Job Description

About the Role

Cybever is an AI-powered 3D creation platform on a mission to revolutionize how videos are made. We harness cutting-edge AI to push the boundaries of 3D and video generation, enabling creators and businesses to generate high-quality content with unprecedented speed and consistency. As a Video Generation Researcher , you will join our core team to build and scale the next generation of AI-driven video generation systems. This is a full-time, engineering-focused role working on-site in Sunnyvale, CA, where you’ll collaborate closely with our founding team to turn breakthrough research into real-world products.

In this role, you will be at the nexus of research and engineering – developing novel video generation models and deploying them in production. We are particularly interested in candidates passionate about diffusion models and autoregressive approaches for video generation, and who love the challenge of optimizing these models for real-world performance. If you’re excited about creating the future of AI video (and perhaps fresh from CVPR with new ideas!), we want to talk to you.

What You’ll Do

• Build state-of-the-art video generation models: Develop and refine algorithms for AI-driven video generation, exploring both diffusion-based and autoregressive (e.g. transformer) approaches to achieve high-quality, coherent video outputs.
• Scale up real-world systems: Design and implement large-scale training pipelines for video models, leveraging distributed computing and big datasets to train models that are both cutting-edge and reliable in production.
• Optimize for consistency and control: Tackle core research challenges like temporal consistency, spatial coherence, and controllability in generated videos.
• Collaborate across teams: Work closely with our engineers and 3D content team to integrate new models into Cybever’s pipeline.
• Stay ahead of the curve: Keep up with the latest research in generative video (diffusion models, transformers, GANs, etc.), and prototype innovative ideas.
• Deliver impact: Continuously test and tune your models in our end-to-end system, pushing updates that immediately impact creators using our platform.

Who You Are

• Ph.D. in Computer Science or a related field (required).
• Strong publication record at top-tier conferences (e.g. CVPR, ICCV, SIGGRAPH, NeurIPS, ICLR).
• Strong software engineering skills and deep knowledge of modern deep learning.
• Hands-on experience with deep learning frameworks, especially PyTorch .
• Experience with generative models such as diffusion models and autoregressive transformers.
• Experience with training large-scale models, distributed systems, and optimizing training/inference performance.
• Clear communicator and effective collaborator in fast-moving, interdisciplinary teams.

Why Join Us

• Shape the Future of AI Video: At Cybever, you’ll be contributing to first-of-its-kind technology that reimagines video creation with AI-built 3D worlds.
• High Impact & Ownership: As an early member of our team, you will work side-by-side with the founders and have significant ownership of critical projects.
• Broad Industry Applications: The technology you build will impact a wide range of industries – from filmmaking and visual effects to gaming, robotics, and beyond.
• Growth and Learning: We foster a culture of continuous learning and encourage publication and open-source contributions.
• Competitive Benefits: We offer a competitive salary and full benefits to support your well-being and productivity.

If you’re excited by the prospect of building cutting-edge video generation systems that bridge research and real-world impact, we’d love to hear from you. Join us at Cybever and help create the future of cinematic AI video, one groundbreaking model at a time.

Job Tags

Similar Jobs