About the Role:
Client is seeking a senior AI Governance Architect to translate a client s AI governance program into the technical structure required to support and enforce it. This is the bridge role between governance and technology. It takes the decision rights, oversight requirements, and policy defined by the governance function and turns them into a target-state architecture the organization can operate. This role does not build AI models, agents, or retrieval systems. Instead, it designs the catalog, lineage, classification, inventory, logging, and integration patterns that make AI activity observable, auditable, and governable through the same systems already used for data. The work is predominantly design and direction, with hands-on involvement limited to validating that a given integration pattern works before it is scaled.
Key Responsibilities:
- Governance-to-Architecture Translation: Design the target-state technical architecture that supports and enforces AI Governance (decision rights, oversight requirements, and policy)
- Integration Design: Own the integration design between the governance, catalog, and lineage stack and the enterprise AI/ML environment, so AI activity is observable and auditable through the systems Data Governance already uses.
- Oversight and Traceability Requirements: Specify the requirements and the architecture required to define how model and agent inputs, outputs, prompts, metadata, and lineage must be captured and traced.
- Catalog and Lineage Extension: Define how the enterprise catalog, lineage, classification, and metadata stack extends to AI artifacts such as model cards, training datasets, prompt templates, and agent registrations.
- Governed Data Foundation: Define the governance and control requirements for data flowing through AI systems, including sensitive-data handling; ensuring data is governed and traceable as it feeds AI, rather than engineering retrieval performance.
- Reference Patterns and Validation: Produce decision records and reference architecture that the build team implements; remain hands-on only to the extent of proving that a given integration pattern works before it is scaled.
- Accountability for Enforcement: Own the outcome that governance is enforceable in the technical environment, with the build team accountable to this role for faithful implementation.
Required Qualifications:
- 8 to 12 years of progressive experience as a data architect, governance architect, or solution architect with significant scope spanning both data and AI/ML environments.
- Demonstrated experience designing how governance, catalog, and lineage capabilities integrate with an enterprise AI/ML environment.
- Strong foundation in data architecture, metadata management, lineage, and classification on a major cloud platform.
- Working understanding of at least one major cloud AI platform (Google Vertex AI / Gemini, Microsoft Azure AI Foundry, or AWS Bedrock) sufficient to design integration and oversight not to build models.
- Hands-on familiarity with enterprise catalog and metadata platforms (e.g., Collibra, Alation, Atlan, Microsoft Purview, Informatica).
- Track record translating governance policy and oversight requirements into operational technical architecture, not solely advisory or reference-architecture work.
- Bachelor s degree in Computer Science, Information Systems, Engineering, or a related field, or equivalent professional experience.
Preferred Qualifications:
- Experience in regulated industries such as agriculture, life sciences, pharmaceuticals, financial services, or healthcare.
- Familiarity with agentic AI and retrieval patterns from a governance, observability, and control standpoint.
- Familiarity with MLOps and model-governance tooling and how it integrates with enterprise metadata and lineage.
- Relevant certifications such as CDMP, Google Cloud Professional Data Engineer, or a major cloud architecture certification.
- Consulting experience with large, complex, multi-platform clients.
What Success Looks Like:
- A documented target-state architecture that connects the AI governance program to the technology estate, adopted by the client s engineering and governance teams.
- Governance that is enforceable in the environment, with catalog and lineage extended to AI artifacts so model and agent activity is visible through the systems already used for data.
- A clear integration design and set of reference patterns that the build team can implement and scale without further architectural guesswork.
- Oversight and traceability requirements that the client s team can run independently after the engagement closes.