AI Solutions Architect

Description You will partner with IT Director and internal client teams to translate complex business requirements into scalable, production-grade architectures - spanning pro-code Azure solutions, low-code Power Platform experiences, and emerging agentic AI frameworks. This role is the keystone that unblocks a high-performing development team by owning end-to-end solution design: from initial client discovery and MVP scoping through to architecture governance, observability strategy, and developer guidance. You will modernize existing AI workloads - including LangChain/LangGraph pipelines and RAG systems - while establishing a forward-looking architecture practice built on the latest AI, integration, and cloud-native patterns. The Role Client engagement & discovery — Work directly with internal clients to deeply understand their use cases, identify the core problem and success criteria, and translate requirements into a clearly scoped MVP. Act as the technical voice in stakeholder conversations, bridging business need to technical possibility. Solution architecture ownership — Design and own end-to-end architectures for AI solutions across the full delivery spectrum: pro-code applications on Azure, low-code solutions on Power Platform & Copilot Studio, and third-party platforms such as Lyzr or Moveworks. Produce architecture artefacts (HLD, LLD, ADRs) that guide delivery teams. AI & agentic framework design — Lead the architecture of advanced AI capabilities: multi-agent systems, agentic workflows, advanced RAG (contextual retrieval, hybrid search, re-ranking), MCP integration, and next-generation AI orchestration patterns using Azure AI Foundry, LangGraph, and adjacent frameworks. Modernization of existing AI workloads — Assess and evolve current LangChain/LangGraph and OpenAI-based pipelines and Google Cloud AI assets. Define a roadmap to advance these toward production-grade, observable, and maintainable architectures aligned with enterprise standards. Backend & integration architecture — Design scalable APIs, event-driven integrations, and enterprise connectors that underpin AI solutions. Ensure AI capabilities integrate cleanly with enterprise systems (M365, ServiceNow, ERP, HR platforms, etc.). Observability & operational excellence — Embed observability-first thinking into every architecture: define logging, tracing, evaluation, and monitoring frameworks for AI systems using tools such as Azure Monitor, Promptflow evals, LangSmith, or equivalent. Ensure AI solutions are auditable and trustworthy at scale. Developer enablement & technical governance — Work hands-on with the engineering team as a trusted design partner. Conduct architecture reviews, provide hands-on guidance during delivery, establish reusable patterns and reference architectures, and reduce technical debt through principled design decisions. Technology radar & innovation — Maintain an active awareness of the AI tooling landscape. Evaluate and recommend emerging platforms, frameworks, and patterns that could improve delivery speed, capability, or cost-efficiency for the team. Qualifications The Requirements Cloud & Infrastructure architecture 7+ years of solution or cloud architecture experience; strong preference for Azure (AKS, Azure OpenAI Service, Azure AI Foundry, Azure Functions, API Management, Service Bus, Azure AI Search, Cosmos DB, Azure Data Factory). Equivalent GCP or AWS considered. Demonstrated experience designing cloud-native, enterprise-scale applications. Familiarity with Well-Architected Framework principles (reliability, security, cost optimization, operational excellence). AI, ML & agentic systems Proven hands-on experience designing and deploying production RAG systems. Deep knowledge of advanced RAG patterns: hybrid search, re-ranking, contextual chunking, graphRAG, and long-context strategies. Understanding of MCP (Model Context Protocol) as an emerging integration pattern. Hands-on experience with LangChain, LangGraph, and agentic orchestration patterns (ReAct, Plan-and-Execute, multi-agent supervisor patterns). Experience working with LLM providers: Azure OpenAI, Google Gemini, and open-weight models via Azure AI Foundry model catalogue. Familiarity with AI safety, responsible AI principles, and enterprise guardrail patterns (content filtering, grounding checks). Experience designing AI evaluation frameworks (ragas, offline evals, online monitoring, LLM-as-judge). Nice to have: Experience designing or advising on predictive ML models (classification, forecasting) — not necessarily model training, but understanding the architecture around data pipelines, feature stores, and model serving in an enterprise context. Low-code, automation & platform tools Architecture-level knowledge of the Microsoft Power Platform: Power Apps, Power Automate, Copilot Studio (formerly PVA), and AI Builder. Exposure to enterprise third-party AI platforms such as Lyzr (agent builder), Moveworks (enterprise AI assistant), or comparable (ServiceNow AI, Glean, Workato). Ability to assess fit-for-purpose versus build vs. buy for AI automation scenarios. Backend, integration & software architecture Hands-on experience designing backend services in Python and/or Node.js/.NET. Familiarity with enterprise application integration: M365 ecosystem (SharePoint, Teams), identity & security (Entra ID, OAuth 2.0, managed identity), and data platforms (Azure Data Lake, Fabric, Purview) as AI data sources. Communication & ways of working Exceptional ability to communicate complex technical concepts to non-technical senior stakeholders. Collaborative mindset with experience guiding and mentoring engineering teams WTW is an Equal Opportunity Employer #J-18808-Ljbffr