I make data trustworthy, then I make it predictive.
Engineering data. Modelling outcomes.
Analytics engineering · Data engineering · Manchester, UK
I build the pipelines, warehouses and tested models that reporting and applied ML depend on.
Python
R
SQL
PySpark
dbt
Dagster
scikit-learn
HuggingFace
Power BI
Sources → Ingestion → Transformation → Serving → Consumption
Where the wrong number starts.
A wrong number is rarely a reporting mistake. It is a decision someone made further upstream: which rows to join, what counts as a customer, when a period closes.
By the time it surfaces on a dashboard it has been wrong for weeks. That upstream layer is where I work, across three disciplines.
Analytics engineering.
I model business logic into curated tables analysts and downstream apps actually use. dbt for transformations, Kimball for the modelling pattern, version-controlled and tested. I structure models in raw, staging, intermediate and mart layers.
SQLKimball
GitData engineering.
I build the orchestration and ingestion layer. Dagster for orchestration, Python for extractors and transformations, PostgreSQL for the warehouse. This includes schedules, data quality tests and monitoring.
PythonPostgreSQLApplied machine learning.
I build prediction models and recommender systems on top of clean data layers. XGBoost for the modelling, SHAP for explainability, MLflow for tracking, PySpark for scale. Football scouting (MSc thesis), Steam game recommendations, churn models on customer intelligence data. SHAP values explain each model's predictions.
scikit-learnSHAPPySpark
MLflowThree of eleven projects that show my work in more detail than my CV.
Account Intelligence Platform
A greenfield Account Intelligence platform built over 16 weeks at Force24 for CSMs, accounts, and stakeholders, shipped with the engineering team into a live production environment. I owned the data layer end to end, integrated it into the FastAPI service through endpoint changes and Redis caching, and shipped features and enhancements across the Angular frontend. I built it on the principle that a dashboard should display data and also lead the user to a specific action. Confidential, sanitised case study.
PostgreSQLPython
Redis
Angular
Agentic ELT Data Platform for Customer Intelligence
MSc dissertation in a live B2B SaaS environment, with the sanitised code public on GitHub. End-to-end JSONB-first ELT platform and three-model churn intelligence stack (survival, XGBoost via PostgresML, and DR-Learner causal inference), surfaced via FastAPI, Angular, and an MCP endpoint for agentic LLM access. Over 1 million records ingested across multiple vendor APIs, 48 dbt models.
PythonPostgreSQLdbtDagsterPostgresMLeconmlMCP
Pharmaceutical Side Effect Classification
Production-grade Python package classifying free-text adverse-event descriptions into a MedDRA-style taxonomy of ten clinical categories across 11,825 marketed medicines. sklearn Pipeline with ColumnTransformer, joblib-serialised end to end, pytest fixtures, GitHub Actions CI matrix on Python 3.10, 3.11, 3.12.
Pythonscikit-learn
pytest
GitHub ActionsArticles about my work.
29 July 2026
The 47th CV.
What building an AI agent for the job market taught me about volume, precision, and building on another developer's open-source project.
3 July 2026
Dashboards inform, agents act.
Putting a Model Context Protocol endpoint on a churn model, and why the data layer underneath decides whether the agent can be trusted.
Summary.
Three years of data work: mortgage analytics in Abuja, CRM data in Istanbul, and most recently a production data layer in the UK. In each role I built the data layer before the reporting on top of it.
Two master's degrees (Salford, Distinction; Bahcesehir University in Türkiye, 3.67 GPA) and an engineering undergrad at the University of Benin in Nigeria. Eight certifications across Google Cloud, Coursera, IBM, and LinkedIn Learning. Two years and nine months as a Microsoft Learn Student Ambassador, running cross-university Azure workshops and a community hackathon with fellow ambassadors and interested learners.
My focus in 2026: dbt, dimensional modelling, CI/CD for data, helping analysts work faster, and ML models with SHAP explanations for their predictions.
Let's talk.
If you work in analytics or data engineering and have a question about dbt, Dagster or warehouse testing, you can contact me.