Senior Ai Data Integration Engineer

NorthHill Technology Resources has an urgent need for a Senior Data Integration Engineer for a cutting-edge opportunity in Mclean, VA.  This is a direct-hire role with our client, a highly respected banking organization,  It is a hybrid role, with 3 days onsite and 2 remote per week.
  Own the bank's approved knowledge and systems layer for AI by implementing secure grounding on enterprise content, approved connectors, permissions-aware retrieval, and DLP-aligned data handling. This role includes the immediate masking and unmasking path for SharePoint-accessed content using Presidio and Protect.

• Build and maintain secure grounding patterns for approved content repositories and internal business systems, starting with SharePoint and other governed enterprise sources.
• Implement approved connectors and retrieval workflows with permissions-aware access, source traceability, and controls that scope retrieval to the minimum necessary data.
• Implement and support Presidio and Protect for immediate masking and unmasking of SharePoint-accessed content before it is released to approved AI workflows.
• Partner with Security, IT, Information Governance, and business owners on DLP, sensitivity labels, metadata, redaction, and data-handling controls for AI retrieval paths.
• Improve retrieval quality, metadata standards, and integration reliability across structured and unstructured sources so answers remain grounded and supportable.
• Create reusable onboarding standards for new content sources, including access review, logging expectations, retention considerations, and validation before activation.
• Engineer retrieval services that preserve permissions inheritance and return provenance metadata sufficient to support reviewer verification, citations, and audit traceability.
• Build validation checklists for new repositories, connectors, content types, and SharePoint masking flows before they are exposed to employee-facing copilots or higher-risk governed workflows.
• Partner with Guardrails and Application teams to tune retrieval quality, reduce hallucination risk, and enforce least-privilege data access across approved workflows.
• Support secure integration patterns for approved enterprise AI platforms, including Claude Cowork and related retrieval-dependent tools, where grounding, connector behavior, provenance controls, and masking services must be enforced.

• Implement connector and retrieval logging that supports audit trails for what data sources were accessed, by whom, and under what approved workflow where the platform supports it.
• Design integrations to respect data minimization, permissions inheritance, read-only access where required, and restrictions on shared file write or delete behavior.
• Help operationalize immediate compensating controls for PII or NPI workflows, including Presidio and Protect for SharePoint-accessed content, with documented fallback redaction controls and QA where needed.
• Coordinate evidence and metadata standards so approval artifacts, retrieval traces, and content-source onboarding records can be retained in the governance repository.
• Contribute technical review for plugins, custom MCP servers, and other integrations that expose enterprise systems to AI workflows.

• 6+ years in data engineering, integrations, enterprise search, retrieval engineering, knowledge systems, or content-platform engineering.
• Hands-on experience with APIs, enterprise content platforms, permissions models, identity-aware retrieval, and reliable integration patterns across heterogeneous data sources.
• Strong understanding of data classification, DLP concepts, metadata, lifecycle and retention considerations, and enterprise content governance.
• Ability to troubleshoot retrieval quality, indexing, connector reliability, and source traceability in document-heavy environments.
• Experience documenting standards so new data sources can be onboarded repeatedly without creating inconsistent control behavior.

• Experience with SharePoint, Microsoft Graph, Copilot Studio grounding patterns, semantic search, vector or hybrid retrieval, and enterprise content systems used in regulated environments.
• Exposure to legal, trust, compliance, HR, or document-heavy operational processes where permissions and provenance matter.
• Financial services or other regulated-data experience with practical awareness of privacy, records, and audit obligations.
• Hands-on experience with Presidio, Protect, or comparable masking, redaction, and data-protection tooling that can support controlled AI workflows.
• Experience supporting Claude Cowork or similar enterprise AI platforms where retrieval, knowledge access, and provenance controls matter.
• Familiarity with Claude Code or comparable AI-assisted engineering tools for connector development, debugging, and integration acceleration.