Video Generation Content Understanding and Feedback Researcher

Lightspeed Tech Center, an R&D department of LIGHTSPEED STUDIOS, seeks a researcher to advance generated-video understanding and feedback for interactive game experiences. The role is based on-site at Singapore-CapitaSky and is full-time.

Responsibilities

• Research how models interpret generated video, including semantics, objects, relationships and spatial structures; implement state tracking and consistency modeling.
• Explore unified generation-understanding architectures, causal consistency and physical constraints in an “action input → video output” pipeline. Research hybrid architectures that provide low-latency feedback for real-time interactive generation.
• Collaborate with Agent and reinforcement learning teams on a closed loop of generation, understanding, control and feedback. Validate controllable interactive capabilities in game scenarios, establish evaluation systems, track relevant industry research and develop technical roadmaps.

Required qualifications

• Ph.D. in an AI-related field such as video understanding, video prediction, reinforcement learning, or multimodal generation and understanding.
• Deep understanding of video-generation and vision-language model mechanisms, familiarity with diffusion models, and research experience in interactive or controllable video generation.
• Ability to design complex model-training pipelines independently, strong engineering skills, proficiency in Python and PyTorch, and publications in video understanding, controllable generation or reinforcement learning.

Preferred qualifications

Experience with Game AI, simulation environments or closed-loop systems, and relevant publications in top-tier conferences or journals.