Portfolio Projects
Demonstrating practical application of skills to solve real-world problems.
Enterprise Data Warehouse for a Healthcare Leader
Problem: A leading consumer healthcare company faced data fragmentation and inefficient reporting, leading to slow decision-making and high operational costs.
Actions & Role: I led the end-to-end design and implementation of a centralized enterprise data lake and data warehouse on Azure. My team and I architected a robust, scalable platform to ingest, store, and structure data from numerous disparate sources, creating a single source of truth.
Outcome: The unified platform eliminated data silos, improved data consistency, and reduced manual reporting cycles by 75%. This enhanced data accessibility for all departments and enabled advanced analytics, leading to faster, data-driven decisions and an estimated 30% reduction in operational costs.
Technologies: Azure Databricks Azure Data Lake Gen2 Azure Synapse Analytics Microsoft Purview GitHub Azure DevOps Python Spark
Unified Cybersecurity Asset Intelligence Platform
Problem: A major pharmaceutical company needed a single source of truth for its cybersecurity assets to enable consistent reporting, improve risk assessment, and strengthen its overall security posture.
Actions & Role: I designed a Cybersecurity Asset Intelligence Platform to consolidate data from diverse security sources into the Azure data ecosystem. The solution included developing a comprehensive control posture reporting system with drill-down functionality and heat map visualizations.
Outcome: The platform provided a unified, high-integrity view of the asset inventory and control postures. This directly improved risk assessment capabilities by an estimated 40% and established a modular foundation for efficiently adding new data sources.
Technologies: Azure Databricks Python Power BI Data Integration Data Lakehouse Unity Catalog Data Governance GitHub GitHub Actions
Strategic Cloud Migration & Infrastructure Intelligence
Problem: A global pharmaceutical giant's cloud migration was stalled by complex, hidden dependencies in its IT environment. Decommissioning servers based on utilization alone was too risky, potentially causing outages for critical, undocumented services.
Actions & Role: I architected a multi-layered intelligence engine that ingested data from CMDBs and performance monitors to programmatically map the entire service hierarchy (from business apps down to physical servers). This system used a logical, multi-step process to safely identify decommissioning candidates by verifying application status before analyzing cluster utilization.
Outcome: This intelligence-driven approach provided a zero-impact guarantee, enabling the confident decommissioning of underutilized hardware. It unlocked significant cost savings, mitigated business continuity risks, and improved the overall integrity of the enterprise CMDB.
Technologies: Python SQL PySpark Azure Databricks Azure Data Lake Gen2 Configuration Management Database (CMDB) GitHub GitHub Actions
AI-Powered Strategic Business Case Generator
Problem: The manual process for creating strategic business cases was slow and often resulted in proposals that were not meticulously aligned with the organization's key objectives, hindering their funding appeal.
Actions & Role: I developed an advanced AI tool that leverages project intake information to automate and optimize the creation of business cases. The tool ensures every case is strategically sound, clearly articulates its value, and aligns with funding criteria.
Outcome: The generator significantly streamlined the path to funding by reducing manual effort in business case creation by an estimated 75%. This accelerated approval cycles and increased the success rate for securing resources for critical projects.
Technologies: Azure OpenAI Azure Databricks Azure SQL Server Natural Language Processing (NLP) Python Streamlit Flask