Data analyst and bioinformatician based in Dublin. I build careful, reproducible analyses on biological and health data, and the pipelines and systems that run them.
MSc Microbiology (completing 2026), Carl von Ossietzky University of Oldenburg · Diploma in Data Analytics and Business, CCT College Dublin
Reducing False Positives in Metagenomic Pathogen Detection · MSc thesis Comparative genomics, a 61-genome Anvi'o pangenome and an explainable XGBoost classifier that separates E. coli O157:H7-specific markers from regions shared with harmless relatives. AUROC 0.761 against 0.552 for a BLAST screen, with a shuffled-label null control and SHAP explanations. DOI
Protein Repertoire Analysis: A. aromaticum EbN1 on Benzoate · proteomics Anoxic versus oxic protein repertoires across gel-slice, membrane and shotgun fractions, from Mascot identifications to condition-specific proteins and their functions.
Irish Acute Healthcare Intelligence · in progress Public HSE, Department of Health and CSO data (HIPE discharges, waiting lists and inpatient healthcare-associated infection prevalence) analysed in Python, SQL and Power BI, with valid denominators, documented data lineage and data-quality checks.
Tradox · private A live FX signal and alerting system: Python and FastAPI service with TradingView webhooks, PostgreSQL, CI/CD deployment to a Linux server, and a governed release process.
Multi-agent research tooling · private Agents built on the Anthropic Claude and OpenAI APIs with MCP servers, for autonomous research, monitoring and email digests.
Languages Python · R · SQL · Bash
Data and ML pandas · NumPy · scikit-learn · XGBoost · PyTorch · SHAP · statistical modelling
BI and cloud Power BI · Tableau · Databricks · ETL · AWS · Azure · GCP
Engineering FastAPI · PostgreSQL · Docker · Linux · Git · CI/CD
Bioinformatics NUCmer · Anvi'o · BLAST · SPAdes · BWA · SAMtools · Bioconductor · Illumina QC
AI Anthropic Claude API · OpenAI API · MCP