Data Analyst / Insight Architect
Hi, I'm Maggie! I specialize in exploratory data analysis and automating messy manual workflows, turning data into visualizations that are easy to understand. I bring a practical, truth-driven mindset to every project. Outside of writing SQL or Python, my interests revolve around art, animals, and a good story-driven video game. Based in Halifax, NS — open to remote.
End-to-end analysis pipelines from raw data ingestion to polished, stakeholder-ready dashboards. Proficient extracting insights from messy data, tracking KPIs and automating reports for long-term use.
Regression, hypothesis testing, and correlation analysis. Experienced in diagnosing data distributions, applying transformations to address skew, and clearly communicating statistical significance.
Relational database design and management. Experience writing complex multi-table queries and optimizing query performance using clustered and non-clustered indexes on large datasets.
Building automated ETL pipelines for data ingestion, wrangling, and transformation. and building reusable analysis frameworks that scale across teams.
Evaluating program impact through statistical effect size and establishing accurate KPI reporting. Familiar with AWS cloud architecture concepts and linking live data sources like SharePoint to reporting tools.
Supervised learning for regression and predictive modeling. Skilled in temporal feature engineering, Tukey outlier filtering, and optimizing compute-intensive models for production pipelines.
Statistical analysis and correlation visualization revealing a significant positive relationship (r = 0.581, p < 0.001) between childhood lead exposure and violent crime across Chicago neighborhoods with 15-year lag.
Web-optimized raster pipeline processing 220+ LIDAR-derived flood models. COG reprojection, XYZ tile generation, and real-time tide datum alignment for public emergency preparedness.
Random Forest ML model optimizing vehicle allocation across Halifax transit routes. ETL pipeline processing 2.5M+ passenger records with temporal feature engineering and outlier filtering.
Replace this with a full description of your project — the problem you solved, your approach, tools used, and the impact delivered.