Rohit Kumar

Synechron Technology (Client: Morgan Stanley) | Mumbai, India | Jan 2023 – Present

Project: Portfolio Data Pipeline – End-to-End Ingestion, Validation, and Processing of Trading Files

Tech Stack: Apache Spark, AWS S3, Python, SQL, Erwin Data Modeler

• Designed and implemented scalable data pipelines to ingest and validate daily trading files with portfolio details from Amazon S3, ensuring accurate and timely delivery to downstream systems.
• Automated the complete pipeline workflow — from data ingestion to database insertion — using Apache Spark and custom orchestration, creating a fully hands-free processing environment.
• Developed a robust data validation framework that detects and isolates corrupt or inconsistent records, routing them to an Errors folder and ensuring only clean data is loaded into the Datamart.
• Worked closely with stakeholders from Trading, Portfolio Management, and Risk teams to refine business rules and evolve validation logic in line with changing requirements.
• Optimized Spark transformations for large-scale portfolio datasets to improve performance and minimize runtime in a resource-efficient manner.
• Designed and delivered a configurable Data Platform supporting both batch and real-time data processing, allowing engineers to define new pipelines via a simple configuration file.
• Built scalable data models using Erwin Data Modeler, enabling robust reporting and deep analytics for the Data Science and BI teams.
• Established an iterative correction workflow, automatically rerouting invalid records to source teams for review and reprocessing — maintaining long-term data quality and compliance.

Capgemini (Client: Baker Hughes) | Bengaluru, India

Project: GlueDynamo ETL Pipeline – Transforming Data with AWS Glue and Storing in DynamoDB

Tech Stack: AWS Glue, AWS Lambda, AWS DynamoDB, Python, PySpark, SQL Server

• Designed and developed a serverless ETL pipeline to trigger a SQL Server stored procedure via API call, fetch the processed data, and store it in Amazon DynamoDB using AWS Glue and Python.
• Configured AWS Glue jobs to securely connect with SQL Server and execute stored procedures for dynamic data extraction.
• Wrote custom Python scripts to handle stored procedure execution and parse result sets for further transformation.
• Leveraged Boto3 (AWS SDK for Python) to programmatically write the transformed data into a target DynamoDB table.
• Integrated the pipeline with AWS Lambda to trigger Glue execution upon receiving API requests, enabling event-driven data processing.
• Ensured the pipeline was scalable, fault-tolerant, and aligned with security best practices for data access and logging.
• Evaluated customer needs and feedback to drive product and service improvements.

• Developed and maintained backend systems for an e-commerce platform specializing in laptops and Huawei electronic equipment.
• Designed and implemented data models and core business logic to support product listings, inventory management, and order processing.
• Collaborated on the client-server architecture, ensuring efficient communication between frontend interfaces and backend services.
• Contributed to building a scalable and reliable platform tailored to online retail needs in the electronics domain.
• Built databases and table structures for web applications.
• Conducted data modeling, performance and integration testing.