Large language models have emerged as one of the most impactful applications in software engineering. However, their integration into the software development ecosystem remains fragmented and suboptimal. This study aims to bridge this gap by adapting and fine-tuning existing large language models for real-world software development scenarios. Rather than building new models from scratch, our approach leverages advanced techniques that allow Large Language Models to interact seamlessly with existing development tools and Integrated Development Environments (IDEs). This research, conducted in collaboration with JetBrains, focuses on personalizing code generation by adjusting the model to the unique features of a project and the individual coding style of each developer.