Data Analyst · Data Scientist · ML / NLP · Financial Data
MSc Data Science & Analytics (Merit) · London, UK · Turning financial and behavioural data into decisions: from Python, SQL and dashboards to ML pipelines and deployed software.
I turn messy, real-world data into decisions by querying and cleaning it, analysing it, modelling it where it helps, and communicating the results through clear dashboards and honest reporting. My background bridges data analytics, machine learning, and software engineering, with a specialism in financial and behavioural data: pricing analytics, credit risk, and market/sentiment trends.
- 🎓 MSc Data Science & Analytics — Merit, Brunel University London (2024–25)
- 📊 Analytics · SQL · pandas · Tableau · Power BI · Excel · EDA · statistical testing · BI storytelling
- 🤖 ML / NLP · scikit-learn · PyTorch · FinBERT / DistilBERT / VADER · walk-forward validation · SHAP
- 💻 Engineering · Python (CherryPy) · REST APIs · JavaScript · D3.js · Docker · Git · Grafana
- 📍 London, UK · Open to Data Analyst, Data Scientist and ML/NLP Engineer roles across the UK, including financial data, risk analytics and NLP work
Each repository takes a real question, works it end-to-end on real data, and reports the findings honestly — the numbers, the visualisations, and the caveats.
| Repository | What it does | Tools | Headline |
|---|---|---|---|
| news-stock-prediction ⭐ | MSc dissertation — 14-phase ML/NLP pipeline predicting next-day index direction across 5 markets from news sentiment | FinBERT · scikit-learn · PyTorch · SHAP | AUC-PR up to 0.587; signal concentrates to 0.69–0.79 on four of five indices in an identifiable market regime (n = 31) |
| credit-risk-ml-hpc | Credit-risk classification on 100,000 records — single-node (R) vs distributed (PySpark), with an actionable risk rule | R · PySpark · MLlib | AUC 0.8748 (R) / 0.8460 (PySpark); debt > $30K + payment delays > 12 days → 3.6× default risk |
| djia-financial-visualisation | Interactive DJIA dashboard with a separate FRED macro chart — hand-coded front end, zero build tooling | D3.js v7 · Bootstrap · vanilla JS | Live interactive demo |
| airbnb-pricing-analytics | Persona-driven pricing dashboard over 12,800 multi-country listings — Tableau primary, Power BI cross-validation | Tableau · Power BI · pandas | Decision-oriented BI storytelling with Munzner-based encoding |
| crypto-twitter-sentiment-2021 | NLP pipeline over 50,293 Bitcoin and Ethereum tweets (2021) — NLTK cleaning and lemmatisation, VADER scoring, Pearson correlation with BTC/ETH returns at daily, weekly and monthly scales | Python · NLTK · VADER · pandas · SciPy | No significant same-period link; weak negative correlation only after smoothing or lagging, clearest for Ethereum (7-day r = −0.15) |
| covid-sentiment-stocks | BSc dissertation — correlation study of public sentiment over 89,632 tweets vs equity-index movement during COVID-19 | Python · VADER · scipy | Statistical (Pearson) analysis of sentiment–market linkage |
Querying & data · SQL · Python (pandas, NumPy) · Excel · R Analysis & ML · EDA · statistical testing · regression · scikit-learn · XGBoost · LightGBM · PyTorch (LSTM, 1D-CNN) NLP · FinBERT · DistilBERT · VADER · 🤗 Transformers BI & visualisation · Tableau · Power BI · D3.js · matplotlib · seaborn Engineering & data eng · Python (CherryPy) · REST APIs · JavaScript · Docker · Git · PySpark · Grafana · MySQL · MongoDB
Beyond analytics, I've shipped real software: a year-long industrial placement at STFC (UKRI) building a monitoring dashboard for its research computing platform (Python/CherryPy back end, REST APIs, Grafana, and a JavaScript/HTML/CSS front end, delivered Agile with JIRA), plus hand-coded front-end data-viz work in D3.js and vanilla JS. I hold a BSc (Hons) Computer Science with Professional Practice (2:1) alongside the MSc, so I read and write code as an engineer, not just a notebook author.
- 📚 Working toward the Google Advanced Data Analytics Professional Certificate (Google) and the Microsoft Data Analysis with SQL, Excel & Power BI Specialization (Microsoft)
- 🎯 Building depth in financial data analytics and applied ML — pricing, risk, and market analysis
- 💬 Open to conversations about Data Analyst and Data Scientist roles in the UK, including financial data, risk analytics and NLP work
