This thesis explored the use-cases of Large Language Models for their effectiveness for Text Simplification, focusing on readability and comprehension for English as a second language users. Since complex texts create difficulties for readers. This work aims to simplify complex languages while keeping semantic integrity intact. The thesis reviews the theoretical foundations of text simplifications, exploring lexical and syntactic simplification techniques, and evaluates transformer-based models. Such as BART, T5, and Qwen. The thesis has a three-step methodology, with an initial minimum viable product benchmarking multiple pre-trained models, a second phase using parameter-efficient fine-tuning, and the utilization of low-rank adaptation and quantization. Lastly, a third stage that uses user-centered feedback with L2 readers. Evaluation metrics such as SARI, BLEU, and FKGL, combined with empirical strategies. Findings suggested that the fine-tuned Qwen model significantly improved the readability without distorting meaning. The results emphasize the importance of user involvement in Text Simplification system development and highlight potential applications in educational and accessibility technologies. This thesis contributes a replicable framework for building adaptable, efficient, and user-focused text simplification using Large Language Models.