Software Engineer @ TD Bank · Data Engineering · Machine Learning · Toronto, Canada
I build data systems that turn messy, high-volume data into something reliable and useful — pipelines, automation, and ML models for fraud and anomaly detection.
- 🚧 Real-time Product Analytics & Integrity Platform: streaming live Wikipedia activity through Kafka, Spark, Iceberg, Trino, dbt and Airflow to track engagement and detect bots (in progress)
- 📄 Published in IEEE (SNPD 2025): Quantum-Enhanced Data Analytics for Crime Prediction · PDF
Languages: Python · SQL · DAX · Bash Data Engineering: Apache Spark · Databricks · Azure Data Factory · Airflow · dbt · Kafka ML & Analytics: scikit-learn · XGBoost · MLflow · PyTorch · TensorFlow · Pandas Cloud & Tools: Azure · AWS · Docker · Git · CI/CD · Power BI
| Project | Description | Stack |
|---|---|---|
| Loyalty Program Fraud Detection | Anomaly detection across 7 ML algorithms on 284K+ transactions; 98.6% ROC AUC | Python, XGBoost |
| Fraud Detection in Unlabeled Data | Compares Isolation Forest, One-Class SVM, LOF and DBSCAN for unsupervised fraud detection | Python, scikit-learn |
| MLflow Recommendation Engine | Recommendation system with experiment tracking | Python, MLflow |
| Automated Data Pipeline | Scheduled pipeline that ingests a Kaggle dataset weekly with GitHub Actions | Python, GitHub Actions |
I'm open to collaborating on data engineering and open-source projects, and always happy to talk data systems.



