3.2 Fabric Platform Operations & Administration

• Operate and support Fabric capacities, tenant settings, domains, workspaces, gateways and connectivity, OneLake, Lakehouses, Warehouses, SQL endpoints, Data Factory items, notebooks, semantic models and related Power BI services.
• Monitor capacity utilization, throttling, job concurrency, pipeline and notebook execution, refreshes, query performance, failures, costs and service health; tune or scale workloads within approved limits.
• Investigate incidents and recurring failures across ingestion, transformation, storage, semantic models, gateways, access and reports; document root cause, corrective action and prevention when required.
• Maintain schedules, service identities, connections, credentials, operational checks, recovery procedures and support runbooks for production services.
• Provide level 2/3 support and coordinate with Microsoft, UNICC, vendors and internal teams when an issue requires escalation.
• Operate and administer the Microsoft Fabric estate on a daily basis, including tenant settings, domains, workspace structures, item lifecycle, licensing assignment, and platform configuration in line with PAHO governance requirements.
• Manage capacity provisioning, utilization monitoring, throttling and smoothing analysis, workload isolation, and cost optimization on F-SKU capacities using the Fabric Capacity Metrics App; maintain a consumption forecast against contracted entitlement.
• Monitor and support pipelines, Dataflows Gen2, notebooks, warehouse and lakehouse workloads, and semantic model refreshes; resolve failures, latency, and performance degradation; identify root causes and sustainable corrective actions.
• Maintain and improve deployment pipelines, workspace Git integration, and environment promotion across development, test, and production.
• Own backup, resiliency, and recovery for OneLake data and workspace artifacts, including documented and periodically tested recovery procedures for production workloads.
• Handle incidents, service requests, problems, and changes through PAHO’s corporate IT service management platform, respecting agreed service levels and change control requirements.

3.3 Data Engineering, Integration & Semantic Modeling

• Design, build, and maintain ingestion and transformation processes using Fabric Data Factory pipelines, Dataflows Gen2, Spark notebooks, shortcuts, and mirroring, applying incremental and cost-efficient patterns.
• Implement and maintain a medallion (bronze / silver / gold) structure with explicit ownership, contracts, and quality controls between layers.
• Develop, maintain, and troubleshoot integrations with PAHO corporate systems — in particular the Workday-based PMIS covering human resources, finance, procurement, and grants data — as well as Salesforce, SharePoint, SQL sources, and external data providers, using approved APIs, report-as-a-service extracts, event-driven patterns, and secure authentication mechanisms.
• Design and maintain dimensional and semantic models (star schema, slowly changing dimensions, Direct Lake), DAX measures, certified datasets, and row-level and object-level security aligned to PAHO’s data access rules.
• Implement data quality, reconciliation, validation, and lineage controls; investigate and resolve data discrepancies with the relevant data owners and document the outcome.
• Provide second-level technical support and guidance to PAHO report developers and business power users on modeling, performance, and reporting standards.

3.4 Platform Baseline, Standards & Evolution Roadmap

• Establish and thereafter maintain a complete and current baseline of the Fabric environment, covering architecture, domain and workspace structure, capacity allocation, data sources and flows, semantic models, security model, integrations, dependencies, operational procedures, and ownership, so that the platform can be operated and handed over by any qualified engineer without reliance on undocumented knowledge.
• Review the current configuration of capacity, workspaces, pipelines, semantic models, and security against Microsoft reference architecture, well-architected guidance, and PAHO standards; document deviations, risks, technical debt, and remediation options with associated effort and benefit estimates.
• Identify, quantify, and — once approved — implement optimization opportunities across capacity consumption and cost, refresh and query performance, pipeline design, storage, and licensing.
• Define, document, and apply platform engineering standards covering naming conventions, workspace and domain strategy, environment separation, source control and CI/CD, promotion and change gates, monitoring and alerting, data quality, and documentation.
• Produce and maintain a prioritized twelve-to-eighteen-month evolution roadmap for the data and analytics platform, covering architecture, governance, capacity, integration, semantic layer, and AI enablement, sequenced with effort, dependencies, risks, cost implications, and resourcing requirements, for ITS management decision.
• Provide ITS management with objective, evidence-based analysis of platform health, risk exposure, and improvement priorities on a recurring basis.

3.5 AI-Enabled Analytics & Innovation

• Assess, pilot, and where approved operationalize Fabric-native AI capabilities, including Copilot in Fabric and Power BI, data agents, AI functions in notebooks, and semantic model preparation for natural-language querying.
• Build AI-ready data products and semantic layers that can be securely grounded and consumed by PAHO’s Copilot Studio and Power Platform agent portfolio, Azure AI services, and other approved AI solutions.
• Propose and prototype practical, measurable use cases — such as automated data quality and anomaly detection, forecasting, natural-language reporting, and document or text analytics — presenting expected benefit, consumption impact, and cost before any scaling decision.
• Apply PAHO’s AI governance, data protection, information classification, and responsible AI requirements; document grounding sources, guardrails, evaluation scenarios, known limitations, and human oversight arrangements.
• Monitor the Microsoft Fabric and Azure AI roadmap and advise ITS on which capabilities are sufficiently mature for institutional adoption and which should remain experimental, with a clear rationale in each case.

3.6 Security, Governance & Compliance

• Apply least-privilege access controls across tenant, domain, workspace, item, and data levels, including workspace roles, sharing settings, and conditional access aligned with PAHO requirements.
• Manage Microsoft Entra ID service principals and app registrations used for Fabric and Power BI automation, including permission scoping, API access configuration, and credential and secret rotation.
• Administer and validate row-level and object-level security in semantic models, including membership-based security patterns and deployment-time integrity checks.
• Support Microsoft Purview integration, sensitivity labels