As a Software Engineering Intern at Different.ai, I develop scalable backend systems and cloud-based data pipelines that power customer-facing analytics. I work with Java, Spring Boot, Apache Spark, AWS Glue, Kinesis, Redshift, Terraform, and Temporal to build reliable and efficient data infrastructure. I re-architected the LinkedIn Ads reporting pipeline, significantly reducing workflow history, activity calls, and database reads while improving scalability. I also built real-time CDC streaming pipelines, enabling near-real-time data availability instead of batch processing. Additionally, I helped privatize the company's Redshift Serverless infrastructure by migrating ETL jobs to secure private VPC connectivity with zero downtime. I also optimized Spark ETL jobs to eliminate memory issues, improving throughput and reducing infrastructure costs, while creating customer-facing analytics dashboards using ThoughtSpot.
