AI LLM Technology Architecture Associate Director

About the Role

Accenture Middle East is seeking an AI LLM Technology Architecture Associate Director to join their team in Riyadh, Saudi Arabia. This full-time role requires over 10 years of experience and involves serving as a definitive technical authority on AI architecture within client engagements and across the practice. The Associate Director will own the complete, end-to-end architecture of advanced AI platforms and solutions, ensuring cohesive design, technical soundness, and alignment with client business objectives and enterprise-grade standards.

Key Responsibilities

This position operates at an executive level, partnering with CIOs, CTOs, and senior business leaders to shape enterprise AI strategy. The role involves connecting business goals to a coherent technical vision, ensuring AI investments deliver competitive advantage. Key responsibilities include:

• Partnering with CIOs, CTOs, and business leaders to shape enterprise AI strategy and connect business goals to a coherent, sequenced technical vision.
• Leading enterprise AI assessments and building implementation roadmaps that sequence investments for lasting competitive advantage.
• Owning the complete, end-to-end technical solution for complex AI platforms, ensuring cohesive design and alignment with business objectives and enterprise standards.
• Translating governing architecture principles into concrete, defensible technical solutions for domain teams.
• Building innovative prototypes and proofs of concept hands-on, using emerging technologies to de-risk decisions and prove early value.
• Performing technology assessments and comparisons, providing evidence-based recommendations on tools, frameworks, and platforms.
• Setting the architectural direction for model- and tool-agnostic multi-agent ecosystems, including orchestration, memory, and tool/skill use, governed through a registry-bound AI Gateway.
• Establishing the agent registry and certification model to ensure only certified agents reach production.
• Defining memory as a first-class abstracted platform service, decoupled from underlying vendor engines.
• Defining the foundation model and inference strategy, including adaptation, fine-tuning, and dynamic cost/quality/latency-aware routing.
• Setting standards for high-throughput, low-latency inferencing and classical ML deployment within unified, production-ready platforms.
• Owning the architecture of the enterprise context layer, including knowledge graphs, ontologies, vector search, and semantic retrieval.
• Setting the design direction for context assembly and memory that manages prompts, context windows, and conversational state across the platform.
• Ensuring security, governance, observability, performance, and scalability are addressed holistically and consistently across every domain.
• Establishing the identity and authorization model, including per-agent identity, IAM/IAP binding, and defense-in-depth enforcement.
• Defining the layered guardrail framework applied at every boundary, balancing protection with performance.
• Governing the MCP control plane, including registry, gateway, and risk scoring, across all internal and third-party servers.
• Mandating adopt-over-build for productized evaluation and observability stacks.
• Establishing FinOps as a first-class concern, including usage labeling, gateway-enforced budgets, and cost-per-archetype as a planning input.
• Making definitive decisions on design patterns, reference architectures, frameworks, and technology selections, balancing innovation with pragmatism.
• Leading and integrating the work of domain architects and specialists, resolving cross-domain tensions into a unified, enterprise-ready system.
• Building the practice's reusable reference architectures, frameworks, and assets, with a deliberate adopt-over-build stance.
• Conducting deep-dive architecture workshops and working sessions with client executives and engineering teams.
• Producing and governing authoritative architecture artifacts, such as blueprints, reference architectures, ADRs, and integration specifications.
• Serving as a recognized thought leader in AI, shaping the practice's point of view and representing the firm externally through publications and conference engagements.

Required Qualifications

• Proven experience in designing and deploying enterprise-grade advanced AI solutions using agentic, generative, and classical AI/ML with at least one cloud vendor.
• Proven experience in the LLM and Generative AI space.
• Proven experience architecting and operationalizing LLM-driven application architecture patterns.
• Proven experience in engineering, machine learning, deep learning, and NLP solutions and applications.

Experience and Location

This role requires over 10 years of relevant experience. The position is full-time and based in Riyadh, Riyadh, Saudi Arabia.