SQL analysis identifying $1.2M in recoverable IT spend — ghost licenses, cloud anomalies, and AWS vs Azure overspend for a simulated PE firm.

Operational supply chain analytics project using SQL and Python to analyze demand trends, procurement activity, delivery performance, inventory classification, and operational KPIs.

Python engine that analyzes 25 years of WTI oil and Henry Hub gas prices to output BUY/WAIT/WATCH procurement recommendations using FRED API data.

End-to-End Python implementation of Christodoulides's (2026) interpretable forensic decision-support engine for institutional procurement integrity. Implements: graph-based supplier entity resolution, robust IQR standardization, Gaussian Mixture Model regime extraction, and a decomposable PHI composite score for prioritized audit triage.

Data-driven analytics and ML clustering of Uzbekistan's public procurement registry (STIR: 201122919). Features spend optimization, vendor risk intelligence, and interactive visualizations.

Data engineering system for explainable prioritization of Colombian public-procurement review using SECOP open data

End-to-end procurement analysis covering €60K supplier spend, OTIF performance tracking, weighted bid evaluation and material cost forecasting — identifying €1,490 in potential cost savings across 5 suppliers and 6 materials.

Africa's first AI-native construction operating system — instant BoQ generation, live material procurement, and professional project management unified in one platform. From design brief to material order in seconds, not weeks.