Feb 2022 – Jan 2024 · 1 yr 11 mos · San Francisco Bay Area

• Implemented a data ingestion framework using Java Spark on EMR serverless. This system manages data ingestion from over 800 sources, including multi billion row relational tables and Kafka topics. Critical in providing data to the Data Lake and Redshift for our ML, platform, and analytics teams.
• Led a migration project, creating data pipelines for more than 7 core AncestryDNA and PetDNA databases. The approach leveraged Airflow, Spark, EMR, and Glue to streamline data loading into the data warehouse, greatly enhancing support for our analytics teams.
• Delivered various data products and scalable APIs.
• Key role in designing and developing essential components for the Ancestry Data Lake that facilitated ML, Data Science workflows, and analytics for the company.

Jan 2021 – Feb 2022 · 1 yr 1 mo · San Francisco Bay Area

• Engineered services within the UBE infrastructure, facilitating real-time processing of over 100k Ancestry business events per minute. Streamlined data ingestion into data warehouse and analytics platforms, leveraging a custom Kafka schema registry while achieving 40% performance improvement through tuning and load testing.
• Enhanced verifi-gateway service with critical functionalities such as JWT authorization and Redis rate limiting, contributing to annual savings exceeding $1 million in fraud detection and dispute prevention.

Jun 2020 – Sep 2020 · 3 mos · Utah, United States

• Built essential Java functionalities for Ancestry API Gateway and AncestryDNA.
• Constructed external partner APIs and custom route serialization within the platform SDK/gateway leveraging Java, ReactiveX, and the Netflix Zuul, facilitating over 25 development teams.
• Introduced AWS Lambda to core platforms to support low-traffic services.
• Implemented Circuit Breaker using Resilience4J to ensure fault tolerance when handling 110,000 requests per minute from the API Gateway, improving system reliability and performance.

Sep 2017 – Jul 2019 · 1 yr 10 mos · Pune District, Maharashtra, India

• End-to-end design and development of the Phoenix Life Insurance, UK web application as a full-stack developer, developing Java microservices and libraries using Spring Boot and Hibernate, facilitating 12M+ insurance policies.
• Spearheaded a successful integration of Drools rule engine to Java Spring modules with JBoss Kie, configured Headless Magnolia CMS v5.7, and constructed reusable/responsive UI using Angular and TypeScript. Delivered 150+ enhancements, issue resolutions, and change requests to improve back-end performance and UI.

Jul 2016 – Aug 2016 · 1 mo · India

• Led development for service request management app with CodeIgniter (MVC) + MySQL supporting 4 user roles, built reusable modules reducing development effort by 30% and delivered responsive dynamic UI with Bootstrap/JS cutting page load times by 20%.