Video Generation Researcher: Understanding & Feedback

L&Q OASIS LIMITED; Greater London; On-site; GBP 75,000 - 110,000; Full time

Job summary

Tencent Lightspeed Studios' Tech Center is seeking a PhD-level researcher to advance AI-based video understanding and controllable generation for next-gen gaming experiences. You will work on diffusion-based and multimodal architectures, shaping research directions and collaboration with Agent/RL teams.

The role emphasizes independent design of complex training pipelines, strong Python/PyTorch skills, and a track record of publications in related fields.

Qualifications

• PhD in AI-related fields with strong publication record.
• Deep understanding of video generation and VLM models incl. diffusion principles.
• Experience with interactive/controllable video generation research.

Responsibilities

• Research model understanding of generated content, parsing semantics, objects, relationships, and spatial structures.
• Implement state tracking and consistency modeling for generated videos.
• Explore unified generation-understanding model architectures.
• Research causal consistency control and physical constraints for action→video pipelines.
• Develop low-latency, real-time interactive generation architectures.
• Collaborate with RL teams to close loop generation→understanding→control→feedback.
• Validate controllable interactive capabilities in game scenarios and shape evaluation systems.
• Keep abreast of industry work and help define technical roadmaps.

Skills

PhD in AI

Video understanding

Diffusion models

Interactive video generation

Python

PyTorch

Publications in AI/ML

Game AI / simulation

Education

PhD in AI/ML

Tools

Python

PyTorch

Job description

Tencent Lightspeed Studios' Tech Center is seeking a PhD-level researcher to advance AI-based video understanding and controllable generation for next-gen gaming experiences. You will work on diffusion-based and multimodal architectures, shaping research directions and collaboration with Agent/RL teams.

The role emphasizes independent design of complex training pipelines, strong Python/PyTorch skills, and a track record of publications in related fields.