Oct 2023 – Apr 2025 · 1 yr 6 mos · Chicago, Illinois, United States · Remote

• LLM-based SQL code Review System:
• Implemented an LLM-based SQL code review system using Ollama, enabling automated analysis of SQL scripts to identify optimization opportunities, improve query performance, and enhance code quality. The model provided detailed feedback on indexing strategies, query structure, and best practices for maintainability and scalability.
• Data Integration in Palantir Foundry:
• Integrated structured and unstructured data from AWS and other platforms using Palantir Foundry’s Data Connection Service, enabling seamless data accessibility and integration efficiency. Designed scalable pipelines leveraging Pipeline Builder to ingest data from relational databases, APIs, flat files, and streaming sources, and automated workflows with Health Check Workflows and Data Quality Metrics, reducing data inconsistencies by 25%.

Mar 2018 – Aug 2021 · 3 yrs 5 mos · Bengaluru, Karnataka, India · On-site

• Enhancing Data Migration:
• Successfully orchestrated large-scale data migrations encompassing seamless migration of 800 Terabytes of data from on-premise Hadoop infrastructure to AWS platform addressing performance bottlenecks in legacy Hive/Impala scripts. Orchestrated the deployment of PySpark scripts for data processing, harnessing the power of cost-effective EMR clusters and AWS Glue, resulting in a 23% improvement in processing efficiency compared to legacy Hive/Impala scripts.
• Customer Lifetime Value (CLV) Prediction for Targeted Marketing:
• Developed CLV models utilizing customer transaction data and predictive analytics techniques such as RFM (Recency, Frequency, Monetary) analysis and Customer Segmentation. Predicted future CLV with 85% accuracy, enabling the marketing team to tailor campaigns and offers to high-value customers. This led to a 25% increase in customer retention and a 30% boost in cross-selling and upselling opportunities.
• Monkey Typed (Randomly Typed) Addresses Classification:
• Developed Classification Models using Natural Language Processing (NLP), TFIDF, N-grams and BERT to segregate Monkey Typed Addresses that decreases the monthly delivery failure rates by 6%.
• Customer Segmentation & tailored strategies:
• Coached and managed a team of Interns on a cross-functional project with the Marketing team to create customer segmentations based on demographic information, purchase history, browsing behavior and preferences using k-means clustering and random forest algorithm to provide insights on how to tailor marketing efforts for each segment to increase annual customer engagement by 9%.

Sep 2015 – Mar 2018 · 2 yrs 6 mos · Bengaluru, Karnataka, India · On-site

• Optimized sales performance by partnering with the sales department by analyzing sales data leveraging Tableau and Power BI. unveiling valuable patterns, trends, and growth opportunities. Offered actionable insights into optimal sales strategies, target customer segments, and product positioning. This resulted in a 12% surge in sales revenue and guided the sales team in achieving their targets.