Data Science & Analytics Consulting

Turning complex data into strategic insight

I build machine learning systems, NLP pipelines, and analytical tools that help organisations make better decisions. Specialising in financial analytics, risk assessment, and applied AI.

View Projects Get in Touch
14+
Years in Technology
Cambridge
Level 7 Data Science (2025)
London
Based in the UK
About

Data science built on industry experience

I spent over 14 years as a Senior Test and QA Leader at Sky, where I led cross-functional teams, optimised complex technical processes, and delivered a 25% reduction in resource demand through systematic quality engineering.

That experience taught me how large organisations actually work, where data gets lost, where decisions get made on instinct rather than evidence, and where machine learning can genuinely move the needle.

In 2025, I completed the University of Cambridge Data Science Career Accelerator (Level 7 Masters-level certificate), formalising the analytical and engineering skills I now apply through RJ Data Voyage. My capstone project, a prototype G-SIB Risk Assessment System, combined advanced NLP with financial metrics extraction to demonstrate how AI can support regulatory analysis at scale.

I work with organisations that need practical, production-ready data solutions rather than slide decks and theory. If the problem involves unstructured text, complex classification, time-series forecasting, or turning messy data into clear strategic insight, I can help.

Industry-Tested Approach
14+ years leading technical teams at Sky, delivering under real-world constraints with measurable outcomes.
Cambridge-Qualified
Level 7 Data Science certificate from the University of Cambridge (2025). Rigorous academic foundation in ML, NLP, and statistical methods.
Full-Stack Data Science
From data ingestion and NLP to model deployment and interactive dashboards. End-to-end delivery, not just notebooks.
0%
Resource demand reduction at Sky
0
Quarterly reports analysed (G-SIB capstone)
0
End-to-end data science projects
0
Years in technology leadership
Featured Work

Selected projects

Each project addresses a genuine business or regulatory challenge, built to production-ready standards with measurable outcomes.

Financial Regulation / NLP

Prototype G-SIB Risk Assessment System

81 quarterly reports analysed 3 G-SIBs

Regulators face a growing challenge: monitoring systemic bank risks at scale when the primary data sources are unstructured financial reports and earnings transcripts. Manual analysis is slow, inconsistent, and prone to oversight.

Analysed 81 quarterly reports across 3 G-SIBs (2023-2025) with automated NLP pipeline
FinBERT BERTopic ARIMA VADER GPT-2
View Case Study
Financial data analysis and risk assessment
Natural Language Processing

Customer Sentiment Classification Engine

27,586 reviews analysed 92% accuracy

A wellness centre was drowning in unstructured customer feedback with no systematic way to identify service improvement opportunities. Critical signals were buried in free-text reviews, unread and unanalysed.

92% classification accuracy with BERT-based model, surfacing actionable service insights
BERT PyTorch NLTK Transformers
View Case Study
Natural language processing and text analysis
Business Analytics / ML

Customer Behavioural Segmentation System

5 distinct clusters K-Means + PCA

Without data-driven segmentation, marketing teams rely on intuition to target customers. This leads to wasted budget, generic campaigns, and missed engagement opportunities across distinct customer clusters.

Identified distinct behavioural clusters enabling targeted marketing strategies
K-Means Scikit-learn PCA
View Case Study
Data-driven customer segmentation analytics
Anomaly Detection / Operations

Maritime Engine Anomaly Detection System

6 sensor channels Isolation Forest + PCA

Conventional monitoring systems miss subtle degradation patterns in operational machinery. By the time rule-based alerts fire, the damage is often already costly. Early, intelligent detection is the gap.

Automated detection of anomalous engine behaviour using statistical and ML methods
Isolation Forest PCA Pandas
View Case Study
Maritime vessel engine monitoring and anomaly detection

Additional Projects

Further work spanning forecasting, deep learning, and predictive analytics.

Forecasting
Demand Forecasting with ARIMA and Prophet
Time-series analysis of retail sales data, achieving 15% improvement in forecast accuracy over baseline methods.
+15% forecast accuracy
Predictive Analytics
Student Retention Prediction System
XGBoost and neural network models to identify at-risk students early, enabling proactive intervention by institutions.
XGBoost + SHAP explainability
Time Series / Public Safety
Baltimore Crime Pattern Forecasting
ARIMA-based forecasting system for predicting crime hotspots, supporting more efficient resource allocation.
SARIMA time series model
Financial Services
Bank Customer Churn Prediction
Classification models identifying at-risk banking customers for proactive retention strategies.
10,000+ customer records
Statistical Analysis
Statistical Hypothesis Testing Framework
Rigorous statistical testing applied to organisational data, distinguishing correlation from causation.
p < 0.05 significance testing
Deep Learning
Custom Neural Network Architecture
Deep network built from scratch with custom forward/backward propagation, hyperparameter optimisation, and SHAP interpretability.
From scratch forward + backprop
Expertise

Core competencies

Six areas where I deliver the most value, each backed by project work and production experience.

Machine Learning

Supervised and unsupervised methods including XGBoost, Random Forests, SVM, ensemble techniques, and clustering (K-Means, DBSCAN, HDBSCAN). End-to-end pipeline development.

Natural Language Processing

Transformer-based models (FinBERT, BERT, GPT-2), topic modelling (BERTopic), sentiment analysis, text classification, and domain-specific NLP for financial and customer data.

Forecasting and Time Series

ARIMA/SARIMA, Prophet, LSTM networks, and decomposition techniques. Applied to demand planning, financial projections, and operational forecasting.

Data Visualisation and BI

Interactive dashboards and visual analytics using Matplotlib, Seaborn, Plotly, and Power BI. Clear, decision-ready presentation of complex findings.

Statistical Analysis

Hypothesis testing (parametric and non-parametric), correlation analysis, causal inference, model validation, and experiment design including A/B testing frameworks.

Anomaly Detection

Isolation Forests, autoencoders, and statistical methods for detecting irregular patterns in operational and financial data. Real-time and batch processing approaches.

Contact

Let's work together

I am open to consulting engagements, contract roles, and permanent positions in data science, analytics engineering, and applied ML. If you have a complex data challenge or a team that could use senior analytical capability, I would be glad to hear from you.

Based in London, available for remote and hybrid work across the UK.

GitHub
LinkedIn
London, United Kingdom