Azure Data Platform

Fragmented sources and brittle pipelines make dashboard figures difficult to explain. Documenting dependencies, owners and acceptance conditions helps an organisation decide whether to repair an existing component, replace it or leave it in place.

Architecture & Platform Design

• Review data lake and lakehouse architectures, including Delta Lake, medallion patterns and Data Vault models
• Compare Azure Synapse Analytics SQL pools and serverless SQL against workload and cost requirements
• Define OneLake storage, ownership and access for Microsoft Fabric or Power BI environments
• Document subscription, tenancy and environment boundaries for hybrid access and isolation

Data Ingestion, Integration & Pipeline Setup

• Trace event, streaming and batch pipelines across Azure Data Factory, Synapse Pipelines and Azure Functions
• Compare integration patterns for on-premises, hybrid and SaaS sources
• Configure versioned pipeline deployment and infrastructure-as-code checks for agreed environments
• Record metadata, lineage, data-product owners and support responsibilities

Governance & Operational Foundations

• Define role-based access, configure agreed Microsoft Purview controls and test access conditions
• Set cost thresholds, owners and alert routes for compute and storage
• Configure health checks for compute, storage and transformation pipelines, with documented response owners

• Configure Azure Data Factory, Synapse Pipelines or Azure Functions to orchestrate specified ingestion and API tasks
• Configure Databricks Unity Catalog, Delta Live Tables and Workflows for named lakehouse pipelines and access rules
• Define Microsoft Fabric OneLake shortcuts, item owners and discovery metadata for agreed data products
• Create or amend Terraform, ARM and Azure DevOps templates for reviewed platform components
• Configure Azure Monitor and Microsoft Purview views for agreed lineage, quality and cost measures

1. Review the estate: Inventory workloads and consumers, trace priority pipelines and record security, regulatory and cost constraints.
2. Compare and configure: Compare architecture options, document the selected design and configure the components and controls in scope.
3. Test and hand over: Test agreed acceptance conditions, record owners and operating procedures, and walk the internal team through the handover.

Rory leads the architecture review, records recommendations and remains accountable for the engagement. Existing pipelines and legacy processes are reviewed before replacement is proposed. If a defined implementation task needs an associate, that role, scope and acceptance condition are agreed in advance. The handover identifies what changed, who owns it and how it is tested.