A365 AI Governance Architect

SNI; Polska; Remote; PLN 180,000 - 260,000; Full time

Job summary

SNI sp. z o.o. is seeking an experienced A365 AI Governance Architect for a remote role with an international client.

You will design end-to-end governance for Agent365, establish identity and access controls, and build risk-based lifecycle processes across multiple platforms. The role requires deep knowledge of Microsoft security tools, cross-platform governance, and regulatory alignment with EU AI Act. You will drive TOMs, RACI matrices, and board-level governance activities in a PoC

Qualifications

• Proven ability to design governance architectures for AI platforms.
• Experience with risk-based lifecycle governance and audit-ready controls.
• Strong documentation skills for playbooks and governance charters.

Responsibilities

• Design end-to-end Agent365 platform architecture for agent lifecycle.
• Establish identity models and delegated access controls.
• Build classification models for autonomy, data sensitivity and tool access.
• Operationalize a multi-stage risk-based lifecycle with owners and evidence rules.
• Maintain cross-platform discovery and governance across Copilot Studio and Azure Foundry.
• Formulate TOMs, RACI matrices, decision rights and governance charters.
• Align agent controls with Entra, Purview and Defender with security teams.
• Incorporate EU AI Act classification and works-council constraints into lifecycle gates.
• Validate governance and lifecycle controls on live agents in PoC environments.
• Author governance playbooks and standing work instructions for registration, reviews, recertification, retirement.
• Co-run governance board meetings with client leadership and manage handovers.

Skills

AI governance

Architectural design

Microsoft Entra

Governance frameworks

Tools

Microsoft Agent365

Copilot Studio

Azure AI Foundry

Entra

Purview

Defender

Job description

SNI serves as a trusted IT outsourcing partner for some of the world's most prestigious companies, successfully delivering projects worldwide.

For one of our international clients, we are looking for A365 AI Governance Architect.

Job details:

• Work Model: Remote

Responsibilities:

• Design end-to-end Agent365 platform architecture for agent registration, deployment, inventory, containment, and residual-free retirement.


• Establish agent identity models, scoped consent, and delegated access using Microsoft Entra.


• Build an archetype classification model evaluating autonomy, data sensitivity, tool access, and business criticality with an associated control matrix.


• Operationalize a 10-stage lifecycle with risk-based gates, assigned owners, and evidence consumption rules.


• Maintain an authoritative agent register and cross-platform discovery model across Copilot Studio, Azure AI Foundry, and non-Microsoft frameworks.


• Formulate Target Operating Models, RACI matrices, decision rights, and board charters across governance domains.


• Align agent controls with Microsoft Entra, Purview, and Defender alongside enterprise security teams.


• Integrate regulatory review checkpoints into lifecycle gates, covering EU AI Act classification and works-council constraints.


• Validate governance, identity, and lifecycle controls on live agents within a PoC environment.


• Author operational governance playbooks and standing work instructions for registration, gate reviews, recertification, and retirement.


• Co-run governance board meetings with client leadership and execute a structured operational handover.

Skills:

• Deep architectural expertise in Microsoft Agent365, Copilot Studio, and Azure AI Foundry.


• Strong background in Microsoft security technologies, including Entra, Purview, and Defender.


• Working knowledge of non-Microsoft agent platforms to enable cross-platform discovery and governance.


• Proven capability to design Target Operating Models, RACI frameworks, lifecycle gates, and audit-ready control matrices.


• Deep understanding of identity governance, scoped OAuth delegation models, and residual-free object decommissioning.


• Experience conducting risk assessments on novel agent behaviors, including delegation depth, tool reach, and multi-agent interactions.


• Familiarity with EU AI Act classification frameworks, Responsible AI guidelines, and works-council compliance requirements.


• Exceptional technical documentation skills for producing operational playbooks, work instructions, and governance charters.

SNI sp. z o.o. will process personal data for the purpose of the recruitment process in accordance with Data Privacy Policy. The data may also be stored and processed for future recruitment purposes, in accordance with the given consent.