AI programmes for organisations that cannot afford to get it wrong.
Terence Kok designs and delivers structured AI engagements for government ministries, national infrastructure operators, and large organisations — where governance, auditability, and multi-jurisdiction compliance are prerequisites, not afterthoughts.
Organisations where AI failures have consequences beyond the balance sheet.
Government programmes, national infrastructure, and regulated enterprise environments share a common requirement: AI systems that are auditable, governable, and defensible under regulatory scrutiny — not just functional in a sandbox. The work here is built for that standard.
Government ministries and agencies
National AI strategy, sector-specific governance standards, and AI programme design for public-sector delivery environments.
Infrastructure operators and developers
AI-first operational architecture for airports, smart cities, utilities, and national infrastructure — designed before go-live, not retrofitted.
Multi-regional enterprises
Organisations deploying AI across multiple business units, regulatory environments, or jurisdictions — where a single compliance stack is required across different legal contexts.
Regulated sectors
Financial services, engineering, healthcare, and other sectors where AI deployments require documented compliance with sector-specific regulatory frameworks.
Selected deployments across government and enterprise.
All projects are publicly announced or verifiable through the referenced organisations. Detailed case studies are available on the Projects page.
Meinhardt AI Centre of Excellence
Domain-specific RAG platform deployed across 6,000+ engineers in six regional offices. Governance model adopted as the Group template for all subsequent AI deployments. Zero security incidents.
Global AI Governance Framework
Single governance standard satisfying Singapore IMDA, Saudi NDMO, and Oman TRA simultaneously — without separate compliance stacks. Referenced in the Group's sustainability report as evidence of responsible AI practice.
NEOM Bay Airport Digital Strategy
AI-first operational architecture for a greenfield international airport — passenger flow, baggage, and airside management designed before opening. Zero-legacy architecture adopted as the reference model for future NEOM infrastructure builds.
Sultan Haitham City IOCC
Federated data architecture connecting traffic, utilities, emergency services, and civic systems into a single operational view. Multi-agency coordination protocol adopted as the national standard for new city developments in Oman.
DIFC Enterprise Campus Network Review
Independent peer review establishing the technical and regulatory baseline for DIFC 2.0 network infrastructure. Assessed against DIFC ICT Law, UAE TDRA, DESC, and SIRA standards. Delivered RFP-ready procurement outputs.
Infrastructure Capability Diagnostic Framework
Four-phase diagnostic framework informing over US$60B in infrastructure investment decisions. Twenty-plus improvement opportunities identified across six domains. Framework adopted as the basis for a continuous technical excellence programme.
Three structured programme types. All scoped individually.
Engagements are not off-the-shelf. The following descriptions define what each programme type produces and the typical parameters — not a fixed scope.
AI Governance Framework Design
A single governance standard that satisfies multiple regulatory environments simultaneously — without separate compliance stacks per jurisdiction. Covers decision authority, audit trails, agentic AI controls, data handling, and jurisdiction-specific annexes. Designed to hold up under regulatory review, not just internal sign-off.
The approach unified Singapore IMDA, Saudi NDMO, and Oman TRA requirements under one framework at Meinhardt Group — enabling legal sign-off on agentic deployments across three regulatory environments without duplicating compliance infrastructure.
- Typical duration
- 8–14 weeks
- Deliverables
-
- Unified AI governance standard
- Jurisdiction-specific compliance annexes
- Regulatory mapping documentation
- Decision authority matrix
- Audit trail requirements and structure
- Staff training materials
- Engagement lead
- Terence Kok — direct engagement on all deliverables
- Working format
- Remote, on-site, or hybrid depending on data sensitivity requirements
AI Centre of Excellence Setup
Design and deployment of an internal AI capability function — from operating model and governance structure through to technical platform architecture and team operating procedures. Built for organisations deploying AI across multiple business units or regional offices, where consistency and governance cannot rely on individual judgement calls.
The Meinhardt AI Centre of Excellence (KELIX platform) deployed across 6,000+ engineers in six regional offices with zero security incidents. The operating model and governance framework became the Group template for all subsequent AI deployments.
- Typical duration
- 12–24 weeks depending on scale
- Deliverables
-
- CoE operating model and governance structure
- Technical platform architecture specification
- Tool selection and vendor evaluation
- Domain-specific knowledge architecture
- Team operating procedures and escalation protocols
- Platform deployment and handover documentation
- Engagement lead
- Terence Kok — direct engagement; Meinhardt Group specialist resources available for larger programmes
- Working format
- On-site preferred for initial phases; hybrid for delivery and handover
Digital Infrastructure Strategy
AI-first operational architecture for infrastructure programmes — where the measurement framework, data collection schema, and governance model are designed before systems go live. Covers passenger and logistics operations, smart city control systems, IoT architecture, and multi-agency integration.
Applied at NEOM Bay Airport (greenfield airport operational intelligence architecture), Sultan Haitham City IOCC (multi-agency national standard, Oman), SOHAR Smart City, and the DIFC campus network review (RFP-ready regulatory-aligned infrastructure baseline).
- Typical duration
- 8–20 weeks depending on scope
- Deliverables
-
- AI-first operational architecture specification
- Measurement framework and KPI design
- Data collection schema and governance
- Implementation roadmap with dependency sequencing
- Independent technical review (if applicable)
- Procurement-ready requirements documentation
- Engagement lead
- Terence Kok — direct engagement; infrastructure specialist resources available for technical review phases
- Working format
- Determined by programme requirements; on-site for site-specific infrastructure work
Four steps from initial contact to engagement.
No commitment is required at any stage before the proposal. The initial conversation is designed to establish fit, not to close a deal.
Initial conversation
No commitment, no NDA required. A 30–60 minute call to describe your programme brief and establish whether the engagement type and available timeline are a fit. Most initial conversations can be arranged within five business days.
Scoping session
A structured session to define scope, deliverables, timeline, team composition, and working format. Conducted under mutual NDA. Output is a shared scoping document that forms the basis of the proposal.
Written proposal
A detailed proposal covering scope, deliverables, timeline, team composition, data handling terms, and engagement pricing. Proposals are issued within five business days of the scoping session.
Engagement
Work begins on agreed terms. Weekly progress reporting is standard on all engagements. Milestone reviews are built into the engagement structure for programmes exceeding eight weeks.
Confidentiality and data handling terms
- Initial conversations are held in confidence without a formal agreement in place.
- A mutual NDA is executed before the scoping session. Engagement documentation, client data, and programme materials are covered from that point.
- On-site working is available for government and regulated environments where document transfer restrictions, security classifications, or data residency requirements apply.
- Client data and engagement materials are not used for AI model training, third-party publications, or reference cases without explicit written consent. Published case studies reference only publicly announced programmes.
- Data handling appendices aligned to specific regulatory frameworks (e.g. PDPA, GDPR, sector-specific requirements) are available on request.
Reference contacts and verification
References from past government and enterprise engagements are available on request following execution of a mutual NDA. Reference contacts are provided at the proposal stage for programmes above a minimum engagement threshold.
Publicly verifiable references — including the Meinhardt Group Annual Sustainability Report, Singapore IMDA governance framework alignment, and publicly announced NEOM and Oman government programmes — are cited in each project case study and do not require NDA access.
View publicly verifiable case studies →The initial conversation is without commitment.
Describe your programme brief in a few sentences. You will hear back within two business days with availability for an initial call.
Start a Conversation