Apple

• Developed and maintained aggregated ETL pipelines using Spark SQL and PySpark on Hadoop file systems as part of Apple's Health Software team.
• Build an end-to-end data application that would allow QA/QE teams to validate the user data or detect anomalies in the user data streams uploaded by the research health app.
• Processed 10 TB of large data in a distributed parallel computing architecture and extracted meaningful features from it to promote data analytics.
• Responsible for scheduling, maintaining, and orchestration of data pipeline using Apache Airflow.
• Skills: Data Modeling, Data Analytics & Visualization, Workflow orchestration, and ETL processes

Amazon

• Implemented static code analysis recommendation rules in Java for python files to improve code quality, application performance and detect code smells in repositories. The detection percentage of the python rules was 78 % in the amazon GitHub repository.
• Incorporated unit test in JUnit cases to validate python rules.
• Led junior software engineering interns and collaborated with the Amazon Codeguru Seattle team for proper implementation and testing of python rules.

Persistent Systems

• Implemented PII Data Masking and Obfuscation Pipeline using Python on client databases. Optimized data masking pipeline performance by 50% and obfuscated 450 GB of large data.
• Ingested data from various data clusters using a combination of AWS Athena and AWS Glue to create data views to be used in BI tools.
• Worked closely with ML solution architects for proper implementation of data-preprocessing pipeline and to derive insights from the data which could be further useful for the ML models.
• Responsible for the creation of various dashboards on Tableau for data analytics.

Persistent Systems

• Responsible for deploying Facebook’s Deep Learning NMT Transformer model on torch-serve (open-source tool for serving PyTorch models) and created generalized custom handlers for all transformer models enabling fast and easy deployment of the model.
• Responsible for deployment and testing of various Pytorch models on container technologies like Docker & Kubernetes.
• Acted as a liaison between teams in Amazon-Sagemaker and Facebook ML teams for the proper deployment of the transformer model.

Persistent Systems

• Implemented several time-critical features while migrating IBM users from mainframe-based to a distributed computing architecture.
• Consistently deliver structured and maintainable front-end codes to build an intuitive user interface and reusable components on the React.js.
• Implemented REST services by Golang with microservices architecture.
• Excellent knowledge of NO-SQL on Mongo and Cassandra DB and expertise in working with SpringlOC, SpringMVC, and SpringAOP.
• Contributed to developing an event-driven and time-triggered report generation module using Java and Spring.
• Leveraged Apache Spark for processing and aggregating high-volume historical data for generating reports.
• Interacted with clients across various global locations like the USA, Canada, China, Australia.
• Leveraged knowledge in Java, J2EE, Spring-MVC, Spring Boot, Golang, JavaScript, HTML, SQL, DOJO, Node.js, React.js.

Languages and Databases

Python

Java

C/C++

JavaScript

Shell Scripting

MySQL

PostgreSQL

MongoDB

Libraries

Pytorch

Tensorflow

Keras

NumPy

Pandas

OpenCV

scikit-learn

matplotlib

Frameworks

Django

Flask

Other

Git

AWS