Legal document analysis aims to leverage the capabilities of large language models in a Python-based environment to work towards automating legal document review, information retrieval processes, and to improve the efficiency of legal research. The value and importance of this thesis lie in its contribution to the development of natural language processing tools tailored for the legal domain. These advancements have the potential to benefit the legal industry by reducing the time and effort required for comprehensive document reviews. Additionally, they can enhance the accuracy and reliability of legal research. In this research, we focus on fine-tuning pre-trained LLMs to achieve three key task outcomes: Named entity recognition, Sentiment analysis, and Summarisation. The resultant models' performance measures up acceptably compared to state-of-the-art NLP models and are ready for research and production purposes.