Dr. Sangdon Park
LLM Application Research | Optimization & Game Theory | Game Design & Prototyping
Current) Sayberry Games Senior Researcher (May 2, 2025 ~)
LLM Application Research | Optimization & Game Theory | Game Design & Prototyping
Current) Sayberry Games Senior Researcher (May 2, 2025 ~)
Over the past few months, I have intensively researched the potential applications of AI Large Language Models (LLMs). Utilizing cutting-edge models like ChatGPT, Claude, and Gemini, I systematically analyzed their practical utility and limitations across the entire process of research, planning, and development.
As a result of this intensive research, I successfully implemented the Edge Computing GUI Simulator, a project I had envisioned for the past four years, in just one month. Additionally, the AI Character Conversation Project, initially started out of personal motivation, yielded results beyond expectations. This project enabled the implementation of natural character interactions beyond simple dialogue, paving the way for exploring new application areas like game development.
By effectively utilizing AI coding tools (like Cursor), I significantly enhanced development productivity and could rapidly implement ideas into prototypes. Based on this experience, I joined Sayberry Games on May 2, 2025, to participate in the research and development of AI-based interactive experiences. Furthermore, I successfully proposed a 2025 8-month, ~$220K (300M KRW) AI game development project.
I wish to share these practical experiences and expertise with the broader community. (Please refer to the AI Seminars section below for more details.)
Designing and developing service architectures that utilize Large Language Models as their core. Optimizing LLM performance through prompt engineering, fine-tuning, RAG, etc.
Developing AI characters and conversational systems using LLMs for natural and meaningful user interaction. Applications in various fields like gaming and education.
Research on applying AI/LLM technologies for simulation and resource optimization in complex systems, such as edge computing environments.
Researching and applying methods to enhance development process efficiency and code quality using AI coding tools (e.g., Cursor).
Based on the experience gained from deeply applying LLMs and AI tools in real-world settings for over 8 months, I conduct various talks and seminars. The content I share is not just theory but practical methodologies and know-how accumulated through direct engagement, validation, and hands-on experience.
Topics range broadly from AI utilization methods that maximize research paper writing efficiency to technical challenges and solutions encountered while integrating and optimizing LLM APIs into actual services. In particular, methodologies for endowing AI characters with human-like personality and consistency using RAG technology have received significant positive feedback from professionals in the gaming and content industries.
Furthermore, I share practical approaches for effectively utilizing LLMs in everyday tasks (e.g., apartment representative duties) and legal-related areas such as contract review. I also provide in-depth comparative analyses on strategically selecting various AI tools like ChatGPT, Claude, and Gemini based on different situations, and how to construct optimal prompts to maximize the strengths of each tool.
Compared to the initial seminar materials (View 8-month-old version), the current content is much deeper and more systematic. Based on insights gained through intensive research and development experience, the talks are structured around the latest technology trends and practical application cases.
These talks are offered in a customized format for diverse audiences ranging from individual users to corporations and research institutions. The focus is on delivering practical knowledge and skills applicable in the field immediately, rather than academic theory, aiming to help attendees gain a competitive edge in the AI era.
Key Talk Topics:
May 2, 2025 - Present
Sep 2017 - Apr 2025
KAIST, Information and Electronics Research Institute
Mar 2013 - Aug 2017
KAIST
Dissertation: Dynamic Energy Trading Schemes for Future Smart Grids
Mar 2011 - Feb 2013
KAIST
Thesis: Throughput Performance Analysis of Optimal Random Access Policy for Cognitive Radio Networks
Mar 2006 - Feb 2011
KAIST