Controllable Video Generation Researcher

Tencent; Greater London; On-site; GBP 90,000 - 140,000; Full time

Job summary

Lightspeed Studios in London seeks a PhD-level AI researcher to design a multimodal controllable generation framework for gaming, supporting keyboard, controller, trajectory, and voice inputs.

You will study high‑fidelity generation with sparse controls, decouple trajectory, camera, motion and posture control, and create modular methods for different game genres, collaborating with Engine/Agent teams to standardize interfaces.

Qualifications

• PhD in AI with focus on CV/DL, strong research track record.
• Proficient in conditional generation and multi-player control tech.
• Experience with diffusion models and real-time video generation.
• Projects in camera/trajectory/posture control and game AI a plus.
• Solid understanding of game engine I/O and frame synchronization.

Responsibilities

• Design a multimodal controllable generation framework for gaming inputs.
• Research high-fidelity generation under sparse control signals.
• Decouple and combine trajectory, camera, motion, and posture control.
• Develop modular control methods for different game genres.
• Build video generation capabilities for multi-player interactive control.
• Collaborate with Engine/Agent teams to define standardized interfaces.
• Advance joint optimization of controllable generation and RL post-training.

Skills

Python

PyTorch

Diffusion models

Conditional generation

Game AI

Education

Ph.D. in AI (CV/DL)

Tools

PyTorch

Job description

Business Unit

LIGHTSPEED STUDIOS is made up of passionate players who advance the art & science of game development through great stories, great gameplay, and advanced technology. We are focused on bringing next generation experiences to gamers who want to enjoy them anywhere, anytime, across multiple genres and devices.

About the Hiring Team

Lightspeed Tech Center is a R&D department under Lightspeed Studios which develop PUBG Mobile and other high-quality games. Our Tech Center leads the research, exploration, and discovery of innovative technologies and provides technical services for all games during all phases of life cycle, including engine, audio, QA, AI, next generation game, technical cooperation, etc.

What the Role Entails

1. Design a multimodal controllable generation framework supporting various inputs such as keyboard, controller, trajectory, and voice.
2. Research high-fidelity generation under sparse control signals and explore the extension of conditional control paradigms in the video domain.
3. Achieve the decoupling and combination of trajectory control, camera control, motion control, and posture control.
4. Design dedicated control modules tailored to different game genres.
5. Develop video generation capabilities for multi-player collaborative control, ensuring low-latency response and inter-frame state retention.
6. Collaborate with the Engine/Agent teams to define standardized interfaces for controllable signals.
7. Drive the joint optimization of controllable generation and Reinforcement Learning post-training.

Who We Look For

1. Ph.D. in AI-related fields, focusing on Computer Vision/Deep Learning.
2. Proficient in conditional generation models and multi-player/multi-channel control technologies; familiar with the principles of diffusion models.
3. Project experience in camera control, trajectory control, or posture control.
4. Understanding of game engine input/output logic and frame synchronization mechanisms.
5. Proficient in Python/PyTorch, with publications in controllable video generation or conditional generation.
6. Experience in Game AI or game development is preferred.
7. Experience in real-time/interactive video generation projects is preferred.

Equal Employment Opportunity at Tencent

As an equal opportunity employer, we firmly believe that diverse voices fuel our innovation and allow us to better serve our users and the community. We foster an environment where every employee of Tencent feels supported and inspired to achieve individual and common goals.