Data scientist and analyst working in R — from raw, messy public datasets to statistical machine learning models, interactive Shiny dashboards, and findings that non-technical stakeholders actually understand.
I came to data science via a scenic route: a BA in History, an MA in History and Heritage, and then a graduate certificate in Data Science at American University. I am now a incoming PhD student at the University of Southampton's Winchester School of Art researching matrilineal heritage, utilising my data science skills. That mix is my superpower — I can build and validate a model, and I can tell you what it means,why it matters, and what to do next.
My happy place is the full pipeline: cleaning stubborn real-world data, choosing the right method, validating it, and turning the output into something interactive and genuinely usable. I've led a project team end-to-end and written the documentation needed to make findings accessible.
I'm based in the UK and looking for part-time data science, data analytics, and research-analysis roles where rigour and clear communication both count. I would consider any full-time work that would commence before September of 2026.
The methods and tools I reach for, grouped by what they're for.
Supervised learning for prediction and classification — linear and logistic regression, regularised models, KNN, discriminant analysis, splines, and classification trees — chosen and compared properly.
Honest model assessment with k-fold and leave-one-out cross-validation, bootstrap resampling, and clear-eyed reporting of error rates instead of cherry-picked accuracy.
Interactive Shiny apps, Quarto reports, vignettes, and presentations — built so that the person reading them doesn't need a statistics degree to act on them.
Mixed-methods grounding: qualitative research from history and heritage training — archival work, primary sources, critical analysis — alongside quantitative statistical methods, so numbers and context get equal rigour.
Real datasets, real constraints, working deliverables.
A multi-tab Shiny app exploring HM Land Registry price-paid data. I led the team, wrote the data-cleaning script, built the full application, and authored the README, Quarto vignette, and presentation.
An exploratory analysis of global life expectancy, income, and population using Gapminder open data — visualising 50+ years of change across continents, modelling the income–longevity relationship, and publishing the findings as a Quarto report.
A standalone Shiny app that recommends books with content-based filtering — turning a recommender-systems concept into an interactive tool anyone can play with.
Where the quantitative skills and the storytelling instincts come from.
Graduate coursework in statistical machine learning, data science, and applied analytics in R.
Research methods, archival work, and long-form analytical writing.
Foundation in evidence-led argument and primary-source analysis.