[AI Research Div.] Research Engineer - Vision Language Action Game Agent (2년 이상 / 인턴)
Core
Research and implement Vision-Language-Action (VLA) agent models for real-time game environments to create AI teammates that interact with players and provide guidance.
Role type
Research Engineer (Vision Language Action Game Agent)
Builds
Real-time AI agents that act as teammates in video games
Domain
Gaming + Multimodal AI (Vision-Language-Action)
Deliverable
production ML models
Required skills
Deep learning research, multimodal model design, real-time inference optimization, large-scale data pipeline construction, experimental design
Preferred skills
Top-tier conference publications (NeurIPS, ICLR, ICML), VLA model project experience, cross-functional communication
Technologies
LLM, SLM, VLM, Deep Learning frameworks
Responsibilities
Design instruction-following data and learning recipes, analyze failure cases, optimize model serving under real-time constraints, build and operate data collection/cleaning pipelines
Seniority
Researcher (Master's/PhD level)