Data Engineer with 4.3 years experience in scalable data pipelines and ETL development

The Automated Reconciliation Process is designed to eliminate manual data comparison and ensure accuracy between multiple data sources, Using Power Automate, Dataverse, and Power BI, the process automatically extracts, validates, and compares Actual vs Expected data to identify mismatches and exceptions.

Architected an automated data extraction pipeline integrating Salesforce, SQL Server, and email sources to enable end-to-end reconciliation without manual intervention.

Developed usage calculation, validation, and exception-check logic within Power Automate flows, reducing billing discrepancy resolution time significantly.

Designed and maintained Dataverse tables to store processed reconciliation results, audit logs, and exception records with full traceability across processing stages.

Implemented a segment-level re-run mechanism to archive, truncate, and reload specific data segments, preventing unnecessary full reloads.

Built five structured Power BI reports in a dedicated reporting workspace, each covering a specific aspect of data validation and process monitoring.

Enforced data governance by maintaining detailed audit logs in Dataverse, ensuring full traceability of data processing stages and exception handling.

Implemented a fall back mechanism to ensure backup availability of source data for recovery and validation purposes.

Contributed on a project for ASL Aviation’s, a UK-based cargo airline, where development of dashboard using Qlik BI based on client requirements. Data was extracted from Oracle DB using Debezium (CDC) and processed through Dremio Lakehouse, with final storage in AWS S3.

Developed optimized SQL queries in Dremio for business analysis purpose.

Migrated legacy Talend-based ETL workflows to Dremio Cloud, enabling direct SQL queries on S3 data lake and reducing dependency on ETL tools.

Delivered Inventory, Operational, Engine Status, and Daily Fleet performance reports.

Improved query performance by implementing Reflection in Dremio, reducing response time refresh in Qlik dashboard.

Created Virtual Dataset (VDS) from physical datasets (PDS) stored in the S3 data lake.

Collaborated with team of 4 to understand business requirements and create reports.

Resolved data discrepancies in Qlik dashboards, enhancing data accuracy and reliability.

Designed and implemented a 3-tier Dremio architecture (Dev, Staging, Prod) to separate development and production workloads, improving data governance and preventing accidental changes

Implemented Role-Based Access Control (RBAC) in Dremio to secure sensitive production data and enforce controlled access for different teams.

Collaborated with Functional Business Analyst to define data requirements and refine business logic.

Led the design and implementation of a metadata-driven data warehouse in Azure, using Azure Data Factory (ADF) as the ETL engine. The solution processes ~500 MB of structured and semi-structured data weekly through incremental loads. SCD Type 2 logic was implemented across 92 tables to capture historical changes. Oversaw end-to-end delivery, team coordination, and source- destination integration using Azure Key Vault for credential management.

Russell and Smith is a US-based company that handles parts for Ford automobiles. Migrated inventory and invoice data from AWS S3 to PostgreSQL RDS using AWS Glue to streamline inventory tracking and financial reporting.

Led the data migration and integration effort for a global textile and apparel group headquartered in Hong Kong. Consolidated data from diverse on-premises databases into Azure SQL Database to enable centralized reporting and analytics using Power BI.

Led the transformation and loading of data from various sources into Snowflake DW for Australian-based Real Estate Company.

Designed and developed ADF Pipeline to extract data from various data sources such as SQL Server, Excel spread sheets, Flat files &

APIs into the Snowflake DW for further Data Analysis and Reporting by using multiple transformations provided by ADF.