I turn raw data into decision-ready insights — designing Power BI dashboards, building semantic models, and writing SQL that holds up at scale. I also build end-to-end machine learning pipelines in Python, from data collection through model training and evaluation.
- Data & BI Analyst at E.ON Energy Solutions, working with Microsoft Fabric, Power BI, and T-SQL
- Hands-on with the full ML workflow in Python — preprocessing, modelling, and evaluation
- M.Sc. Digital Transformation @ FH Dortmund
- English (C1) · German (B1, telc certified) · Bengali (native)
- LinkedIn · sarazmin28@gmail.com
| Data & BI | Power BI · Microsoft Fabric (Fabric SQL) · Data Warehousing · Semantic Modelling · Data Visualization |
| Data Science & ML | Python · pandas · NumPy · scikit-learn · TensorFlow · NLP · Jupyter |
| Databases | SQL · T-SQL · MySQL |
| Languages & Web | TypeScript · JavaScript · C# · React · Next.js · Node.js · ASP.NET |
| Tools & Methods | Git · Azure DevOps · Jest · Cypress · Figma · Agile/Scrum |
YouTube Comment Sentiment Analysis — Python · NLP · Machine Learning End-to-end sentiment pipeline on real YouTube comments pulled via the YouTube Data API: data collection, manual labelling, preprocessing, and a model comparison across Naive Bayes, Logistic Regression, SVM, Random Forest, and a neural network.
Customer Sales Analytics Dashboard — Power BI Interactive sales dashboard on the Kaggle Superstore dataset — modelling sales, profit, and regional performance with drill-downs that surface the metrics a business actually decides on.