Microsoft Foundry Architect

New Value Solutions, a national IT consulting company, is seeking a Microsoft Foundry Architect to lead the architecture, design, and hands-on implementation of enterprise AI solutions on Microsoft Foundry, formerly Azure AI Foundry.
The successful candidate will establish the platform foundation and develop secure, scalable, and reusable patterns for AI agents, retrieval-augmented generation (RAG) applications, model deployments, and AI-enabled business workflows. This role combines architecture and engineering delivery, taking solutions from initial design through production deployment while addressing security, privacy, governance, performance, and cost requirements.
Responsibilities
Design and document Microsoft Foundry architecture, including platform resources, projects, development and production environments, and Azure landing zone integration.
Define reference architectures, engineering standards, and reusable patterns for AI agents, copilots, RAG applications, and AI-enabled workflows.
Build and deploy agents using Foundry Agent Service, including tool integration and multi-agent orchestration.
Select, deploy, and manage AI models, including model versioning, deployment options, quotas, and capacity planning.
Design and optimize RAG pipelines, including chunking, embeddings, indexing, retrieval, and re-ranking using Azure AI Search and vector storage.
Implement enterprise security controls, including private networking, managed identities, Microsoft Entra ID, RBAC, Azure Key Vault, and audit logging.
Incorporate Responsible AI, content safety, privacy, and human review controls into AI solutions.
Develop AI evaluation processes, automated regression tests, red-teaming practices, and continuous production evaluations.
Implement monitoring, tracing, and observability using OpenTelemetry, Azure Monitor, and Application Insights.
Monitor and optimize token consumption, latency, model capacity, and platform costs.
Automate infrastructure provisioning and deployments using Terraform or Bicep and CI/CD pipelines.
Integrate AI solutions with enterprise applications, APIs, Microsoft 365 Copilot, Copilot Studio, and Microsoft Teams.
Review designs and implementations against architecture standards, production-readiness criteria, and enterprise change controls.
Produce architecture decision records, runbooks, operating procedures, and reusable templates.
Provide knowledge transfer and mentoring to engineering and platform teams.
Requirements
8+ years of experience in cloud architecture, software engineering, or platform engineering, including 2+ years designing and delivering AI/LLM solutions.
Hands-on experience architecting and implementing solutions using Microsoft Foundry, Azure AI Foundry, Azure AI Studio, or Azure OpenAI Service, including at least one production deployment.
Experience designing LLM, RAG, and agentic solutions while balancing quality, latency, cost, security, and privacy.
Hands-on experience with Foundry Agent Service and agent frameworks such as Microsoft Agent Framework, Semantic Kernel, AutoGen, or LangGraph.
Strong knowledge of Azure AI Search, embeddings, and vector-based retrieval.
Strong Azure architecture experience across networking, private endpoints, identity, RBAC, Azure Key Vault, and Azure Monitor.
Proficiency in Python and/or C#/.NET, including experience with Foundry/Azure AI SDKs and REST APIs.
Experience with Infrastructure as Code using Terraform or Bicep, CI/CD pipelines, Git, and Agile delivery.
Ability to communicate architecture decisions clearly to technical and non-technical audiences.
Strong collaboration and communication skills within cross-functional teams.
Undergraduate degree in Computer Science, Engineering, or a related STEM discipline;
an equivalent combination of education and experience may be considered.
Preferred qualifications include relevant Microsoft certifications, MCP and agent-to-agent integration, containerized agent deployments, model fine-tuning, AI evaluation and content safety tooling, and experience in regulated industries.
ID# 5606
#J-18808-Ljbffr