B.Sc. (Hons) Software Engineering · M.Sc. Big Data Analytics
I am a Software Engineer and Solutions Developer with an academic foundation spanning a B.Sc. (Hons) in Software Engineering and a Master's in Big Data Analytics.
I specialize in building high-reliability desktop systems, offline-first mobile platforms, and data-intensive reconciliation engines. My focus is on transactional integrity, resilient CI/CD pipelines, and crafting clean, purpose-driven tools that solve concrete real-world operational problems.
- 🔭 Building: Offline-first ecosystems, enterprise desktop suites, and automated financial/data reconciliation tools.
- 💡 Engineering Philosophy: Offline resilience, strict data schemas, zero-bloat architecture, and comprehensive observability.
- 🎯 Target Domains: Full-Stack Architecture, Production Desktop (.NET 8 / WPF), Data Engineering & Cloud Integrations.
| Domain | Technologies & Libraries |
|---|---|
| Languages |       |
| Desktop & Mobile |     |
| Web & Back-End |     |
| Databases & ORM |      |
| Data & BI |     |
| DevOps & Quality |     |
🍽️ InventoryApp
Production-grade Windows Desktop Application (.NET 8 / WPF)
- Architecture: MVVM architecture on .NET 8 WPF with Entity Framework Core and embedded SQLite persistence.
- Observability & Diagnostics: Full Sentry error and crash monitoring integration with automated diagnostic runtime log and memory dump collection.
- Robust CI/CD: GitHub Actions workflow orchestrating build, unit/smoke testing, database migration verification, headless desktop startup tests, and automated win-x64 artifact distribution.
🛡️ MccShepherd
Offline-First Student Conduct & Pastoral Care Ecosystem
- Architecture: Built for institutional boarding-school environments with intermittent connectivity using a local-first SQLite mutation queue that automatically mirrors to Supabase Postgres.
- Multi-Tenant Security: Rigorous Row-Level Security (RLS) partition by school institution with PL/pgSQL database triggers driving dynamic progressive sanction rules and linear heat decay algorithms.
- Multi-Platform: Expo 57 / React Native cross-platform mobile client paired with a Next.js 14 executive leadership analytics dashboard.
A collection of lightweight, single-purpose utilities engineered for high performance, local execution, and strict integrity:
| Project | Description | Stack | Status |
|---|---|---|---|
| Sightline | Developer contextual navigator for AI skills, plugins, and MCP tools. Runs 100% locally with zero external npm dependencies. | Node.js 20+ | Active Engine |
| MergeMind | Offline inventory 120+ Day Age Analysis reconciliation engine. Cuts manual multi-workbook consolidation to <2 mins with 100% row transactional integrity. | Python · Streamlit · Pandas | Production Tool |
| AVES | Automated Vehicle Expenditure System for accounting ledgers. Ingests Pastel workbooks, classifies pooled costs, and produces reconciled reports. | Python · Streamlit · SQLite | Production Tool |
| NutriPlan | Nutrient-first, constraint-aware smart meal planning engine matching dietary requirements against in-stock ingredients. | Next.js 14 · Fastify · Prisma | In Development |
| PigeonPrompt v2 | Lightweight interface for rapid prompt engineering, structure testing, and model prompt refinement. | TypeScript · Web | Active Tool |
- Computer Vision: Access Control with Facial Recognition — Biometric facial authentication using OpenCV and Python.
- Data Warehousing & BI: NorthwindETL / etl-ssis — Relational data extraction, transformation, dimensional modeling, and OLAP star-schema design.
- Statistical Analytics: Billboard 100 Data Analysis — Exploratory and inferential data modeling in R.