Research Engineer - 3D World Models

About the role

SpAItial is pioneering the next generation of World Models, pushing the boundaries of generative AI, computer vision, and realistic simulation. The team is moving beyond 2D pixels to build models that natively understand physics and geometry, enabling industries such as robotics, AR/VR, gaming, and cinema to interact with physically‑grounded 3D environments.

Key responsibilities

• Design and develop cutting‑edge generative 3D machine‑learning methods for high‑quality content creation from images, video, and other inputs.
• Build, train, optimize, and evaluate models for 3D reconstruction, novel view synthesis, and world generation.
• Implement and experiment with state‑of‑the‑art 3D representations, including point clouds, meshes, and 3D Gaussian splatting.
• Develop training pipelines and loss functions that improve geometry accuracy, visual fidelity, and spatial consistency.
• Collaborate with researchers to integrate physics‑aware priors and world‑model capabilities into generative systems.
• Analyze model performance, debug failure cases, and iterate rapidly to enhance quality and robustness.

Required profile

• Bachelor's or Master's degree, or equivalent research/project experience, in computer science, machine learning, computer vision, graphics, robotics, or a related field.
• Strong fundamentals in deep learning and generative models, especially diffusion models and transformers.
• Solid understanding of 3D processing concepts such as camera geometry, depth, reconstruction, point clouds, meshes, or Gaussian splats.
• Proficiency in Python and deep‑learning frameworks such as PyTorch, with experience in model training and optimization.
• Ability to implement research papers, run experiments, and iterate quickly on new ideas.
• Strong coding skills and a passion for building reliable, scalable ML systems.

Required skills

• Python
• PyTorch
• Deep learning
• Diffusion models
• Transformers
• Point clouds
• Meshes
• 3D Gaussian splatting
• Camera geometry
• Depth reconstruction