Every project, in one table.
A selection of my dashboard case studies (21 pages, 68,545 rows) built from raw files, plus 7 Python and machine-learning notebooks. Filter by the tool you need, sort by size, or hover a row to preview the dashboard.
ProjectToolLink
1MetroBank Customer InsightsA 5-page Excel intelligence report on who MetroBank serves, where its money moves, which branches lose money and why customers complain.Excel.xlsx
2Maven Electronics Sales PerformanceA 4-page Power BI report for a global electronics retailer: what sells, who buys, which stores carry the business and where revenue is slipping.Power BILive
3HeartSync Product AnalyticsA 5-page Power BI product report for a global dating app, following the user journey from sign-up to match, subscription and safety reports.Power BILive
4Nexa Workforce AnalyticsA 3-page Excel HR report that showed an ageing workforce, middling engagement and a $1.59M training budget that wasn't paying off.Excel.xlsx
5Finance Job Market IntelligenceA live Power BI report on the finance job market: who's hiring, what they pay, which skills keep coming up and how fast recruiters hire.Power BILive
Total 5 shown
Python & ML
Notebooks
Machine-learning work from the Hamoye AI internship, the Omdena team project and personal projects. Results are from the original notebooks.
#NotebookResultLink
1Early autism detectionWorked with a global team on a predictive model for early autism detection. My part was exploratory analysis, data validation and feature assessment to improve data quality for the model.F1 0.98on autism indicators
2Student performance prediction (GitHub, opens in a new tab)Predicted student outcomes from 649 records and 33 variables. Compared six models (Random Forest, Linear, Ridge, Lasso, Decision Tree and SVR) and used feature engineering and correlation analysis to find the key drivers.R² 0.81best model, Random Forest
3Wildlife image classifier (GitHub, opens in a new tab)A CNN that tells apart buffalo, elephants, rhinos and zebras. Data augmentation, regularisation and tuning took test accuracy from 61.7% to 96.7% and reduced overfitting.96.7%test accuracy, up from 61.7%
4Bank customer churn (GitHub, opens in a new tab)Machine-learning churn analysis to find which bank customers are likely to leave, so retention teams can act before they do.—
5Proactive credit risk (GitHub, opens in a new tab)Predictive analytics on a credit-risk dataset to help lenders spot likely defaults early and protect their loan portfolio.—
6Financial distress prediction (GitHub, opens in a new tab)A Random Forest classifier that flags companies at risk of financial distress, packaged with a small prediction app.—
7Heart disease prediction (GitHub, opens in a new tab)A support vector machine model that predicts whether a patient has heart disease from clinical measurements.—
Contact
=IF(you_have_data, "let's look at it together", "say hello anyway")
Whether it's a role, a messy spreadsheet or a question about one of the case studies, messages come straight to my inbox.
Open to data analyst and BI work: full-time, contract or freelance, remote or on-site