World Wide Technology (WWT) — a global technology solutions provider with over US$20 billion in annual revenue, 10,000+ employees, and operations across Americas, EMEA, and APAC, recognized as the eleven-time NVIDIA Partner of the Year for AI and Deep Learning — is seeking an exceptional End-to-End AI Infrastructure Architect for its Abu Dhabi, UAE operations. This is a named Design Authority role on WWT’s most complex AI infrastructure programmes in EMEA — responsible for integrating compute, networking, storage, facilities, platform, and security into coherent architectures that perform when thousands of GPUs train models together. The role requires 15+ years of infrastructure and solution architecture experience, 3–5+ years in AI/ML or HPC infrastructure at scale, TOGAF or equivalent framework experience, and deep vendor-neutral evaluation capability across GPU compute (NVIDIA, AMD), networking, storage, and platform.
About WWT — World Wide Technology | 11× NVIDIA Partner of the Year
Company: World Wide Technology (WWT) — US$20B+ Global Technology Solutions Provider | 10,000+ Employees | Americas · EMEA · APAC | Founded: 1990
NVIDIA Partnership: Eleven-time NVIDIA Partner of the Year for AI and Deep Learning — unmatched depth in AI infrastructure at enterprise scale
AI Infrastructure Practice: Designs, deploys, and operates the physical and logical foundations of enterprise AI — GPU compute clusters · High-performance networking · Storage architectures · Power and cooling · AI platform software
Vendor Partners: NVIDIA · HPE · Dell · Cisco and leading AI infrastructure vendors as strategic partners — direct vendor access for design validation
Client Base: Fortune 500 enterprises and regulated industry clients — the most demanding, highest-consequence AI infrastructure programmes in EMEA
Why This WWT End-to-End AI Infrastructure Architect Role Is the Career Opportunity of 2026
Named Design Authority Status: Being the named Design Authority on WWT’s most complex EMEA AI infrastructure programmes — the single technical accountability point for end-to-end architectural integrity across GPU superclusters, liquid cooling, InfiniBand fabrics, and AI platforms — is one of the most professionally prestigious AI infrastructure architecture titles available in the global technology market in 2026
Most Advanced AI Deployments in EMEA: WWT’s EMEA AI infrastructure pipeline includes GPU superclusters, liquid cooling systems, InfiniBand fabrics, and sovereign AI deployments at scales and specifications that represent the absolute frontier of enterprise AI infrastructure engineering
NVIDIA Partner Access: Eleven-time NVIDIA Partner of the Year status gives WWT’s architects direct access to NVIDIA’s engineering team, early access to DGX SuperPOD and HGX reference architectures, and the vendor relationship depth that enables genuinely differentiated AI infrastructure design
Tax-Free Abu Dhabi: Highly competitive senior architecture compensation — zero personal income tax in Abu Dhabi, UAE
Position Overview — The Integrator Who Prevents AI Infrastructure Failure
Most AI infrastructure failures happen at the seams — where compute meets network, where storage meets platform, where facilities meet hardware. This End-to-End AI Infrastructure Architect role at WWT exists to be the integrator who prevents this. As the named Design Authority on WWT’s most complex AI infrastructure programmes in EMEA, you will integrate compute, networking, storage, facilities, and platform layers into unified AI infrastructure architectures that function coherently when thousands of GPUs need to train a model together. You will map AI infrastructure architectures to client business outcomes including cost-per-training-run, time-to-inference, and utilization efficiency; evaluate and recommend vendor-neutral component stacks across GPU compute (NVIDIA, AMD), networking, and storage; design reference architectures for training clusters, inference farms, and hybrid training/inference deployments; coordinate between domain specialists across networking, storage, MEP, and platform to resolve cross-domain design conflicts; conduct architectural reviews and risk assessments at programme milestones; and present architecture decisions and trade-offs to client CTO and VP-level stakeholders. The role requires the systems-level thinking that understands how component interactions affect overall system performance at GPU supercluster scale.
Why This WWT End-to-End AI Infrastructure Architect Role Is the Most Prestigious AI Architecture Career of 2026 in Abu Dhabi: End-to-end AI infrastructure architects who combine 15+ years of compute, networking, and storage infrastructure architecture experience with 3–5+ years of AI/ML or HPC infrastructure platform design at scale, TOGAF or equivalent framework certification, proven design authority experience on large-scale programmes, NVIDIA DGX/HGX SuperPOD reference architecture familiarity, Kubernetes and Slurm AI platform software understanding, and the CTO-level client stakeholder communication skills to present architectural trade-offs with authority and clarity — represent the absolute pinnacle of the enterprise AI infrastructure talent market globally. WWT’s 11× NVIDIA partnership, EMEA AI infrastructure portfolio, and Abu Dhabi’s extraordinary AI investment environment make this the most technically significant and professionally career-defining AI infrastructure architecture opportunity in the region in 2026.
Key Responsibilities
End-to-End AI Infrastructure Integration — Named Design Authority
- Integrate compute, networking, storage, facilities, and platform layers into unified AI infrastructure architectures — serving as the systems-level integrator who understands how the choices made in each domain affect the performance, reliability, and scalability of the overall system, preventing the cross-domain mismatches (compute-network bottlenecks, storage-platform incompatibilities, facilities-hardware thermal mismatches) that cause AI infrastructure to underperform or fail at the seams where domain boundaries meet in ways that no single domain specialist can detect or prevent alone
- Serve as the named Design Authority on major WWT AI infrastructure programmes — taking formal, accountable technical ownership of architectural integrity from initial design through delivery and operation on WWT’s most complex EMEA AI infrastructure programmes, providing the single technical accountability point that ensures architectural decisions are made with full systems-level understanding rather than domain-by-domain optimization that creates unexpected cross-domain problems
- Conduct architectural reviews and risk assessments at programme milestones — applying the TOGAF-structured architectural review methodology and AI infrastructure domain knowledge to systematically evaluate programme designs at each stage gate, identifying architecture risks, unresolved design conflicts, incomplete specifications, and vendor commitment gaps that represent material delivery risk before they become programme-disrupting problems during hardware installation or system commissioning
- Design reference architectures for common AI infrastructure patterns — producing the documented, validated, and vendor-qualified reference architecture blueprints for GPU training clusters, inference farms, hybrid training/inference deployments, and sovereign AI installations that give WWT’s delivery teams the proven architectural starting points that accelerate programme design quality and reduce the risk of novel architecture selection on each new programme
GPU Compute, HPC Networking & Storage Architecture
- Evaluate and recommend vendor-neutral component stacks across GPU compute, networking, storage, and platform — applying genuine multi-vendor technical depth to compare NVIDIA DGX H100/H200 and HGX configurations against AMD Instinct-based alternatives, evaluate InfiniBand HDR/NDR fabric architectures against RoCE Ethernet alternatives, assess NVMe-oF and Lustre parallel file system storage architectures against their alternatives, and make technically grounded architecture recommendations that serve the client’s specific AI workload requirements and economic constraints rather than defaulting to a single-vendor preference
- Design GPU training cluster architectures — specifying the GPU node hardware (NVIDIA DGX/HGX or equivalent), inter-node high-bandwidth networking fabric (InfiniBand NDR, RoCE 400GbE), storage architecture (parallel file system, NVMe-oF), cluster management software (Slurm, Base Command Manager), and AI platform layer (Kubernetes, Kubeflow or MLflow) configurations that deliver the training throughput, model parallelism, and resource utilization efficiency that the client’s AI training workloads require
- Apply understanding of NVIDIA DGX SuperPOD reference architectures — leveraging familiarity with NVIDIA’s validated DGX SuperPOD blueprints (compute, networking, storage, and software stack specifications) to accelerate design validation and hardware compatibility verification on WWT programmes that adopt or adapt SuperPOD reference architectures for client-specific deployment contexts including data centre facilities, rack density constraints, power availability, and network topology requirements
- Design inference farm architectures — specifying the GPU compute configuration, batching and scheduling approach, model serving framework (Triton Inference Server, TensorRT-LLM, vLLM), load balancing architecture, and storage design that delivers the latency, throughput, and cost-per-inference performance requirements that production AI inference workloads demand, applying knowledge of inference-specific GPU selection trade-offs that differ significantly from the optimal hardware choices for model training
Facilities, Power, Cooling & Sovereign AI Considerations
- Map AI infrastructure architectures to facilities requirements — translating GPU compute cluster specifications into the specific power density, cooling capacity, rack weight, floor loading, and physical space requirements that data centre facilities teams need to provision or adapt the facility infrastructure that the AI cluster will occupy, identifying facility upgrade requirements early enough to address them within the programme schedule rather than discovering them during hardware delivery
- Apply understanding of advanced cooling technologies for high-density GPU deployments — designing the liquid cooling infrastructure (direct liquid cooling, rear-door heat exchangers, immersion cooling) required for high-density NVIDIA H100/H200 deployments where air cooling alone cannot sustain the thermal management performance that maintains GPU performance at rated specifications without throttling
- Address sovereign AI and air-gapped deployment requirements where applicable — designing the network isolation, data sovereignty compliance, supply chain verification, and operational security requirements that government and regulated industry AI infrastructure programmes in the UAE and wider EMEA market increasingly impose on AI infrastructure architecture and operational procedures
TCO Modelling, Client Stakeholder Engagement & Cross-Domain Coordination
- Map AI infrastructure architectures to client business outcomes — translating the technical architecture into quantified business impact metrics including cost-per-training-run across build, colocation, and cloud deployment models, time-to-inference latency and throughput projections, GPU utilization efficiency estimates, and total cost of ownership over the asset lifecycle, connecting architectural decisions to the commercial and operational outcomes that drive client investment decisions rather than treating architecture as a purely technical exercise
- Present architecture decisions and trade-offs to client CTO and VP-level stakeholders — communicating complex AI infrastructure architectural choices, inter-component trade-offs, risk assessments, and vendor evaluation rationales in the clear, business-outcome-framed presentation style that enables senior non-technical executives to make confident and well-informed AI infrastructure investment and procurement decisions
- Coordinate between domain specialists across networking, storage, MEP, and platform teams to resolve cross-domain design conflicts — managing the technical interfaces between specialist contributors whose domain-optimal design choices may create cross-domain incompatibilities, facilitating the structured technical design conflict resolution that produces coherent integrated architectures rather than domain-optimal but system-suboptimal component combinations
Qualifications & Requirements
Must-Have Requirements
- 15+ years in infrastructure and solution architecture — spanning compute, networking, and storage, with demonstrated progression into senior design authority roles on large-scale, complex technology infrastructure programmes
- 3–5+ years specifically designing AI/ML or HPC infrastructure platforms at scale — with production AI cluster or HPC supercomputer design and deployment experience that demonstrates the specific AI infrastructure domain depth this role requires
- TOGAF, SABSA, or equivalent architecture framework certification and demonstrated application on major programmes
- Vendor-neutral evaluation capability across GPU compute, networking, storage, and platform — with genuine multi-vendor technical depth rather than single-vendor familiarity
- Client-facing Design Authority experience on programmes of significant scale — presenting architecture decisions to CTO and VP-level client stakeholders with credibility and clarity
About WWT’s AI Infrastructure Practice & Abu Dhabi’s AI Ecosystem in 2026
World Wide Technology‘s position as the eleven-time NVIDIA Partner of the Year for AI and Deep Learning is not an honorary title — it reflects a depth of technical capability, a scale of AI infrastructure delivery experience, and a strategic vendor partnership with NVIDIA that gives WWT’s architects access to engineering resources, reference architectures, and technical validation support that independent architects and smaller integrators simply cannot replicate. In Abu Dhabi in 2026 — surrounded by the UAE’s extraordinary AI investment programs, anchored by MBZUAI, TII, G42, Microsoft’s UAE AI investment, and the UAE’s determination to establish itself as a global AI superpower — WWT’s AI Infrastructure Practice is engaged with some of the most consequential, most technically demanding, and most geopolitically significant AI infrastructure programmes available anywhere in the world. The End-to-End Architect who joins WWT in Abu Dhabi will be working on AI infrastructure that matters at a national scale — GPU superclusters that train foundation models, sovereign AI deployments that protect strategic data assets, and inference infrastructure that makes AI capabilities accessible to millions of users across the region. This is the AI infrastructure architect career that 2026 is offering to the professional who has spent fifteen years building toward it.
Your Career Growth Path: End-to-End AI Infrastructure Architect → Distinguished Engineer — AI Infrastructure → Chief AI Infrastructure Architect → VP AI Infrastructure Practice → Chief Technology Officer — AI — a globally recognized, technically elite, and professionally consequential AI infrastructure architecture career trajectory at the absolute frontier of enterprise AI infrastructure design and delivery.
Who Should Apply?
- Principal AI Infrastructure Architects (15+ Years): With NVIDIA DGX/HGX SuperPOD, InfiniBand fabric, and enterprise AI platform design authority experience at major technology consultancies, system integrators, or hyperscalers who want the most technically ambitious AI infrastructure architect role available in the EMEA market
- HPC Infrastructure Architects — AI Transition: With proven HPC supercomputer and high-performance networking design experience who are transitioning into enterprise AI infrastructure architecture and want to apply their systems-level architecture expertise within WWT’s AI Infrastructure Practice
- Enterprise Solution Architects — NVIDIA Ecosystem: With deep NVIDIA compute platform expertise, TOGAF certification, and Fortune 500 client design authority experience who want to lead AI infrastructure architecture on WWT’s EMEA portfolio of the most advanced GPU supercluster deployments
- Sovereign AI Infrastructure Architects — UAE/GCC: With experience designing air-gapped, sovereign, or regulated industry AI infrastructure deployments in the UAE or GCC who want to apply their sovereign AI architecture expertise within WWT’s Abu Dhabi operations on the region’s most consequential AI infrastructure programmes
- Data Centre & Cloud Architects — AI Infrastructure Specialization: With enterprise data centre and cloud infrastructure architecture backgrounds and deep AI infrastructure technical depth who want a named Design Authority role on WWT’s most complex EMEA AI infrastructure programmes with direct NVIDIA, HPE, Dell, and Cisco vendor partner access
AI Cybersecurity Red Team Expert Jobs Dubai 2026
Senior SAP BTP AI Integration Engineer Jobs Abu Dhabi UAE 2026
