Systems and Infrastructure Engineer at LinkedIn Media Infrastructure

• A quick note on a part of our research: https://bridgingbarriers.utexas.edu/news/whole-communities-whole-health-study-goes-mobile.

• Developed a full stack mobile and web application named Hornsense for the Whole Health Whole Communities Project undertaken by the University of Texas at Austin using Django, AWS, Terraform, MySQL and UI using React Native and React.

• Streamlined the backend application efficiency, resulting in app efficiency, better user experience and increasing the user app usage by 30% with push notifications, scheduled purging and archival of backend data using AWS Lambda and cron jobs.

App link: https://apps.apple.com/us/app/wcwh-hornsense/id1589284088

• Developed a post ETL data quality pipeline that reports the data quality metrics for the Platform Security team by using Python, Apache Spark, Databricks, AWS Deequ and Azure Data Factory.

• Developed a configurable pipelined template driven report generation engine for creating a snapshot of Tesla Infrastructure dashboard using Python Jinja templating, Azure Data Factory and Databricks.

• Developed a REST framework for the LinkedIn feed infra team for comparing the quality of indices among replicas using Java, Rest.li, Spark, Kafka, Rocks DB and Espresso DB.

• This framework enables the LinkedIn feed to be more relevant to users’ preferences by 11%.

• Developed workflow orchestration engines and schedulers for an ETL process in payments transaction processing using Temporal, Golang, containerization and virtualization using Docker and Kubernetes jobs.

• Built from scratch an auto-rollback recovery workflow of an in-memory KV store incase of transaction declines ensuring high availability. This helped in decreasing the time and effort of manual rollback and error detection by 5x.