I am a BSc (Hons) Artificial Intelligence undergraduate with practical experience in data analytics, machine learning, database management, and business intelligence reporting.
I work with Python, SQL, Power BI, Excel, Power Query, and DAX to prepare data, build ETL pipelines, create data models, and develop interactive dashboards that support actionable decisions.
I have completed end-to-end machine learning projects using real-world datasets, including classification, forecasting, clustering, anomaly detection, and computer vision. My work includes evaluating models with accuracy, R-squared, precision, recall, F1-score, and cross-validation metrics.
I have also designed normalized SQL Server databases, created role-based reporting queries, and applied relational database design principles such as normalization, integrity constraints, and CRUD operations.
I am seeking a Data Analyst internship where I can apply my analytical, technical, and problem-solving skills to transform data into meaningful business insights while continuing to grow in a professional environment.
I designed an ETL pipeline using Power Query and developed a snowflake schema data model for warehouse inventory analysis.
I created DAX measures for total revenue, average price, and total quantity, then produced a 10-visualization Power BI dashboard covering stock, pricing, and warehouse KPIs.
I developed an end-to-end machine learning pipeline using 119,000 real-world records for hotel and tourism analysis.
Logistic Regression achieved 82.0% accuracy for cancellation prediction using 5-fold cross-validation, while an ANN achieved 92.7% R-squared for revenue forecasting. I also used K-Means and DBSCAN to identify three customer segments and HBOS anomaly detection to flag 1,168 anomalous bookings.
I designed a nine-table, 3NF-normalized Microsoft SQL Server database for Kandy General Hospital.
I implemented integrity constraints and wrote eight role-based SQL reporting queries to support hospital data management and reporting needs.
I built a five-class indoor image classifier using transfer learning and the MIT Indoor Scenes Dataset. The model classified kitchen, bedroom, bar, living room, and warehouse scenes.
I evaluated three CNN architectures: ConvNeXt-Tiny achieved 92.77% test accuracy, EfficientNetB0 achieved 91.53%, and ResNet50 achieved 90.50%. I used class-wise precision, recall, F1-score, confusion matrices, and a 70/15/15 train-validation-test split for model selection.
I led the classification domain of a group systematic literature review based on 20 peer-reviewed sources. I reviewed YOLO, R-CNN, and SVM approaches for vehicle detection and road-surface classification.
I co-proposed IITMS, a four-layer machine learning pipeline incorporating SHAP interpretability for real-time transport deployment.
BSc (Hons), Artificial Intelligence
BSc (Hons) Artificial Intelligence undergraduate program.
Google AI Professional Certificate, Artificial Intelligence
Completed seven courses and built more than 20 AI artifacts.
Diploma, Spoken English
Completed with a Distinction Pass and received A+ grades in reading and listening comprehension.
G.C.E. Ordinary Level, English Medium
Achieved 6A, 1C, and 2S results.
This professional hasn’t added portfolio projects yet.
This professional hasn’t listed any services yet.
