At Brickclay, I built and maintained a Python ETL pipeline processing 500K records daily in production. I also identified and resolved pipeline bugs and documented pipeline logic and validation findings.
For CrickVision, I designed an end-to-end data system ingesting about 10M records, with an ETL pipeline, API layer, and frontend components. I built a FastAPI backend with modules for auction simulation, fatigue tracking, and bowling-synergy analytics.
On my MLOps Pipeline project for Fake News Detection, I built automated data drift detection and an observability stack using MLflow, Prometheus, Grafana, and Alertmanager. CI/CD integration reduced deployment time by 70%.
I also designed a near-real-time retail data warehouse using star-schema modeling and built a Python ETL pipeline with an extended HYBRIDJOIN algorithm. As a Database Management Teaching Assistant, I assisted students with database concepts and practical coursework.
