Video Generation & Understanding Research Scientist

LightSpeed Studios; Singapore; On-site; SGD 150,000 - 210,000; Full time

Job summary

LightSpeed Studios' Lightspeed Tech Center seeks a PhD-level AI researcher to advance generation and understanding of video content. You will explore model architectures, track state, and collaborate with RL teams to build end-to-end generation–understanding loops for interactive games.

Candidates should have deep hands-on expertise in Python/PyTorch, publications in video understanding or RL, and a track record of designing complex training pipelines.

Qualifications

• Ph.D. in AI-related fields with focus on video understanding or multimodal generation.
• Strong Python/PyTorch skills and publications in relevant areas.
• Experience with interactive video generation or controllable generation is preferred.

Responsibilities

• Research model understanding for generated content and spatial relationships.
• Develop state tracking and consistency modeling for generated videos.
• Collaborate with RL teams to close the generation–understanding–control loop.
• Lead evaluation systems and monitor industry developments to shape roadmaps.

Skills

Python/PyTorch

Video understanding

Reinforcement learning

Research pipelines

Education

Ph.D. in AI-related fields

Tools

VLM models

Diffusion model principles

Job description

LightSpeed Studios' Lightspeed Tech Center seeks a PhD-level AI researcher to advance generation and understanding of video content. You will explore model architectures, track state, and collaborate with RL teams to build end-to-end generation–understanding loops for interactive games.

Candidates should have deep hands-on expertise in Python/PyTorch, publications in video understanding or RL, and a track record of designing complex training pipelines.