Deep Learning Engineer, World Models

Join Humanoid, a company on a mission to build the world's most capable and safe humanoid robots. As a Deep Learning Engineer on the World Models team, you will design and train multimodal world models, develop learned-simulator evaluation, generate synthetic rollouts, establish fidelity metrics, build data pipelines, and collaborate with other teams. You should have a track record of training large generative models, deep hands-on experience with modern generative architectures, experience with large-scale distributed training, strong Python + PyTorch/JAX skills, and excitement about grounding generative models in physical reality.

• Design and train multimodal world models, including action-conditioned video prediction and dynamics models.
• Develop learned-simulator evaluation to score candidate policies offline and predict real-world success rates.
• Build data pipelines that turn fleet telemetry, teleoperation logs, and internet-scale video into training corpora for world models.