LLM-Based AI Companion for Multi-Agent Collaboration

This PhD opportunity at the University of Edinburgh's School of Informatics focuses on developing an LLM-based AI companion for multi-agent collaboration, with co-funding from Tencent Games, a global leader in game development and publishing. The project aims to advance the integration of large language models (LLMs) into multi-agent systems, particularly within the context of online gaming environments. Students will explore cutting-edge research in artificial intelligence, machine learning, software engineering, and human-computer interaction, with a special emphasis on collaborative intelligent agents and their application in videogames. The successful candidate will work closely with Professor A Storkey, leveraging the expertise and resources of both the University of Edinburgh and Tencent Games. Research topics may include natural language processing for agent communication, reinforcement learning for adaptive behaviors, and the design of AI companions that enhance user experience in multiplayer games. The project offers a unique opportunity to contribute to the future of AI-driven game development and multi-agent collaboration, with potential for real-world impact in the gaming industry. Applicants should possess a strong academic background in computer science, artificial intelligence, or software engineering, with demonstrated programming skills and familiarity with machine learning frameworks. Experience in game development or multi-agent systems is beneficial but not essential. International candidates are welcome and may need to provide proof of English language proficiency. While funding details are not specified, the position benefits from industry collaboration and access to state-of-the-art research facilities. The application deadline is 18 February 2026. Interested candidates should prepare a CV, academic transcripts, and a statement of research interest, and submit their application via the University of Edinburgh's online portal. For further information, refer to the project link or contact the School of Informatics.