Research Engineer - Vision Language Action Game Agent

Responsibilities

• Research and implement VLA agent models that receive multimodal inputs (screen, text, game state, etc.) and output actions in a gaming environment
• Design data, establish training recipes, and analyze/improve failure cases to enhance instruction-following performance
• Optimize modeling, inference, and serving for real-time constraints and verify performance
• Build and operate large-scale training data collection and refinement pipelines

Requirements

• Currently pursuing a Master's or PhD in AI, Computer Science, or a related technical field
• Ability to lead research and quickly understand/design experiments through deep learning papers
• Ability to adapt quickly to new domains
• Strong communication skills for smooth collaboration

Preferred Qualifications

• Experience publishing papers in top-tier AI/ML conferences (NeurIPS, ICLR, ICML, etc.) or journals
• Proven project success related to Vision Language Action Models
• Excellent communication skills for collaborating with other disciplines such as game design, art, and business

About the Company

KRAFTON AI Research Division develops AI solutions and proprietary deep learning technologies to enhance game production efficiency and create diverse applications within and outside of games.