Skip to content
View GarethMackenzie's full-sized avatar

Block or report GarethMackenzie

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
GarethMackenzie/README.md

Gareth Andrew Mackenzie

Data Analytics · Business Intelligence · Analytics Engineering · Applied Machine Learning

I build analytical systems that turn operational data into clear, measurable decisions using Power BI, Python, SQL, data modelling and machine learning.

Johannesburg, South Africa

Email · LinkedIn · GitHub


Featured projects

01 | Insurance Claims Intelligence

Power BI · Analytics Engineering · Data Quality · SQL · Python

A source-controlled Power BI decision-support project built on 75,000 synthetic claims. The repository contains an editable PBIP/PBIR report, a TMDL semantic model, 79 explicit DAX measures, Power Query transformations, SQL analysis, a deterministic Python data pipeline, 8 report pages and 33 automated QA checks. GitHub Actions rebuilds and validates the project on every push and pull request.

View the flagship project · Review the QA workflow

02 | Motor Claims Risk Triage

Applied Machine Learning · Decision Science · Model Governance

A synthetic-data case study for capacity-constrained review prioritisation. The project compares classification approaches through a leakage-safe pipeline, stratified cross-validation, an untouched holdout set and ranking metrics at practical review capacities. It also examines model explainability, sensitive-feature governance and human oversight.

View the machine-learning project

Both repositories use synthetic data. Their analytical results are portfolio evidence, not employer results or production performance.

Technical stack

Area Tools and methods
Analytics and programming Python · SQL · R · Excel
Business intelligence Power BI · DAX · Power Query · Tableau
Data modelling and analytics engineering Dimensional modelling · Semantic modelling · PBIP · PBIR · TMDL · Data quality
Machine learning scikit-learn · XGBoost · Imbalanced classification · Model evaluation
Engineering and delivery Git · GitHub · GitHub Actions · CI/CD · Testing

Professional profile

My background combines business and operations experience with applied analytics. I work on data quality, performance measurement, process improvement, risk-aware decision support and communication for operational and executive audiences. Insurance and claims are established areas of domain knowledge, not the limit of my analytical work.

Professional impact

These outcomes relate to professional claims and operations work. They are separate from the synthetic portfolio results above.

Area Outcome
Claims handling 40% faster following process redesign
Fraud identified or prevented R500K+
Claims managed Approximately 200 motor and agricultural claims per month

Experience

  • Insurance Specialist, Old Mutual Insure (Oct 2023 to present)
  • Claims Specialist, Auto & General Australia (Oct 2022 to Jun 2023)
  • Customer Support Specialist, Bob Group (Oct 2021 to Sep 2022)
  • Associate Claims Coordinator, Innovation Group South Africa (Sep 2018 to Aug 2020)

Certifications

  • Microsoft Certified: Azure Data Scientist Associate, issued Nov 2025, expires Nov 2026
  • Microsoft Certified: Power BI Data Analyst Associate, issued Oct 2025
  • Oracle Cloud Infrastructure 2025 Certified Data Science Professional, issued Jul 2025, expires Jul 2027
  • Introduction to Data Analytics, TAFE NSW, issued Oct 2025
  • Lean Six Sigma Black Belt

Credential links will be added only when the corresponding public verification URLs are available.

Career focus

Senior Data Analyst · Data Analytics · Business Intelligence · Analytics Engineering · Operations Analytics · Decision Support · Data Science

Insurance analytics remains a secondary domain specialism.

Pinned Loading

  1. insurance-claims-intelligence-powerbi insurance-claims-intelligence-powerbi Public

    Executive-grade Power BI insurance claims analytics portfolio featuring PBIP, TMDL, DAX, Power Query, SQL, synthetic data quality, financial exposure and risk-based review prioritization.

    Python

  2. motor-claims-triage motor-claims-triage Public

    Machine-learning claims risk triage using Python, leakage-safe modelling, ranking metrics, lift analysis, testing and human-in-the-loop review prioritisation.

    Python

  3. GarethMackenzie GarethMackenzie Public

    Data analytics, business intelligence, analytics engineering and applied machine learning portfolio.