Member of Technical Staff, Microsoft Robotics (Spatial AI) | Microsoft Careers

What you'll do

• Design and develop physical world models capturing 3D spatial structure and object relationships.
• Train video prediction models and neural physics simulators to predict future environmental states.
• Implement evaluation frameworks for model accuracy and planning performance benchmarks.
• Build data pipelines for multi-sensor data fusion and dataset curation for diverse environments.
• Write efficient code in Python using PyTorch, JAX, or TensorFlow for large-scale experiments.
• Contribute to the research roadmap for world modeling and spatial AI technologies.

What we're looking for

• Strong background in robotics perception, navigation, and sensor systems integration.
• Experience with world models, video prediction, neural physics simulators, or 3D generative models.
• Expertise in 3D computer vision techniques including depth estimation and point cloud processing.
• Proficiency in PyTorch, JAX, TensorFlow for training large-scale models on GPU clusters.
• Familiarity with model-based reinforcement learning and imagination-based planning approaches.
• Published research or contributions to world models and spatial AI in top-tier venues.
• Experience with Microsoft Azure AI toolset or equivalent cloud ML platforms.