TikTok
Research Engineer/Scientist (all levels), World Models
San Jose, California · Full-time · Posted Oct 28, 2026
- Experience
- Extensive research experience; no number of years specified.
- Education
- M.S. or Ph.D. in Computer Vision, Computer Graphics, or Machine Learning, or equivalent experience.
About the job
Develops research methods and infrastructure for large-scale, interactive world models, with a focus on long-horizon video generation and realistic physical dynamics.
What you'll do
- Develop large-scale, diverse, interactive multimodal data-generation pipelines.
- Develop training pipelines for long-context interactive video-generation models.
- Advance video-generation models to capture temporal consistency, physical dynamics, object interactions, and causal relationships.
What they ask for
- M.S. or Ph.D. in Computer Vision, Computer Graphics, or Machine Learning, or equivalent experience.
- Extensive research experience in GenAI, multimodal foundation models, or Embodied AI.
- Ability to communicate complex technical concepts and collaborate effectively within cross-functional research teams.
Nice to have
- Experience in video generation and synthesis, efficient and real-time diffusion models, 3D/physics-based simulation, or reinforcement learning for agentic environment interaction.
- First-author publications in venues including CVPR, ICLR, NeurIPS, SIGGRAPH, or ICML.
Tools
GenAI, multimodal foundation models, video generation, diffusion models, 3D/physics-based simulation, reinforcement learning
AI focus
Builds large-scale generative world models and interactive video-generation systems, emphasizing temporal consistency, realistic physics, object interactions, and causal relationships.
Model-written summary · the original posting is authoritative