In my PhD research at GPT Lab, I combine multi-agent AI systems with large language models (LLMs) to innovate in software engineering. Beyond a professional pursuit, refining LLMs is also my hobby, aiming to secure their application in software data and automate development processes. My work streamlines code generation and data analysis with AI, evaluating LLM effectiveness in engineering tasks. This endeavor not only introduces new technologies but also aims to revolutionize software development by integrating LLMs and AI agents. My goal is to advance AI in software creation, contributing to academia and practical engineering, making development more efficient and innovative.