Kartikey Shaurya

Contributed Significantly to the Data Science team, overseeing the entire Data Management, model development life-cycle and RESTful API development from design to production readiness. Designed and implemented data pipelines to enhance the efficiency of the Internal Data Lake(160 GB) by migrating and consolidating data, resulting in a 40-50% increase in data write speed, a 20% compression in data storage and a 10-20% reduction in response time. Developed a PII Detection model to facilitate anonymous data sharing. Developed and monitored a Regression (ML) Model to predict performance clusters for drivers, boosting driver income by 8-10% and decreasing attrition rates by nearly 15%. Developed a POC for an Llm-based chatbot designed to engage with job seekers. Established the data foundation for Samarth platform comprising 30 million job seekers. Introduced batch processing and integrated Elasticsearch to significantly enhance API responsiveness.

Worked on a React application capable of generating invoices based on real-time data and implemented regression techniques to identify potential payment delays. Built and enhanced the accuracy of the payment delay detection model by nearly 15 %. Additionally, introduced an ML pipeline to automate ML training and manage data drift effectively.