Shreyans Borad

Enhanced scalability and experiment quality on the Catalog Recommendation page by migrating the existing architecture to a Directed Acyclic Graph (DAG) model, allowing product team to launch 10x of experiments

• Migrated core services from Java to Go-lang, achieving a significant cost reduction of 50-70% for multiple services while improving overall system efficiency
• Led the transition from pre-computed recommendation sets to real-time calculation, leveraging advanced ML models designed by the data science team
• This initiative resulted in better product recommendations and a 1.3% increase in company net profit
• Maintained 99.9% uptime during Meesho's Maha Blockbuster sale in the Diwali season, ensuring uninterrupted service during high-traffic periods
• Built a real-time streaming pipeline using Apache Flink to provide near real-time catalog updates, enabling backend systems to operate on the most up-to-date data for better decision-making

• Developed a staging framework that significantly optimized file formation processes for staging connectors like GCS and Snowflake within Informatica, resulting in a 2-3x performance boost
• Implemented bulk API support over the JNI network to enhance data transfer efficiency between C++ and Java applications
• Optimized performance by 10x through Push Down optimization, leveraging Depth First Search (DFS) to generate multiple database queries after traversing the Informatica Logic Workflow
• Created an automated tool integrated with Perforce, accelerating the customer support process by automating bug fix releases for specific versions, enhancing efficiency and reducing turnaround time