I build practical ML systems for messy, real-world data — most recently the LLM core of a financial-advice product at Prentice Wealth Management. MSc Data Science at Bristol. I care about interpretability and shipping simple solutions to hard problems.
Projects I'd most like to talk through.
Re-engineered an Inductive Logic Programming framework in Python + Clingo: fixed long-standing bugs, extended functionality and sped up induction 4×. Shipped as a pip-installable package with an in-browser playground that runs the abduce → deduce → induce pipeline across ILP benchmarks.
LightGBM classifier for early sepsis detection in ICU patients, tuned for the cost-asymmetry of false negatives. SHAP feature attribution keeps every prediction explainable to non-technical decision-makers.
RAG finance assistant running on a local open-source model (Qwen3.5 via Ollama), measured against a labelled evaluation set with MLflow experiment tracking and unit tests in CI.
MobileNetV2 served through FastAPI, containerised with Docker and deployed on Google Cloud Run, with a Next.js frontend on Vercel.
Coursework, side projects and experiments.
Decoder-only transformer (attention, tokenisation, training loop) in PyTorch and an MLP in JAX, built from first principles.
MSc group project: demographic clustering + Bayesian ridge regression to forecast participation across London subpopulations, with uncertainty estimates.
Retrieval-augmented chatbot built on Llama and LangChain, with document chunking, an embedding store and conversation memory.
Species-logging app for the university diving society. Won 1st place in the cohort for software engineering.
Adversarial game-playing agent using minimax with heuristics, plus pathfinding, for the Scotland Yard board game.
Parallel and distributed Conway's Game of Life across AWS instances, benchmarked across configurations.
Production website for a traditional weaver — designed, built and deployed end-to-end.
Statistical Computing & Empirical Methods · Large-Scale Data Engineering · Data Science Methods & Practice · Visual Analytics · AI & Text Analytics
Graduating February 2027
Machine Learning · Applied Data Science · Algorithms · Computer Vision · Software Tools
Mathematics A* · Further Mathematics A · Physics A · Computer Science A
Mathematics 9 · Physics 9 · Computer Science 9 · Chemistry 9 · Electronics 9 · English Language 9 · English Literature 8 · Biology 8 · Spanish 8 · History 8
I'm available now for full-time data science and applied ML roles. Happy to talk through any of the work above — or anything tangential and interesting.