•Marketplace Intelligence - Data Foundation Team
•Designed and built end-to-end Spark job management platform Dimrunner, an internal microservice to support running Uber's new version hexagonal clustering algorithms.
•Optimized Flink streaming ETL jobs of demand minute & supply minute of network pricing pipeline on Gairos(Uber's realtime data platform).
•Maintained data quality and reliability of rider session stats by monitoring consistency between upstream Gairos ElasticSearch cluster and downstream Hive cluster.
•Helped produce high-quality trip-level source data of UberPool service through data backfilling by partition in H1 of 2020.
