Pradumnya Borah

Actively involved in creating data regression models, using predictive data modeling algorithms to deliver insights and implement action-oriented solutions to complex business problems. Performed EDA with Feature engineering to analyze distribution, find correlations and gain understanding of data. Worked on intelligent routing/lead scoring for real estate industry and significantly improved the F1-score by 6%; minimizing the agents involvement time upto 40% and increasing the leads engagement by more than 50%. Initiated the development of an Email spam detection tool using Multinomial Naive Bayes, applying NLP techniques and using TF-IDF. Successfully decreased spam percentage from 11% to 5% with precision-recall tradeoff. Handled imbalanced dataset using under-sampling approach to improve ROC-AUC. Used hypothesis tests (A/B testing), provided business insights to the company's decision making process in area such as customer interaction and the cost of acquiring customers. Participated in complete software development lifecycle, including requirement planning, design, development, testing, deployment and production releases. Also, created and consumed RESTful APIs for software integration.

Collaborated with their employees to develop a warehouse management system and gain knowledge on software industry environment.

State highest marks in statistics 97%