Projects
A collection of my work and research projects
OntoSemantics
Jan 2024 - Jan 2001
NLP pipeline for extracting biological entity relationships from scientific texts using ontology-driven knowledge graphs.
Highlights:
- Designed relation extraction tools using ontologies and custom NLP parsing pipelines
- Represented extracted data as a Neo4J graph model; aimed at bioinformatics research workflows
- Submitted to ISBM and under peer-review for publication
Faust Streaming Contributor
Jan 2023 - Jan 2001
Contributor to an open-source stream processing library for building high-throughput real-time systems.
Highlights:
- Maintained Kafka-based streaming modules written in Python and AsyncIO
- Supported project revival after Robinhood’s deprecation of the original repo
Seal Counting with YOLO
Jun 2020 - Sep 2020
Used deep learning and object detection to automate population surveys of seals from aerial images.
Highlights:
- Trained and optimized YOLOv3 and YOLOv4 models using custom backbones on aerial wildlife datasets for conservation purposes
- Evaluated model precision across scaled architectures using TensorFlow, OpenCV, and Jupyter
- Dissertation project for MSc AI at University of St Andrews
Coronavirus Insights Dashboard
Mar 2020 - Jun 2020
Visualized global COVID-19 trends in a real-time dashboard using D3.js and Tableau.
Highlights:
- Built and deployed a web-based COVID-19 case visualization tool with live updates
- Applied a CI/CD workflow to automate dashboard deployment and updates
Superconductor Temperature Prediction
Feb 2020 - Mar 2020
Built ML models to predict the critical temperature of superconductors based on material properties.
Highlights:
- Applied regression and feature engineering techniques using Scikit-learn and pandas
- Performed EDA and visualizations to understand correlations in experimental datasets