I’m Andrea, and I’ve been working within the data, ML & AI space for nearly a decade.
Throughout my career, I’ve had the opportunity to work on a wide range of projects across different sectors, including healthcare, public policy, and education, utilising a variety of tooling and techniques, some of which I like to write about – (check out my blog and projects section)!
Background
I studied economics and statistics, and started my career in the Government Economic Service (GES). I spent nearly 5 years at the Department for Education across a range of policy areas, starting in economic modelling and gradually moving into broader data science work.
Since then, I’ve worked across the public and private sectors and built experience across the full ML lifecycle: from scoping projects, building prototypes, developing data pipelines, to productionising models, setting up monitoring, and evaluating how they perform in practice.
Outside work, I write about the things I build — mostly statistical methods, ML workflows, and the occasional detour into something I’m trying to understand better. If you want to collaborate or just talk through a problem, get in touch.
Bupa
Recently, I’ve been doing some contract work with Bupa on their AI Emblematic initiative - a multi-year programme to use AI and data science to improve the health outcomes of their customers.
My work there has been focused on building production analytics pipelines, scoping predictive risk models, and generating evidence to support both strategy and member-facing health planning. This has involved a mix of data engineering, model development, and analysis that can inform both internal decision-making and customer-facing tools.
Health Economics and Outcomes Research Ltd.
At HEOR, most of my work was focused on large-scale real-world evidence studies and health economics projects for pharmaceutical clients, using UK healthcare databases and datasets like CPRD Aurum, HES, and ONS.
That included things like parametric survival modelling, healthcare cost modelling, and evidence synthesis. Examples include modelling the cost of obesity-related healthcare, and global meta-analysis of meningitis carriage for vaccine development.
UK Health Security Agency
At UKHSA, I was initially supporting the UK’s COVID-19 response, specifically working on the NHS COVID-19 App, collaborating with academic partners to publish research evaluating and estimating the effectiveness of the app in preventing covid cases, and subsequent hospitalisation and deaths.
Once the programme ended, I was also involved in other health protection efforts, such as Mpox, Avian Influenza, and Hepatitis B, where one of my later projects involved training text classifiers using Transformers-based architectures to extract hepatitis markers status from unstructured clinical reports.
Department for Education
At the Department for Education, I worked predominantly on forecasting and policy modelling, with a focus on policy evaluation and practical economics. I worked across a variety of policy areas, including teachers funding, school infrastructure and capital expenditure, and higher education.
One of my highlights was working on the Teacher Pay-Bill model and the School’s Cost Pressures Model, which were used by the government to predict spending as well as funding growth for schools in England.
Science (2024)
“Drivers of epidemic dynamics in real time from daily digital COVID-19 measurements” showed how daily digital signals can be used to track changing epidemic dynamics in near real time. I contributed to the analytical work that connected those measurements to actionable epidemiological insight.
Nature (2023)
“Digital measurement of SARS-CoV-2 transmission risk for precision epidemiology” demonstrated how digital contact data can sharpen estimates of transmission risk. My contribution focused on turning high-volume behavioural and epidemiological data into evidence that could support more precise public-health analysis.
Nature Communications (2023)
“Epidemiological impacts of the NHS COVID-19 app in England and Wales throughout its first year” quantified the real-world impact of the NHS COVID-19 app over its first year. I contributed to modelling and evidence generation that helped explain the app’s population-level effect in plain, decision-relevant terms.
Birkbeck, University of London
MSc Applied Statistics & Operational Research — Distinction
University of Nottingham
BSc Economics — First Class Honours