Research Scientist – Generative Modeling (Diffusion)

About the role

World Labs is seeking a Research Scientist with deep expertise in generative modeling, especially diffusion models, to advance its 3D world‑building technology. The role blends hands‑on research, engineering, and product collaboration to push the boundaries of visual AI.

Key responsibilities

• Design, implement, and train large‑scale diffusion models for generating 3D environments.
• Develop novel control signals, adapt models to aesthetic preferences, and distill models for efficient inference.
• Collaborate with research, engineering, and product teams to translate requirements into technical roadmaps.
• Lead data curation, experimentation, evaluation, and deployment across the model lifecycle.
• Stay current with cutting‑edge diffusion and generative AI research and integrate findings.
• Mentor colleagues and promote best practices in generative modeling and ML engineering.

Required profile

• 3+ years of experience in generative modeling or applied machine‑learning roles, preferably in fast‑paced research environments.
• Strong publication record or open‑source contributions in generative AI.
• Deep expertise in at least one area of diffusion modeling such as pre‑training, post‑training, distillation, or conditioning.

Required skills

• Generative modeling
• Diffusion models
• Large‑scale model training
• Data curation for pre‑training or post‑training
• Tokenizers and VAEs for image, video, or 3D data
• Long‑context architectures
• 3D vision
• PyTorch
• TensorFlow