Sr. AI Governance Architect

Cyclotron, Inc.•Chicago, IL

1d•$160,000 - $180,000•Remote

About The Position

Cyclotron's AI Transformation Group is hiring a Sr. AI Governance Consultant to drive how our enterprise clients govern AI and agents across the entire Microsoft estate. This is a senior, client-facing role that is both broad and deep. You will own AI and agent governance and data governance as your primary disciplines: Copilot Studio and Agent Builder lifecycle, Agent 365, Microsoft Purview, data readiness for AI, and responsible AI controls. You will also be expected to carry real depth in Power Platform governance and in platform and workload governance across Microsoft 365 and Azure. The right person has designed a governance operating model, sold it to an AI Council or Architecture Review Board, and then configured it in the tenant. They stood in front of an estate of agents and environments and knew where to start. You will lead engagements, shape scope with our sellers, and raise the bar for the consultants around you. We treat governance not as a brake but as the thing that lets AI scale safely.

Requirements

8+ years of enterprise technology experience, including 4+ years leading governance, security, or architecture work on Microsoft platforms in multi-business-unit, regulated, or high-scrutiny environments.
Proven depth in AI and agent governance: Copilot Studio, Agent Builder, Microsoft 365 Copilot, and Agent 365, including publishing models, agent identity, security posture, and monitoring at enterprise scale.
Proven depth in data governance with Microsoft Purview (information protection, DLP, data map and catalog, DSPM for AI) and working knowledge of Microsoft Fabric.
Strong hands-on Power Platform governance: environment strategy, DLP design and enforcement, managed environments, CoE Starter Kit, and ALM with Pipelines or Azure DevOps. You can navigate PPAC quickly and know what each setting actually does.
Solid grounding in identity and security: Entra ID, least privilege, conditional access, RBAC, Defender, and auditability, and how they map onto AI, data, and Power Platform decisions.
Working knowledge of Azure governance (landing zones, Azure Policy, RBAC, cost controls) and Microsoft 365 collaboration governance.
A track record of both architecting and implementing: you designed the model, secured the decision, and configured it. Portfolio evidence expected.
Executive-grade communication and emotional intelligence: you can explain governance trade-offs and risk to a CIO, CISO, or AI Council, in writing and in the room.
Able to work remotely within the United States or Canada with reliable overlap for Eastern and Central business hours.
Governance-fluent bar: we are looking for someone visibly engaged with how Microsoft AI, data, and automation platforms actually get governed at enterprise scale. You don't need to check every box below, but the majority should feel like a clear “yes,” and you should be ready to walk us through the relevant ones in a portfolio review:
You have governed an agent estate at scale, including inventory, risk triage, and bringing unmanaged agents and environments into a governed model.
You have designed Purview controls specifically to make Copilot and agents safe to scale, and can explain how you handled oversharing and knowledge source risk.
You have stood up an AI CoE or AI Council operating cadence at a real client and can describe the intake, escalation, and political work it took to stick.
You have informed opinions on Agent 365, Copilot Studio security, and where Microsoft's native guardrails still fall short.
You have translated a responsible AI framework into concrete controls, evaluation gates, and audit evidence.
You stay current: you read the release plans, try features in preview, and have a view on where Copilot, agents, and Purview are heading next.

Nice To Haves

Azure AI depth: Azure OpenAI, AI Foundry, AI Search, content safety, and evaluation tooling, and how they integrate with Copilot Studio.
Pro-code extensions: Azure Functions, Logic Apps, custom connectors, API Management, and GitHub Advanced Security.
ServiceNow or other ITSM integration for agent intake and approvals.
Adoption and change management: AI Ambassador or champion programs, Viva Insights, and the Copilot Dashboard.
Certifications such as SC-100, SC-401, PL-600, AB-100, or AZ-305, or equivalent demonstrated experience.
Consulting or professional services experience, pre-sales exposure, and a track record of mentoring consultants or growing a practice.

Responsibilities

Lead the design and implementation of enterprise AI and agent governance frameworks: target operating model, governance decisions, policies and standards, roles and responsibilities, and a prioritized, adoptable roadmap.
Govern the full agent lifecycle across Agent Builder, SharePoint agents, Copilot Studio, Frontier agents, and Copilot connectors: intake and registration, routing criteria (Agent Builder vs. Copilot Studio), maker access and publishing models, production promotion checkpoints, ownership change, monitoring, and retirement.
Design Agent 365 governance: agent identity boundaries, policy, monitoring, audit, and alignment with Microsoft Defender, Purview, and Entra ID.
Define Copilot Studio-specific security and DLP controls (event triggers, channels, knowledge sources, HTTP requests, Graph and federated connectors), along with credit consumption and capacity guardrails.
Stand up and formalize AI Centers of Excellence: agent lifecycle and connectors, monitoring and reporting, audit and compliance roles, review cadences (for example, Agent Insights and monthly usage reviews), and escalation into existing ARB and AI Council structures.
Map responsible AI and regulatory expectations (NIST AI RMF, ISO/IEC 42001, EU AI Act, industry-specific requirements) into practical controls, evaluation gates, and audit evidence.
Lead data governance workstreams in Microsoft Purview and Fabric: catalog, lineage, classification, data quality, sensitivity labeling, and data product ownership.
Assess data readiness for AI: data estate assessments, oversharing risk in SharePoint and OneDrive, grounding and knowledge source governance for agents, and the data controls needed before Copilot and agents go broad.
Design Purview protections for AI (DSPM for AI, DLP, information protection, retention, audit) and coordinate them with Power Platform and Copilot Studio policy.
Govern Dataverse and Dataverse-to-Fabric patterns so agent and app data stays classified, owned, and auditable.
Design environment strategy and landing zones: tiered and functional environments, environment groups, Development to Stage/UAT to Production models, naming, ownership, and provisioning standards.
Design classic and advanced DLP and connector classification, run impact analysis against existing resources, and remediate without breaking the business.
Own ALM and deployment governance: solution packaging, versioning, environment variables, Power Platform Pipelines and Azure DevOps approval gates, and rollback.
Deploy and operationalize the CoE Starter Kit and Copilot Studio Starter Kit, managed environments, and security group and role models in Entra ID.
Shape Microsoft 365 collaboration governance: Teams, SharePoint, and Exchange policy, external sharing, information barriers, lifecycle, and retention.
Advise on Azure governance for AI workloads: landing zones, Azure Policy, RBAC, tagging, cost controls, and governance of Azure OpenAI and AI Foundry deployments.
Define application and SDLC governance for AI solutions: API Management, Logic Apps, secrets handling, code review and model release gates, prompt management, and evaluation gates in Azure DevOps and GitHub.
Serve as the governance lead on multi-pillar engagements; direct mid-level consultants and engineers and own the quality of what ships.
Lead executive briefings, AI Opportunity Labs, environment strategy workshops, and security and DLP strategy workshops; run the room without dominating it.
Partner with sellers and solution engineering on discovery, estimates, WBS, and SOWs, making in-scope and out-of-scope governance offerings explicit.
Produce client deliverables that get used: governance frameworks, playbooks, runbooks, decision logs, intake forms, and policy templates.
Contribute to Cyclotron's Governance Offering Suite: reusable IP, accelerators, and the quality bar that separates foundational from strategic work.