agent, and observability standards |
| Practitioner / Reviewer | How can I apply, challenge, and strengthen the discipline? | Worked examples, templates, patterns, review guide |