Senior AI Researcher, World Foundation Models

What you'll do

• Research and implement model architecture and algorithm changes to improve video generation fidelity and human-centric quality.
• Prototype improvements in spatial multimodal modeling, modality alignment, and diffusion-based video generation for better controllability.
• Optimize training and inference efficiency through architectural techniques like distillation, pruning, and compression.
• Define training objectives to improve sim-to-real generalization for human motion, contact, and interaction dynamics.
• Develop domain-specific benchmarks to evaluate world models reasoning about video, simulation, and physical environments.
• Translate research results into production-grade checkpoints, training code, and demos that showcase capability gains.
• Diagnose visual artifacts using perceptual metrics and human preference signals to ensure high-quality output.

What we're looking for

• PhD in Computer Science, Graphics, Computer Engineering, or a closely related field (or equivalent experience).
• 8+ years of applied research and/or industry experience in vision, graphics, or adjacent ML domains.
• 3+ years of direct experience designing, training, and evaluating generative models for image, video, or audio.
• Hands-on experience improving generative models with a focus on perceptual quality and temporal stability for human generation.
• Advanced proficiency in Python, PyTorch, C++, and CUDA with strong research-engineering practices.
• Experience training and debugging large models in multi-GPU/multi-node environments and distributed training workflows.
• Practical knowledge of inference/runtime bottlenecks and optimization techniques like distillation, pruning, and compression.
• Strong "eye for quality" to diagnose visual artifacts using perceptual metrics or human preference signals.