From Biology to Game AI: An Unconventional Journey

I am a Master student at the University of Chinese Academy of Sciences (UCAS), currently focusing on game AI research. My academic journey has been anything but conventional - it’s a story of following one’s passion and transforming competitive gaming experience into artificial intelligence research.

Early Years and Gaming

During my undergraduate years (2016-2020), while officially studying Biotechnology, my heart was elsewhere. I spent countless hours mastering StarCraft II, achieving Grandmaster rank on the Korean server and competing in various professional tournaments. My initial dream was to become a professional StarCraft II player, and though I reached significant competitive achievements (including top placements in WESG China Regional Finals and NSL), I eventually found my true calling at the intersection of competitive gaming and artificial intelligence.

Current Research

My research journey in game AI began with a simple question: Could we create AI systems that play games not just effectively, but in ways that reflect human-like strategic thinking? This quest has led me to work on various exciting projects:

• StarCraft II AI: Developing large language model-based agents capable of complex strategic decision-making
• CSGO AI: Creating vision-based AI systems using imitation learning
• AI Town: Exploring social interaction and decision-making in multi-agent environments

The combination of my competitive gaming background and technical expertise has given me a unique perspective in this field. I understand both the competitive player’s mindset and the technical challenges of implementing these strategies in AI systems.

Future Vision

My research goal is to develop human-like and human-friendly game AI systems. I believe the future of game AI lies not in creating unbeatable opponents, but in developing AI that can understand, adapt to, and interact meaningfully with human players. I’m particularly interested in:

• Developing world models for video game environments
• Creating AI systems that can learn and adapt to different play styles
• Building AI that enhances rather than replaces human gaming experiences

I’m excited about the potential of combining large language models, world models, and reinforcement learning to create more intuitive and engaging game AI systems. Through my research, I hope to bridge the gap between competitive gaming strategies and artificial intelligence development.

Future Plan

I am actively seeking opportunities to continue my research journey:

• PhD Positions: I am looking for PhD positions starting Fall 2025, with a focus on Game AI, Multi-agent Systems, and World Models
• Research Positions or intern: Open to research opportunities in Game AI and Machine Learning

My ideal research environment would allow me to:

• Pursue advanced research in human-like game AI systems
• Combine competitive gaming insights with cutting-edge AI techniques
• Collaborate with researchers who share my passion for creating more intuitive and engaging AI

Feel free to contact me if you have any opportunities or would like to discuss potential collaboration:

• Email: sc2meisah@gmail.com
• Google Scholar
• GitHub