AI Solution Architect

The AI Solution Architect defines and governs the architecture of HOYA's common AI layer, the shared foundation on which divisions adopt machine learning, generative AI, copilots and agentic solutions securely, responsibly and at scale. The role translates business strategy, risk requirements and technology opportunities into architecture principles, platform guardrails, reference designs and reusable patterns. These span Azure AI Foundry, Azure OpenAI, Azure Machine Learning, data and knowledge services, orchestration, integration, identity, observability and MLOps/LLMOps. The architect leads architecture assurance for priority solutions, guides technology and build-versus-buy decisions, and defines controls for data protection, model and agent risk, evaluation and operational monitoring. The role works with group engineers, security, governance and division teams, including joint proofs of concept with divisions, to take AI solutions safely from experimentation to enterprise scale.

Job Responsibilities

• Collaborate with Enterprise Architecture to develop and maintain HOYA's enterprise AI target architecture, principles, standards, capability roadmap and reference technology patterns, aligned with the group AI strategy and enterprise architecture.
• Translate business, data, security, privacy, legal, regulatory and responsible AI requirements into enforceable architecture guardrails and design controls.
• Define reference architectures for predictive ML, generative AI, RAG, copilots, agents, multi-agent orchestration, intelligent workflows etc.
• Architect the Azure AI ecosystem across Azure AI Foundry, Azure Machine Learning, Azure AI Search, Microsoft Fabric, integration services, identity and monitoring capabilities.
• Define the end-to-end MLOps and LLMOps architecture, covering experimentation, datasets, model and prompt registries, evaluation, approvals, CI/CD, release, rollback, monitoring, retraining and retirement.
• Set enterprise evaluation standards for quality, accuracy, groundedness, relevance, robustness, bias, safety, security, latency, cost and business outcomes.
• Define secure RAG and agentic AI patterns covering content ingestion, retrieval, indexing, chunking, embeddings, vector search etc.
• Define architecture and governance patterns for identity, authorization, memory, tools, actions, human oversight, audit trails etc.
• Apply Agent Security Posture Management practices to agent inventory, discovery, ownership, permissions, data access, tool usage, behavioral monitoring, vulnerabilities and risk remediation.
• Lead architecture reviews for priority AI initiatives. Document decisions, risks and exceptions, guide remediation and approve production readiness against the defined guardrails.
• Guide technology selection, build-versus-buy decisions, vendor assessment, model choice and portability, while managing lock-in, cost, operability and lifecycle risk.
• Partner with group AI engineers and division teams during proofs of concept and division-led delivery, so that designs remain implementable, supportable and reusable.
• Work with security, privacy, legal, risk, data governance, enterprise architecture and division teams to evolve controls and scale approved patterns.
• Maintain architecture assets, decision records, reference implementations and technical standards, and build architecture capability across HOYA teams.

Qualifications

• Bachelor's degree in computer science, artificial intelligence, data science, engineering or a related discipline.
• At least 10 years of solution or enterprise architecture experience, including significant architecture responsibility for AI, machine learning or advanced analytics solutions.
• Demonstrated ability to influence executives and senior technical stakeholders, govern cross-division designs and communicate complex architecture decisions clearly.
• Microsoft certifications such as Azure Solutions Architect Expert, Azure AI Engineer Associate or Azure Data Scientist Associate.
• TOGAF or an equivalent enterprise architecture certification is beneficial.

Experience

• Extensive architecture and hands-on technical expertise across Azure AI Foundry, Azure OpenAI, Azure Machine Learning, Azure AI Search and Azure data platforms.
• Strong understanding of ML, generative AI, RAG, embeddings, vector search, agentic architecture, orchestration, tool use, memory and human-in-the-loop patterns.
• Proven experience defining enterprise target architectures, reference designs, platform guardrails, reusable patterns and technology roadmaps.
• Deep knowledge of MLOps/LLMOps, evaluation, model and prompt lifecycle, CI/CD, monitoring, drift, incident response and retirement.
• Strong knowledge of AI security, responsible AI, privacy, identity, Zero Trust, threat modelling, content safety, data protection and secure software development.
• Experience with enterprise integration patterns, APIs, event-driven architecture and data/knowledge architectures supporting AI solutions.
• Experience in performing architecture assurance, risk assessment, design governance and exception management across multiple delivery teams.
• Knowledge of and practical experience with Agent Security Posture Management, AI security testing and AI observability.
• Working experience with Microsoft Copilot Studio, Power Platform and Microsoft Fabric is preferred.
• Experience establishing an enterprise AI platform or leading AI architecture in a federated global organization is preferred.