Mansi Bhardwaj

Experience

• 

  Galytix

  Jul 2022 - Aug 2024

  Client : Foyer Group Health (Jan'23 - March'24)Responsibility –• Improved the existing structure of pipeline by adding features in pipeline such as validation and profiling todetect anomaly at early stage.•Validation consists of table and column level analysis. On the other hand, profiling involves comparing currentdata to previous run to check if anomalies present in latest data.•Utilized Object-Oriented programming(OOPs) principal to create data pipeline that follows ETL processesadherence to BDCR(Build, Detect, Correct, Repair) principles.•Creating pipeline run and step information artifacts for end user to determine whether pipeline run issuccessful or not. Automated pipeline execution with Apache Airflow, replacing manual triggers withscheduled runs.•Build test cases to determine the functionality of pipeline and ensuring its reliability.Tech Stack – Git, Linux, Python, Pandas, PyTest, Apache AirflowClient : SocGen (Apr'24 - Present)Responsibility –•Successfully enhanced the performance of data processing pipeline handling 225GB of data.•Utilized Polars to streamline the workflow and significantly reduce processing time from 10 hours to 2.5hours.•Implemented robust handling and logging mechanism to ensure data integrity and facilitate troubleshootingduring the data processing.•Conducted thorough performace testing and benchmarking to identify bottlenecks and optimize the efficiencyof the pipeline further.•Collaborated with cross-functional teams, including data scientists and analysts, to integrate the optimizedpipeline into the border data analytics framework, ensuring seamless operation and usability.Tech Stack – Git, Polars, Python Show less Responsibility –•Perform EDA on semi structured raw data in JSON format using pandas to get the insights and relevantinformation from data.•Interact with client to discuss and propose our understanding of data. Incorporated essential checks such asdata quality score, logging of pipeline runs and step information to enhance efficiency of data pipeline.•Implemented Polars module to optimize the data processing pipeline, resulting in significant reduction inpipeline run time.•Utilizing Git for version control to manage code changes and facilitate collabortion amoung team members.•Employing Linus as the operating environment for running and automating the data processing tasks.Tech Stack : Git, Linux, Python, Pandas, Polars Show less

  • Data Analyst

    Jan 2023 - Aug 2024

  • Data Analyst Intern

    Jul 2022 - Dec 2022
• 

  Fidelity International

  Aug 2024 - now

  Database Developer - Data Analyst