Vyshakh Prasad is pursuing his M.S. in Computer Science at the University of the Pacific, following his B.S.-M.S. Dual Degree in Physics from IIT Kanpur. His research focuses on Machine Learning and Natural Language Processing (NLP) for domain-specific applications. His work includes fine-tuning the Qwen2-1.5B model using the GRPO framework and a custom reward function for legal text summarization, as presented in this paper. He leverages deep learning, reinforcement learning, and efficiency techniques like LoRA to tackle complex computational challenges.