Led the design and implementation of a Bronze/Silver/Gold pipeline on AWS S3, while working with Hadoop ecosystem components (HDFS, Hive) for data processing, consolidating REST API, CSV, and campus databases for 1,900 students. Owned data ingestion, transformation, orchestration, and BI consumption reporting as the sole data engineering resource. Built a custom multi-stage validation mechanism in PySpark / AWS Glue checking schema enforcement, nulls, referential integrity, and duplicates at the Bronze-to-Silver transition, writing failed records to a dedicated S3 DLQ prefix with JSON audit logs. Authored and maintained 8-10 production DAGs on daily/weekly schedules using task-level dependencies, retry logic, and email alerting on failure via Apache Airflow (MWAA). Utilized Linux basic commands (e.g., grep, awk, tail, file permissions) via secure shells to inspect intermediate execution logs and troubleshoot distributed data job failures, maintaining pipeline runtimes of 25–40 minutes. Designed star-schema dimensional models in Amazon Redshift with distribution keys on high-join columns and compound sort keys on date and campus dimensions, using the COPY command and materialized views to optimize query latency. Processed realistic education-sector volumes of approximately 10,000 to 25,000 transactional records per day (spanning attendance logs, LMS engagement, admissions, and fee collections) across 5+ source feeds, maintaining >98% on-time execution reliability. Delivered Power BI and Tableau dashboards tracking student conversion velocity, weekly enrollment trends, and fee collection across primary schools and training centres, directly supporting capacity planning and leadership decision-making.