Computer Vision Researcher

As a Computer Vision Researcher at World Labs, you will join a team led by Fei-Fei Li to develop 'spatial intelligence'—AI that understands and interacts with the 3D physical world. You will push the boundaries of large-scale generative models, moving beyond 2D pixels toward holistic 3D scene representation and reasoning. Your work will directly impact how machines perceive depth, geometry, and semantics to create immersive, interactive digital environments.

- Design and implement novel architectures for 3D world modeling and spatial reasoning. - Develop scalable algorithms for 3D reconstruction and generation from sparse 2D image or video data. - Optimize high-performance training pipelines for massive datasets of 3D objects and environments. - Stay at the forefront of AI research, translating academic breakthroughs into production-ready spatial models. - Collaborate with engineers to integrate vision models into interactive, real-time 3D simulation engines. - Author high-quality research papers and represent World Labs at international AI conferences.

- Ph.D. in Computer Science, Robotics, or a related field with a focus on Computer Vision or Deep Learning. - Strong publication record at top-tier venues such as CVPR, ICCV, ECCV, NeurIPS, or SIGGRAPH. - Expertise in 3D vision topics such as Neural Radiance Fields (NeRFs), 3D Gaussian Splatting, or Point Clouds. - Proven experience training large-scale foundation models or generative models (Diffusion, Transformers). - Proficiency in PyTorch and C++, with experience in high-performance GPU kernels (CUDA). - Strong mathematical foundation in projective geometry, linear algebra, and optimization. - Ability to work independently in a fully remote, globally distributed research environment.