Research Scientist Intern - Video Generation

HoYoverse is committed to providing immersive virtual world experiences to players around the world. We have brought fans popular games including Genshin Impact, Honkai: Star Rail, Honkai Impact 3rd, Tears of Themis, and Zenless Zone Zero, as well as a wide range of entertainment content.
Community is at the heart of everything we do. We are devoted to engaging fans and fostering an enthusiastic and inclusive global community.
Pushing the boundaries of imagination, we consistently explore cutting-edge game development technologies.
In the future, we will continue to expand our content production, technology research, and publishing duties through operations in offices in Singapore, Montreal, Los Angeles, Tokyo, Seoul, and other areas.
When you apply to a position with HoYoverse, we will process your personal data. To learn more, please review our comprehensive Global Applicant and Candidate Privacy Policy.

What You Will Do

• As a Video Generation Research Intern, you will work closely with our research and engineering teams to advance post-training techniques for large-scale video generation models, with a particular focus on controllability, generation quality, and temporal consistency.

• You will explore cutting-edge generative modeling techniques and contribute to the development of next-generation AI-powered content creation and interactive virtual world experiences.

You Will

• Research, develop, and implement post-training methods for video generation models, including supervised fine-tuning, preference optimization, and reinforcement learning-based approaches.

• Explore controllable video generation techniques to improve motion control, camera movement, trajectory following, and action-conditioned generation.

• Design and conduct experiments to improve video generation quality, temporal consistency, motion realism, and alignment with user instructions.

• Collaborate with researchers and engineers to develop training pipelines, evaluation benchmarks, and scalable experimentation workflows.

• Stay up to date with the latest advances in video generation and generative AI, and contribute to research publications and practical applications.

What We Are Looking For

• Currently pursuing a Master's or PhD degree in Computer Science, Artificial Intelligence, Computer Vision, Machine Learning, or a related technical field.

• Strong research background and hands-on experience in video generation, including training or fine-tuning video generative models.

• Solid understanding of modern generative modeling techniques, such as diffusion models, flow matching, Diffusion Transformers (DiTs), and autoregressive models.

• Strong proficiency in Python and deep learning frameworks such as PyTorch.

• Ability to independently implement research ideas, conduct experiments, analyze results, and collaborate effectively with a research team.

Pluses

• Publications at top-tier AI and computer vision conferences, such as CVPR, ICCV, ECCV, ICML, ICLR, or NeurIPS. First-author publications related to video generation, especially controllable video generation, are strongly preferred.

• Research experience in controllable video generation, including camera control, motion control, trajectory-guided generation, action-conditioned generation, or interactive video generation.

• Experience with large-scale video model training, distributed training, or training and inference optimization.

• Contributions to influential open-source video generation projects or research implementations.

Compensation

• Hourly pay ranges from $50 - $80 USD (Mountain View only). Actual pay is based on location and may vary depending on relevant knowledge, skills, and experience from the candidate.

We are an equal opportunity employer that believes diverse backgrounds are key to bringing our concepts to life. If you're looking to play a key role in creating the best immersive virtual world experience for our users, we invite you to join our team.