About KRAFTON AI Research Division
KRAFTON AI Research Division is building a next-generation AI-based game production paradigm centered around large language models (LLM). We manage the entire process, including model design, data structuring, large-scale training, optimization, evaluation, deployment, and operation, aiming to enhance game production efficiency and create new play experiences. Our direction broadly includes:
- Increasing game production efficiency and universality through deep learning technology development.
- Developing Co-Playable Characters that can play games with users.
About the Simulation AI Team
The Simulation AI Team applies LLM/SLM-based agents to KRAFTON's internal games, constructing a simulation environment where virtual characters autonomously think and act within game worlds.
- Designs Agent architectures where virtual characters (NPCs) autonomously perceive situations and behave naturally based on emotions, relationships, and goals.
- Develops Agent systems that assist in generating and modifying game content, allowing players to directly expand and transform the game world through natural language.
- Collaborates on inference optimization, cost efficiency, and building stable serving pipelines for efficient LLM inference operation in large-scale simulation environments.
- Proactively leads the entire process from research and prototyping to actual game application and operation.
Responsibilities
- Research and implement LLM-based agent models for autonomous behavior and interaction of virtual characters within games.
- Research and implement agent models/harnesses for game content generation and expansion.
- Optimize inference/serving and verify performance, considering real-time constraints and large-scale operating environments.
Requirements (Mandatory)
- Currently pursuing a Master's or Ph.D. degree in Artificial Intelligence (AI), Computer Science, or a related technical field.
- Ability to quickly understand LLM/deep learning related papers and translate hypotheses into experiments.
- Ability to adapt quickly to new domains.
- Excellent communication skills for smooth collaboration.
Preferred Qualifications
- Experience in writing papers for top-tier AI/ML conferences (NeurIPS, ICLR, ICML, ACL, AIIDE, etc.) in related fields.
- Research or development experience related to LLM Agent, Multi-agent System, or Tool-use.
- Experience with LLM inference optimization (serving, quantization, KV cache, etc.).
- Experience with or strong interest in applying AI in game/simulation environments.
- Excellent communication skills for interacting with other professions (game planning, art, business, etc.).