AI Architect

WTW•King of Prussia, PA

About The Position

The AI Architect partners with the IT Director and internal client teams to translate business requirements into scalable, enterprise-grade AI architectures, primarily using Azure AI Foundry, Copilot Studio, and the Power Platform, with enterprise agent platforms where appropriate. The role owns end-to-end solution design across pro-code and low-code delivery, from discovery and MVP scoping through governance, integration, production hardening, and team enablement. It is not the primary developer role, but requires technical fluency to review code, challenge designs, set engineering standards, guide engineers, and use AI-assisted development to rapidly prototype and unblock delivery. The architect designs AI agents and agentic workflows, data-grounding and retrieval approaches, integrations, evaluation and observability, and security controls. The role also modernizes existing automation and AI assets, establishes reusable architecture and engineering practices, supports responsible AI governance, and guides model and cost decisions. The position works with internal stakeholders and delivery teams, including vendor and partner engineers, in a regulated-industry context.

Requirements

At least 7 years of solution or cloud architecture experience; Azure is strongly preferred, with equivalent GCP or AWS experience considered.
Hands-on experience with Azure AI Foundry and core Azure AI services.
Demonstrated experience designing cloud-native, enterprise-scale solutions.
Strong understanding of Well-Architected Framework principles, including reliability, security, cost optimization, and operational excellence.
Hands-on architecture experience with Microsoft Copilot Studio, including enterprise agents, topics, tools, and orchestration.
Working knowledge of Azure AI Foundry agent service, model catalogue, and evaluations, and of MCP as an integration pattern.
Solid understanding of RAG and knowledge-grounding design, including chunking, retrieval, Azure AI Search or vector stores, grounding evaluation, and choosing between RAG, fine-tuning, and tool use.
Experience taking AI or agent solutions from prototype to production, including evaluation, security, observability, and release gating.
Familiarity with enterprise agent platforms such as Moveworks or Lyzr and the ability to assess fit-for-purpose and build-versus-buy choices.
Working knowledge of Python or C# sufficient to read, review, and critique production code and guide pro-code engineers; primary development is not required.
Working knowledge of REST APIs, SDKs, and authentication and authorization patterns, including OAuth 2.0/OIDC, Entra ID app registrations, managed identity, and on-behalf-of flows.
Understanding of version control, code review, CI/CD, and infrastructure-as-code sufficient to set standards and hold delivery teams to them.
Architecture-level knowledge of Power Apps, Power Automate, Copilot Studio, and AI Builder.
Experience designing scalable, governed Power Platform solutions with sound ALM across environments.
Strong familiarity with M365, including SharePoint, Teams, and Dataverse, and with Entra ID and identity and security patterns as they relate to AI solution design and integration.
Ability to communicate complex technical concepts to non-technical senior stakeholders.
Collaborative approach and experience guiding and mentoring delivery teams, including vendor and partner engineers.
Employment-based non-immigrant visa sponsorship and/or assistance is not offered for this opportunity.

Nice To Haves

Microsoft certifications such as Azure Solutions Architect Expert (AZ-305), Azure AI Engineer Associate (AI-102), Power Platform Solution Architect Expert (PL-600), or Power Platform Developer Associate (PL-400).
Working knowledge of front-end frameworks and languages such as React, Angular, or TypeScript.
Hands-on CI/CD experience with Azure DevOps Pipelines or GitHub Actions and infrastructure-as-code experience with Bicep or Terraform.
Experience delivering AI solutions in a regulated industry, such as financial services or insurance, with data privacy and audit obligations.
Experience working alongside vendor or partner delivery teams in a multi-vendor model.

Responsibilities

Work with internal clients to understand use cases, identify core problems and success criteria, and scope MVPs; represent technical considerations in stakeholder discussions.
Own end-to-end AI solution architectures using Azure AI Foundry, Copilot Studio, the Power Platform, and appropriate enterprise agent platforms; produce HLDs, LLDs, and ADRs.
Design AI agents and agentic workflows, including MCP-based integration, RAG and knowledge grounding, tool orchestration, and agent evaluation approaches.
Design data-grounding architecture across Dataverse, SharePoint, Microsoft Fabric/OneLake, SQL, Azure AI Search, and vector stores; establish chunking, retrieval, and indexing strategies and choose between RAG, fine-tuning, and tool use.
Guide AI solutions from prototype to production, including evaluation frameworks, golden datasets, regression testing, release gating, tracing, security hardening, and red-teaming.
Assess existing automations and AI assets and define governed, maintainable modernization roadmaps aligned with enterprise standards.
Design integrations between AI capabilities and enterprise systems using connectors, MCP, and APIs, including M365, ServiceNow, ERP, HR platforms, and other enterprise sources.
Embed monitoring, tracing, and evaluation using platform-native tools, Azure Monitor/Application Insights, and Power Platform CoE capabilities.
Ensure solutions follow WTW Responsible AI intake and governance processes and address data classification, privacy, least-privilege access, auditability, and regulatory obligations with Security, Privacy, and Risk colleagues.
Review code and designs across pro-code and low-code workstreams; set authentication, API, CI/CD, and infrastructure-as-code standards; establish reusable components and reference architectures; and drive ALM and environment governance.
Guide model selection, token budgets, capacity decisions, and cost monitoring.
Track AI tooling developments and evaluate emerging platforms, frameworks, and patterns for delivery speed, capability, and cost efficiency.
Guide and mentor delivery teams, including vendor and partner engineers, and use AI-assisted rapid prototyping to validate concepts and unblock work.

Benefits

Competitive benefit package, subject to eligibility requirements.
Medical insurance, including prescription coverage.
Dental insurance.
Vision insurance.
Health Savings Account.
Commuter Account.
Health Care and Dependent Care Flexible Spending Accounts.
Group Accident and Group Critical Illness coverage.
Life Insurance and AD&D coverage.
Group Legal coverage.
Identity Theft Protection.
Wellbeing Program and Work/Life Resources, including Employee Assistance Program.
Paid holidays.
Annual paid time off, including paid state/local leave where required.
Short-Term Disability and Long-Term Disability.
Other leaves, including bereavement, FMLA, ADA, jury duty, military, parental, and adoption leave.
Contributory Pension Plan and Savings Plan (401k).
Some Level 38 and more senior roles may be eligible for non-qualified Deferred Compensation and Deferred Savings Plans.
Annual short-term incentive bonus eligibility.