I am Adeleke Akinrinade Kayode, a Senior Data Scientist, Machine Learning Engineer, and Statistical Researcher with 8+ years of experience delivering high-impact data science solutions remotely to clients in the USA, UK, Canada, Australia, and worldwide. Specialising in machine learning, Bayesian modelling, NLP, and financial analytics.
Six technical disciplines built over 8 years of research, consulting, and applied data science across healthcare, finance, and education sectors.
Bayesian hierarchical modelling, survival analysis (Cox regression, Weibull), discrete choice experiments, and causal inference using R and Python. Applied across clinical trials, public health, and social science research.
End-to-end ML pipelines for fraud detection, NLP text classification, loan default prediction, and customer churn. Production-grade implementations using scikit-learn, XGBoost, and TensorFlow with rigorous evaluation protocols.
Anti-money laundering (AML) detection, transaction pattern analysis, asset-liability management (ALM) gap analysis, and liquidity risk modelling for non-banking financial companies (NBFCs). Current work on TreasuryIQ at Technocolabs.
Interactive Power BI dashboards, ggplot2 publication-quality charts, matplotlib/seaborn analytical reports, and geo-visualisation with Python. Delivered for healthcare, livestock management, and financial sector clients in Nigeria.
Statistical analysis and manuscript preparation for peer-reviewed journals including Elsevier Finance Research Letters, IEEE Access, Virtus Interpress, and Journal of African Business. Proficient in LaTeX typesetting and APA/Vancouver citation systems.
Structured 12-week Data Science Fundamentals curriculum covering Python, statistical inference, machine learning, and SQL. Tailored for professionals entering the UK and Nigerian data markets. 3 sessions per week, project-based learning.
A selection of production-grade data science and analytics projects spanning financial technology, healthcare, and community data systems.
Production-grade asset-liability management (ALM) intelligence platform for non-banking financial companies. Identified a severe maturity mismatch across 5 asset and liability buckets from a 10,000-row synthetic dataset.
End-to-end fraud detection pipeline using XGBoost and Random Forest on 284,807 credit card transactions. Achieved 99.2% accuracy with 0.87 F1-score on the minority fraud class through SMOTE oversampling and threshold tuning.
Sentiment analysis and topic classification system built with TF-IDF vectorisation and a fine-tuned BERT model. Deployed on 50,000 customer reviews for a retail dataset with 94.1% accuracy across 6 sentiment categories.
A complete 12-week curriculum from Python fundamentals to machine learning and capstone projects. Step-by-step guides with real datasets, executable code, and exercises.
Installation, variables, data types, operators, control flow, functions, and Git basics. No prior programming experience required.
Pearson and Spearman correlation, simple and multiple linear regression, logistic regression, and the distinction between correlation and causal inference.
Whether you need statistical analysis for a research publication, a machine learning system for your business, or structured data science training for your team, let us discuss your specific requirements.
Receive tutorials, statistical tips, and analysis methods directly to your inbox. Published twice monthly. No spam.