Pinky Kapoor

Pinky Kapoor is a Data Scientist at Verisk, involved in projects focusing on fraud prediction and bias mitigation in insurance claims data, with expertise in analyzing PDF data for authenticity verification. Previously, Pinky served as a Data Science Intern at Verisk, optimizing code with PyTorch Lightning and utilizing the PDF Arlington model for feature extraction. As a Graduate Research Assistant at The University of Texas at Dallas, Pinky contributed to multi-agent reinforcement learning research and developed a CNN model achieving 94% accuracy in handwritten text recognition. Additionally, Pinky has experience coordinating technical events and improving campus connectivity as a Technical Secretary at the Indian Institute of Technology, Patna, and has served as a Teaching Assistant in various engineering courses. Educational qualifications include a Ph.D. in Reinforcement Learning and Machine Learning from The University of Texas at Dallas and an M.Tech in Communication System Engineering from the Indian Institute of Technology, Patna.