MSc Business Analytics @ University of Exeter · Data Scientist / BI Analyst
Turning messy data into decisions — ML pipelines, BI dashboards, and optimisation models.
I'm an analytics professional with an MSc in Business Analytics (University of Exeter, 2025) and a BCom background. My work sits at the intersection of statistical modelling, machine learning, and business decision-making — from a dissertation-scale ML pipeline on 96k Brazilian e-commerce orders, to Tableau and Power BI dashboards tracking real revenue, to Excel-Solver LP models saving ~£100k in marginal revenue.
- 🎓 MSc Business Analytics — University of Exeter Business School
- 🧠 Focus: applied ML, causal thinking, decision analytics, dashboards
- 🌍 Based in India · open to remote & on-site roles
- ✍️ I publish notes on analytics workflows at my Netlify blog
| Project | Stack | One-liner |
|---|---|---|
| Olist E-Commerce ML (MSc Dissertation) | Python · XGBoost · SMOTE · KMeans | Predicting customer satisfaction across 96,353 Brazilian orders with leakage-resistant temporal validation. |
| RoBERTa Sentiment NLP | PyTorch · HuggingFace · Transformers | Fine-tuned RoBERTa hits 90.5% accuracy on IMDb vs LogReg 81.5% / NB 77%. |
| E-Commerce Sales Dashboard | Tableau | Refined executive dashboard for revenue, category & regional drill-down. |
| Car Price Prediction | Python · RF · XGBoost | Regression benchmark on 10k listings — XGBoost MSE 65,564 vs RF 239,412. |
| Blue City FC Revenue LP | Excel Solver · Simplex LP | LP model finding £4.85M optimal ticket-mix revenue with shadow-price analysis. |
| Chestnut Ridge Segmentation | R · KMeans · Hierarchical | Marketing analytics case study: 4-segment retail strategy backed by NbClust + GE matrix. |
⭐ If a project resonates, a star helps others find it. Thanks for reading!