May 2017 – May 2019 · 2 yrs · Bengaluru, Karnataka, India

• AIR Platform For Qubole : Designed an built a data ingestion framework for collecting key metrics related to a job execution life cycle on the Qubole platform which was later processed to generate Alerts Insights and Recommendation for the workloads. The metric exporter API was built using a drop-wizard framework and we used spark data pipelines for processing the ingested data using Airflow as the job orchestrator.
• Similarity Service: Designed and Built a service for identifying similar Hive/Presto or Spark Programs being executed on the Qubole Platform. The service provide capabilities for identifying recurring workloads across the platform and to track various run time metrics across these different runs enabling users to identify anomalies in run times or other relevant execution metrics.
• CI/CD Orchestrator For Data Pipelines: Built a CI/CD framework for automating schema evolution and deployment of Data Pipelines. The framework had the capability for a tag based deployment of the data pipelines code that would also in turn take care of any schema level changes needed for the functioning of the pipeline code at that version,

Jul 2015 – May 2017 · 1 yr 10 mos

• Designed and developed from scratch, dynamic ads solution of Facebook for Vizury clients. What makes this product different from the existing solutions in the market is it gives a finer control over the ROI for the advertising spends. It currently serves the majority of facebook advertising that Vizury does.
• A REST layer using jersey, jackson, jetty, hibernate which helps in creation and updation of various facebook entities required.

Sep 2013 – Jun 2015 · 1 yr 9 mos

• As a product engineer in analytic's team i have worked on various projects
• Big Data Analytics
• Setting up and maintenance of hadoop cluster that which is widely used for reporting and ad hoc querying.
• Writing custom reports on big data using Apache Pig.
• Improvements for existing ad hoc querying infrastructure using Apache Hive for better performance and stability.
• Hive Source Code Modifications
• Modified Apache hive for a particular use case which has helped in optimizing the query planning phase.

Jun 2013 – Aug 2013 · 2 mos · Bengaluru, Karnataka, India

• - During a short stint of 3 months i underwent a training session which helped a lot in transitioning my college level skills to what is required in industry level.